Optica Online Industry Meeting: Photonics Packaging, Assembly and Testing. Recording 29 August 2022.

Full Transcript

https://www.youtube.com/watch?v=zAjkGhkjppA

[00:04] one to the first ever online industry.
[00:07] one to the first ever online industry meeting at optica to the first ever.
[00:11] meeting at optica to the first ever networking event for two hours where.
[00:13] networking event for two hours where you're here to find potential partner.
[00:15] you're here to find potential partner suppliers customers and maybe even.
[00:17] suppliers customers and maybe even investors we are gonna have two hours of.
[00:20] investors we are gonna have two hours of very very strong of very active.
[00:24] very very strong of very active networking.
[00:27] i would like to get everyone active everyone open up your screen.
[00:30] active everyone open up your screen share your screen constantly i want to.
[00:32] share your screen constantly i want to see you switch on your camera it is time.
[00:34] see you switch on your camera it is time to start welcome to the online industry.
[00:37] to start welcome to the online industry meeting in photonics packaging.
[00:40] meeting in photonics packaging assembly.
[00:41] assembly and test.
[00:43] and test welcome back from your holidays it's.
[00:45] welcome back from your holidays it's time to do business again on monday.
[00:48] time to do business again on monday august 29th we are launching the first.
[00:52] august 29th we are launching the first of a new series of five online industry.
[00:55] of a new series of five online industry meetings and we kick off with the.
[00:57] meetings and we kick off with the hottest topic of all photonics packaging.
[01:00] hottest topic of all photonics packaging assembly and testing.
[01:03] assembly and testing these meetings are deep dive interactive.
[01:06] These meetings are deep dive interactive discussions where we focus on end users.
[01:09] Discussions where we focus on end users needs in several important market segments from co-package optics in data centers to fncw lidar in automotive.
[01:11] Needs in several important market segments from co-package optics in data centers to fncw lidar in automotive.
[01:14] Segments from co-package optics in data centers to fncw lidar in automotive for instance to address co-package and plugable optics.
[01:19] Centers to fncw lidar in automotive for instance to address co-package and plugable optics we'll hear from the major market leaders like cisco and h.g genuine both telling us what they need from the photonic industry and how about leiden.
[01:22] For instance to address co-package and plugable optics we'll hear from the major market leaders like cisco and h.g genuine both telling us what they need from the photonic industry and how about leiden.
[01:24] Plugable optics we'll hear from the major market leaders like cisco and h.g genuine both telling us what they need from the photonic industry and how about leiden.
[01:29] Major market leaders like cisco and h.g genuine both telling us what they need from the photonic industry and how about leiden.
[01:31] Genuine both telling us what they need from the photonic industry and how about leiden.
[01:34] From the photonic industry and how about leiden.
[01:35] Leiden you may know that recently luminar acquired freedom photonics what does this mean for the new roadmap and how can the optica industry network help them develop the right package for fncw lidar.
[01:38] You may know that recently luminar acquired freedom photonics what does this mean for the new roadmap and how can the optica industry network help them develop the right package for fncw lidar.
[01:41] Acquired freedom photonics what does this mean for the new roadmap and how can the optica industry network help them develop the right package for fncw lidar.
[01:44] This mean for the new roadmap and how can the optica industry network help them develop the right package for fncw lidar.
[01:47] Can the optica industry network help them develop the right package for fncw lidar.
[01:50] Them develop the right package for fncw lidar.
[01:54] Lidar i am delighted that young vardaman can join our discussion to explain what photonics packaging can learn from semicon.
[01:57] Can join our discussion to explain what photonics packaging can learn from semicon.
[01:59] Photonics packaging can learn from semicon.
[02:03] semicon she brings us a specialist knowledge from the semicond industry that's essential for hybrid solutions.
[02:05] Knowledge from the semicond industry that's essential for hybrid solutions.
[02:08] That's essential for hybrid solutions just around the corner.
[02:10] Just around the corner and last but not least, Gordon Killer.
[02:14] And last but not least, Gordon Killer from DARPA will explain their photonics.
[02:17] From DARPA will explain their photonics in the package for extreme scalability.
[02:19] In the package for extreme scalability program in our discussion related to.
[02:22] Program in our discussion related to opportunities for our members in this.
[02:25] Opportunities for our members in this DARPA initiative, we will discuss the.
[02:27] DARPA initiative, we will discuss the pipes project which has performance.
[02:30] Pipes project which has performance goals that will blow your mind.
[02:33] Goals that will blow your mind. Get ready for a very interactive.
[02:35] Get ready for a very interactive discussion. Let's get started right now.
[02:39] Discussion. Let's get started right now.
[02:41] So ladies and gentlemen, we are live. We.
[02:44] So ladies and gentlemen, we are live. We are live and we are here to find ways to.
[02:46] Are live and we are here to find ways to connect. This meeting combines all the.
[02:49] Connect. This meeting combines all the entire supply chain. We have companies.
[02:50] Entire supply chain. We have companies who provide.
[02:51] Who provide packaging solutions, equipment.
[02:54] Packaging solutions, equipment, faster fiber to chip assembly, silicone.
[02:57] Faster fiber to chip assembly, silicone photonics, indium phosphide, silicon.
[02:59] Photonics, indium phosphide, silicon nitrate. The entire supply chain is here.
[03:02] Nitrate. The entire supply chain is here and we are here, the Optica staff, to make.
[03:04] And we are here, the Optica staff, to make sure that you all connect. I started.
[03:07] Sure that you all connect. I started working for Optica already six months.
[03:09] working for optica already six months ago my job is to make sure that optica becomes an industry network globally and all the companies in your network find the most important membership benefit of all the other members.
[03:23] our goal is to act as a catalyst and connect them all of you.
[03:27] many of you are optical corporate members just to remind you what you get from the optical membership is access to market reports access to events but most important introductions to those companies that you can do business with and these events are all your tools to make those introductions.
[03:48] i speak on behalf of 160 optical staff.
[03:52] in this slide you can see our optical corporate engagement council with ceos of top companies worldwide.
[03:57] these are the people who tell us how to drive this industry network.
[04:00] i would like to also acknowledge your president professor satoshi kavata and the rest of the optica executive staff.
[04:05] together we are here to bring the optica former usa most
[04:12] here to bring the optica former usa most renowned
[04:13] renowned academic platform to be the leading
[04:16] academic platform to be the leading industry network to connect companies
[04:19] industry network to connect companies with potential suppliers partners
[04:21] with potential suppliers partners customers investors and hirable talent
[04:23] customers investors and hirable talent organize about 400 events per year
[04:27] organize about 400 events per year you can find all of them in the website
[04:30] you can find all of them in the website but i want to just highlight that today
[04:32] but i want to just highlight that today we have the first ever the first ever
[04:35] we have the first ever the first ever online industry meeting at optica these
[04:38] online industry meeting at optica these meetings are two hours in which all of
[04:40] meetings are two hours in which all of you are part of the meeting this is not
[04:42] you are part of the meeting this is not a webinar you can hear just to check
[04:44] a webinar you can hear just to check your email while you tell it work no
[04:46] your email while you tell it work no this is not the point of this meeting
[04:47] this is not the point of this meeting the point of this meeting is that all of
[04:49] the point of this meeting is that all of you find ways to help the five end users
[04:52] you find ways to help the five end users we selected for you today we are
[04:54] we selected for you today we are addressing topic of photonics packaging
[04:55] addressing topic of photonics packaging assembly and test next and next last
[04:58] assembly and test next and next last tutorial september it is all about
[05:01] tutorial september it is all about centos second and picosecond laser
[05:03] centos second and picosecond laser applications you also also remember that
[05:06] applications you also also remember that many of you
[05:07] many of you have are applying for providing industry
[05:09] have are applying for providing industry tutorials we're gonna announce many
[05:11] tutorials we're gonna announce many industry tutorials in the coming months
[05:14] industry tutorials in the coming months if as a company you want to
[05:16] if as a company you want to teach how to use your equipment to solve
[05:19] teach how to use your equipment to solve challenges let us know and we'll provide
[05:22] challenges let us know and we'll provide you with one hour online meeting for you
[05:25] you with one hour online meeting for you to tell to teach the rest of the
[05:27] to tell to teach the rest of the industry how your equipment can solve an
[05:30] industry how your equipment can solve an industry challenge and also remember we
[05:32] industry challenge and also remember we have now online in optica industry
[05:36] have now online in optica industry summits face face optical industry
[05:38] summits face face optical industry summits we're gonna have about 10 per
[05:41] summits we're gonna have about 10 per year in the rest months we have one in
[05:43] year in the rest months we have one in paris 2067 october to talk about the
[05:46] paris 2067 october to talk about the quantum industry we're selecting 30
[05:49] quantum industry we're selecting 30 quantum startups companies in the 5-year
[05:52] quantum startups companies in the 5-year business who are gonna tell us how the
[05:54] business who are gonna tell us how the rest of the supply chain can help them
[05:56] rest of the supply chain can help them and on the 8th and 9th of november in
[05:58] and on the 8th and 9th of november in silicon valley many of you are going to
[05:59] silicon valley many of you are going to be there co-package and plugable optics
[06:02] be there co-package and plugable optics we want to find the entire supply chain
[06:05] we want to find the entire supply chain package and practicable optic spinning
[06:06] package and practicable optic spinning gold together to silicon valley hosted
[06:08] gold together to silicon valley hosted by dupont technology center in san jose
[06:10] by dupont technology center in san jose to find ways to connect but today today
[06:13] to find ways to connect but today today it is all about the photonics packaging
[06:15] it is all about the photonics packaging assembly and test i would like to
[06:18] assembly and test i would like to thank
[06:19] thank all of you
[06:20] all of you thank the amazing optica team that put
[06:22] thank the amazing optica team that put this together i really had a big dream
[06:25] this together i really had a big dream for this first meeting and my
[06:26] for this first meeting and my expectations were
[06:28] expectations were much past we have
[06:30] much past we have 490 registrations and we have selected
[06:33] 490 registrations and we have selected five companies who are driving the
[06:36] five companies who are driving the industry and who are here looking for
[06:38] industry and who are here looking for partners looking for people to help them
[06:40] partners looking for people to help them we have peter developer francisco gordon
[06:43] we have peter developer francisco gordon killer from darpa gordon morrison from
[06:45] killer from darpa gordon morrison from freedom photonics now aluminar company
[06:48] freedom photonics now aluminar company ginseo there is my friend jim from
[06:50] ginseo there is my friend jim from hgtech and last but definitely the very
[06:53] hgtech and last but definitely the very well known young vardaman thank you very
[06:56] much all of you but these are our
[06:58] featuring speakers but all of you all of
[07:01] you today in the room are past of this meeting also would
[07:03] room are past of this meeting also would like to remind all of you that this
[07:05] like to remind all of you that this meeting is also live stream in youtube
[07:06] meeting is also live stream in youtube so let me say hello to the youtubers
[07:09] so let me say hello to the youtubers hello youtubers of the world if you're
[07:11] hello youtubers of the world if you're following this video for youtube just
[07:12] following this video for youtube just make sure that you write all your
[07:13] make sure that you write all your comments and questions in the chat and
[07:16] comments and questions in the chat and we'll read them during the meeting.
[07:18] we'll read them during the meeting throw you in the zoom room remember that.
[07:19] you in the zoom room remember that you're here to connect so make sure that.
[07:22] you're here to connect so make sure that you find who else is in the room as well.
[07:25] you find who else is in the room as well everyone in the room the 125 people.
[07:27] everyone in the room the 125 people already is gonna grow to 400 all of them.
[07:30] already is gonna grow to 400 all of them have shared the company name just you.
[07:32] have shared the company name just you know what kind of companies are in the.
[07:34] know what kind of companies are in the room in the next meetings i'm going to.
[07:36] room in the next meetings i'm going to publish also a supply chain but for this.
[07:38] publish also a supply chain but for this first ever meeting there were many.
[07:41] first ever meeting there were many companies that i wasn't sure where to.
[07:42] companies that i wasn't sure where to position them i want to talk to them.
[07:44] position them i want to talk to them first that's why this is the first ever.
[07:46] first that's why this is the first ever meeting question you're gonna have the.
[07:47] meeting question you're gonna have the supply chain with all the companies.
[07:48] supply chain with all the companies registered and that's all i had to say i.
[07:51] registered and that's all i had to say i want to give the floor to our first.
[07:54] want to give the floor to our first feature speaker a personal friend of.
[07:56] feature speaker a personal friend of mine peter doubler spitted the double we.
[07:59] mine peter doubler spitted the double we know each other for a long time.
[08:01] know each other for a long time from europe thank you very much for.
[08:03] from europe thank you very much for being here from luxtera now a cisco.
[08:06] being here from luxtera now a cisco company telling us how the rest of the.
[08:08] company telling us how the rest of the optical supply chain can help you the.
[08:10] optical supply chain can help you the floor and the attention of everyone goes.
[08:13] floor and the attention of everyone goes to cisco.
[08:15] to cisco okay thanks so much for the introduction.
[08:18] okay thanks so much for the introduction uh
[08:20] uh yeah
[08:25] uh can you see my presentation correctly
[08:27] uh can you see my presentation correctly it is that clear
[08:29] it is that clear okay
[08:30] okay so let me start here since we have a
[08:32] so let me start here since we have a little time so i'll give a little
[08:33] little time so i'll give a little overview of uh what's our approaches for
[08:37] overview of uh what's our approaches for packaging assembly and tests of a
[08:39] packaging assembly and tests of a silicon photonics optical transceivers
[08:43] silicon photonics optical transceivers so it's narrowing down the applications
[08:45] so it's narrowing down the applications a little bit it's for the use of silicon
[08:47] a little bit it's for the use of silicon photonics for optical transceivers
[08:49] photonics for optical transceivers so if you look here
[08:51] so if you look here at the the trends that we see in in
[08:54] at the the trends that we see in in technology requirements and product
[08:55] technology requirements and product requirements is that
[08:57] requirements is that bit rates are steadily increasing okay
[09:00] bit rates are steadily increasing okay so now we are at 56
[09:02] so now we are at 56 gigabyte per seconds we are going 212
[09:05] gigabyte per seconds we are going 212 and in the future potentially we go even
[09:07] and in the future potentially we go even to the double of that
[09:09] to the double of that for both the electrical and the optical
[09:12] for both the electrical and the optical uh interfaces and that means that not
[09:15] uh interfaces and that means that not only the optics have to go faster you
[09:16] only the optics have to go faster you need fast modulators fast photo
[09:18] need fast modulators fast photo detectors and so forth you also need the
[09:20] detectors and so forth you also need the electronics going with that and that.
[09:22] electronics going with that and that means that we will need advanced.
[09:24] means that we will need advanced supporting.
[09:25] supporting um advanced cmos nodes so currently uh.
[09:28] um advanced cmos nodes so currently uh designs are.
[09:30] designs are done at seven nanometers and.
[09:32] done at seven nanometers and in a couple of years or actually maybe.
[09:34] in a couple of years or actually maybe now already we are already working at at.
[09:37] now already we are already working at at five nanometers in the future three.
[09:39] five nanometers in the future three nanometer uh cmos.
[09:41] nanometer uh cmos another thing that is very important and.
[09:43] another thing that is very important and that ties into the the packaging is that.
[09:45] that ties into the the packaging is that the parasitics the electrical parasitics.
[09:48] the parasitics the electrical parasitics of a resistance inductance and.
[09:50] of a resistance inductance and capacitance.
[09:51] capacitance to a large extent affect the performance.
[09:54] to a large extent affect the performance of optical transceivers so you need very.
[09:56] of optical transceivers so you need very advanced low parasitic electrical.
[09:59] advanced low parasitic electrical interconnect and that dies directly into.
[10:01] interconnect and that dies directly into the packaging.
[10:03] the packaging another thing that we see is that and.
[10:05] another thing that we see is that and that we we just heard that there's also.
[10:07] that we we just heard that there's also a lot to talk about.
[10:09] a lot to talk about co-packaged optics.
[10:10] co-packaged optics we need higher levels of integration so.
[10:13] we need higher levels of integration so we need more channels and more optical.
[10:15] we need more channels and more optical in.
[10:16] in ios per square.
[10:18] ios per square millimeter.
[10:20] millimeter that drives that everything has to be.
[10:21] that drives that everything has to be more compact not only the optics but
[10:23] more compact not only the optics but also the electronics you have to come
[10:25] also the electronics you have to come connect everything very closely together
[10:28] connect everything very closely together what this has as a side effect is that
[10:30] what this has as a side effect is that also the power dissipation will get much
[10:33] also the power dissipation will get much more concentrated and that makes that
[10:35] more concentrated and that makes that heat sinking is also a challenge so
[10:37] heat sinking is also a challenge so that's also something that is adding to
[10:39] that's also something that is adding to the packaging
[10:40] the packaging and there again how these things are
[10:42] and there again how these things are solved are um
[10:44] solved are um by using a dense electrical interconnect
[10:46] by using a dense electrical interconnect so copper pillars interposers through
[10:48] so copper pillars interposers through substrate vias and so forth
[10:51] substrate vias and so forth and also it also means that the photonic
[10:53] and also it also means that the photonic die the picks are getting larger than
[10:56] die the picks are getting larger than they are currently that's also uh poses
[10:59] they are currently that's also uh poses some some packaging
[11:00] some some packaging challenges and finally of course there's
[11:03] challenges and finally of course there's the manufacturing we need high volume
[11:05] the manufacturing we need high volume low cost manufacturing i know that in an
[11:08] low cost manufacturing i know that in an optics there are many applications where
[11:10] optics there are many applications where where with different kinds of uh volumes
[11:12] where with different kinds of uh volumes but for optical transceivers you have to
[11:14] but for optical transceivers you have to think about
[11:16] think about a couple of millions per year in
[11:19] a couple of millions per year in production so all your your
[11:21] production so all your your your design must be for
[11:23] your design must be for manufacturing and low cost.
[11:27] manufacturing and low cost and you have to be able to ramp to high.
[11:29] and you have to be able to ramp to high volume in a pretty short time.
[11:31] volume in a pretty short time so time to market is important so.
[11:34] so time to market is important so our approach there is that we want to.
[11:35] our approach there is that we want to leverage as much as possible mature.
[11:37] leverage as much as possible mature technologies automation and try to make.
[11:40] technologies automation and try to make things as simple as possible to to the.
[11:43] things as simple as possible to to the right of the slide you see there one of.
[11:44] right of the slide you see there one of the.
[11:45] the things evolutions that we went through.
[11:47] things evolutions that we went through is that we went from wire bonding the.
[11:49] is that we went from wire bonding the photonic chips towards through substrate.
[11:51] photonic chips towards through substrate via or 3d packaging for the silicon.
[11:54] via or 3d packaging for the silicon photonics.
[11:55] photonics i will show some example in the next.
[11:57] i will show some example in the next slide.
[11:58] slide so.
[11:59] so this slide is showing some examples of.
[12:01] this slide is showing some examples of things that we have done over the years.
[12:03] things that we have done over the years the first one is that we want to.
[12:06] the first one is that we want to use that.
[12:07] use that packaging technologies from the.
[12:08] packaging technologies from the electronics so we started with using.
[12:11] electronics so we started with using micro bumps to attach an electronic ic.
[12:13] micro bumps to attach an electronic ic to a photonic ic and currently.
[12:16] to a photonic ic and currently the current products they are using.
[12:18] the current products they are using through substrate vias.
[12:20] through substrate vias so 3d packaging that is basically where.
[12:23] so 3d packaging that is basically where we have in the photonic ic in the pic we.
[12:25] we have in the photonic ic in the pic we have uh drilled holes and fill them up with copper so that we connect the top side and the bottom side that solves a number of packaging issues electrical interconnect issues.
[12:37] um you see that the cross-section of the a product that we have that's a 400 gigabit uh dr4 or fr4 optical module and that is using that 3d packaging.
[12:53] another element in of optical assembly is wafer level uh packaging and assembly so you see there are a number of examples the first one is laser diode packaging on a wafer level um with with um automated pick and place tools uh we make millions like that that a year and the other one is showing how we then put the light source and electronic die also on a way for scale uh on a silicon photonics wafer.
[13:24] another uh aspect there is that in in
[13:27] another uh aspect there is that in in automation you want to do things on automation.
[13:28] you want to do things on wafers or on panels and there on the bottom you see how we do panel level attach of the fiber interface so that is attach of the fiber interface.
[13:32] so that is where we attach a fiber array with an mt connector uh on on the silicon photonics chip.
[13:41] and then the last one is you see there how when everything is put together what you see there is a 400g dr4 optical module.
[13:51] module with the photonic dye the electronic dye and the light source and the fiber interconnect you see that the assembly is very simple because of the high level of integration.
[14:01] and that is the the product with the tsv's and and the and the copper pillar bump interconnect.
[14:09] that brings me then to my last slide so that this what is our test approach um also here the same as for the assembly.
[14:16] we try to leverage commercial equipment for uh electrical and optical wafer acceptance test at the fab and also electrical and optical sort at us at the at the osat.
[14:28] at the osat so this is a machine that we developed.
[14:30] so this is a machine that we developed with the tokyo electron.
[14:32] with the tokyo electron and it does optical sort it measures.
[14:34] and it does optical sort it measures automatically optical performance of.
[14:36] automatically optical performance of devices on a wafer.
[14:39] devices on a wafer another thing that uh that's related.
[14:41] another thing that uh that's related with the testing is that we have a.
[14:43] with the testing is that we have a built-in self-test so.
[14:46] built-in self-test so in our photonic chip.
[14:48] in our photonic chip we integrate both the transmitter and.
[14:50] we integrate both the transmitter and the receiver and in the electronic die.
[14:52] the receiver and in the electronic die we also have a bit error rate tester.
[14:54] we also have a bit error rate tester actually so we have a prbs.
[14:57] actually so we have a prbs generator and a prbs checker.
[14:59] generator and a prbs checker and.
[15:00] and my use of of an optical switch which we.
[15:02] my use of of an optical switch which we have on our photonic chip we can switch.
[15:04] have on our photonic chip we can switch the output of the transmitter to the.
[15:07] the output of the transmitter to the input of the receiver when we go into.
[15:09] input of the receiver when we go into the the test mode and we can test the.
[15:11] the the test mode and we can test the bit error rate and that allows us to to.
[15:14] bit error rate and that allows us to to test optically at high speed without the.
[15:18] test optically at high speed without the need for high speed.
[15:20] need for high speed probing so that that's a.
[15:22] probing so that that's a a big advantage and a big reduction in.
[15:24] a big advantage and a big reduction in cost when you want to want to build such.
[15:26] cost when you want to want to build such devices.
[15:27] devices and then last but not least i mean it's
[15:30] and then last but not least i mean it's use of the use of the the common knowledge in in manufacturing.
[15:33] the common knowledge in in manufacturing it's a non-good die approach so when you have a flow of manufacturing optical transceivers all the elements you bring in you know that they are good and then you assemble them with a good process flow with process monitoring and so forth and that allows you then to optimize the the the yields and also optimize the the the costs accordingly.
[15:56] and that ends my presentation i'm open for questions if there are any later on.
[16:01] thank you very much peter for a great presentation it is so great to have you here.
[16:07] we have many questions for you because we have the entire the entire supply chain in front of us but the first thing i would like to tell you is that since i joined optic i haven't been stopped traveling and i went to i went to austria to visit a company called bessie.
[16:23] do we have anybody from bessie in the room is is peter with us.
[16:28] i'm here.
[16:30] I'm here. Hey Peter, it's good to see you.
[16:32] Hey Peter, it's good to see you as you have seen in the previous slide.
[16:34] We saw from BC.
[16:37] From BC, sorry, we saw from...
[16:42] We saw from Cisco the need for wafer and panel level assembly leveraging passive and active alignment.
[16:46] We saw from Cisco the need for wafer and panel level assembly leveraging passive and active alignment.
[16:49] Panel level assembly leveraging passive and active alignment.
[16:52] And active alignment when it comes to the passive and active optical alignment.
[16:54] When it comes to the passive and active optical alignment, that's something that I, I, we're going to discuss later.
[16:55] Optical alignment, that's something that I, I, we're going to discuss later.
[16:57] We have companies in the room like Ficontec, for example.
[16:59] For example, but I would like to understand a what does, beside from the equipment perspective and how you could support.
[17:04] What does, beside from the equipment perspective and how you could support?
[17:06] Perspective and how you could support, how you could contribute to this a wafer level assembly that is needed for the next generation optical assembly?
[17:08] You could support, how you could contribute to this a wafer level assembly that is needed for the next generation optical assembly.
[17:10] Contribute to this a wafer level assembly that is needed for the next generation optical assembly.
[17:13] Assembly that is needed for the next generation optical assembly.
[17:16] Generation optical assembly. Yeah, thanks, uh, for asking Jose.
[17:19] Yeah, thanks, uh, for asking Jose. Actually, we're, we're in the passive alignment uh market.
[17:22] Actually, we're, we're in the passive alignment uh market, so we do not do active alignment.
[17:24] So we do not do active alignment.
[17:26] Active alignment, but uh, of course, without.
[17:28] Active alignment, but uh, of course, without.
[17:30] without very flexible platforms at bessie we do very flexible platforms at bessie we do passive alignment and uh mainly we we are delivering to direct attach uh diet attach equipment or also flip chip equipment and if i look at those uh processes we're looking at there's also one very new and interesting platform that we have been working on in the past three years where we do a so-called direct bonding process so it's a real covalent bonding process where it's there is no flux there is no epoxy or neither a bonding agent involved it's basically bringing a quite high quality let's say surfaces together and initiate bonding by uh via atomic forces so that could be one very interesting perspective for the maybe not the so near future but further future and we have a platform that's actually really doing this already now and doing this already in serial.
[18:31] and doing this already in serial production for one very specific product.
[18:33] production for one very specific product yeah.
[18:35] yeah all the meetings that we have so far we.
[18:37] all the meetings that we have so far we always discuss how annoying is epoxy.
[18:40] always discuss how annoying is epoxy could you let us know i'm sorry that i.
[18:42] could you let us know i'm sorry that i interrupted you peter uh how this.
[18:46] interrupted you peter uh how this glueless epoxy-less bonding could be a.
[18:49] glueless epoxy-less bonding could be a solution.
[18:51] solution i'm.
[18:53] i'm peter peter.
[18:55] peter peter sorry.
[18:56] sorry oh okay.
[18:57] oh okay i think that the the technology that was.
[19:00] i think that the the technology that was just mentioned uh is indeed very.
[19:02] just mentioned uh is indeed very interesting not so much because it does.
[19:04] interesting not so much because it does not use epoxy but just because it allows.
[19:07] not use epoxy but just because it allows direct bonding of an electronic diet to.
[19:09] direct bonding of an electronic diet to a photonic dye.
[19:10] a photonic dye and what's important there what i.
[19:12] and what's important there what i mentioned in my first slide is that can.
[19:13] mentioned in my first slide is that can be done with minimum.
[19:15] be done with minimum parasitic capacitance and that is very.
[19:18] parasitic capacitance and that is very important for the high speed.
[19:20] important for the high speed applications when we go to beyond.
[19:23] applications when we go to beyond gigabit per second.
[19:25] gigabit per second having these direct bonds of two dice to.
[19:27] having these direct bonds of two dice to each other without uh.
[19:30] each other without uh a solder connection.
[19:32] A solder connection is very important for reducing the parasitics.
[19:36] And that's indeed a very interesting field to look into.
[19:40] I, I don't know Peter, you have with us one slide that you could show us about the direct bonding dongle Bessie because I really Peter, I have a problem when I have two Peters, Peter and their buddy.
[19:51] Yes.
[19:52] Uh, if you allow me to share my screen.
[19:56] Always stay with you as an Optica member, you can always share Guangxles at all or online industry meetings.
[20:01] Okay, basically, basically I prepared uh for this meeting uh four slides that I would like to to present to you.
[20:09] It's it's really very basic and nothing, no technical slides and this just to give you an idea what we are doing with this platform, that's okay with you?
[20:18] Sorry, presentation mode.
[20:21] Can you all see my screen clear?
[20:24] So direct spawning, I do not know whether you guys are all aware of that, but in principle, is this something if if the components and their surfaces are
[20:32] components and their surfaces are sufficiently smooth and flat
[20:35] sufficiently smooth and flat a bond front will propagate out by it
[20:38] a bond front will propagate out by it outward from the first point of initial
[20:40] outward from the first point of initial contact that's a process well known
[20:43] contact that's a process well known like evt has done that on wafer to wafer
[20:46] like evt has done that on wafer to wafer level and what we are doing is we're
[20:48] level and what we are doing is we're doing this on a diet to wafer level
[20:52] doing this on a diet to wafer level so meaning bringing those two parts
[20:54] so meaning bringing those two parts together and by simply at room
[20:56] together and by simply at room temperature up applying a very low force
[20:59] temperature up applying a very low force bonding is initiated it needs an
[21:01] bonding is initiated it needs an annealing step afterwards to
[21:04] annealing step afterwards to really finalize the bonding but what you
[21:06] really finalize the bonding but what you can see in this video is basically just
[21:08] can see in this video is basically just uh
[21:09] uh this done by two transparent surfaces
[21:13] this done by two transparent surfaces by initiating a little bit
[21:14] by initiating a little bit force in the middle it actually
[21:16] force in the middle it actually propagates the whole bonding way
[21:18] propagates the whole bonding way throughout the whole wafer and what
[21:19] throughout the whole wafer and what bezier does in this case is doing this
[21:22] bezier does in this case is doing this on a die to wafer level meaning in our
[21:26] on a die to wafer level meaning in our machine we do have
[21:28] machine we do have a source wafer or a source uh carrier of
[21:32] a source wafer or a source uh carrier of dice for example can be a framed wafer
[21:35] dice for example can be a framed wafer uh with our already diced eyes and then
[21:37] uh with our already diced eyes and then we flip chip on this onto a target wafer
[21:40] we flip chip on this onto a target wafer of course there is a lot more involved
[21:43] of course there is a lot more involved those wafers and all parts involved need
[21:45] those wafers and all parts involved need prior treatments activation cleaning and
[21:48] prior treatments activation cleaning and activation plasma treatment
[21:50] activation plasma treatment and once we have done this there is
[21:53] and once we have done this there is another step an annealing step where
[21:56] another step an annealing step where actually
[21:57] actually the difference in thermal expansion
[21:59] the difference in thermal expansion between the silicon
[22:01] between the silicon and the metal let's say the copper
[22:05] and the metal let's say the copper will lead to those two
[22:07] will lead to those two copper parts really making contact and
[22:10] copper parts really making contact and making a bond itself
[22:13] making a bond itself so there is no agent or no epoxy or
[22:15] so there is no agent or no epoxy or nothing else involved we have a very
[22:17] nothing else involved we have a very high quality interconnect uh silicon
[22:21] high quality interconnect uh silicon oxide to silicon oxide as well as copper
[22:23] oxide to silicon oxide as well as copper to copper
[22:24] to copper and i think is quiet both is quite
[22:26] and i think is quiet both is quite important
[22:27] important on the silicon photonics world because
[22:30] on the silicon photonics world because uh on on the electronics you want you
[22:32] uh on on the electronics you want you want uh best possible coupling but also
[22:35] want uh best possible coupling but also on the optical side you do want to have
[22:38] on the optical side you do want to have um minimized losses when you're trying
[22:41] um minimized losses when you're trying to couple in light from
[22:43] to couple in light from from actives or a laser source
[22:47] and yeah that's that's just the
[22:49] and yeah that's that's just the advantages i do see uh with with this
[22:53] advantages i do see uh with with this technology involved there is no flux uh
[22:55] technology involved there is no flux uh also quite important uh in our case is
[22:58] also quite important uh in our case is high accuracy we can go down to 200
[23:00] high accuracy we can go down to 200 nanometers of
[23:03] nanometers of alignment accuracy three sigma it's a
[23:06] alignment accuracy three sigma it's a very clean uh platform
[23:08] very clean uh platform down to iso 3 even in the bonding area
[23:11] down to iso 3 even in the bonding area we can do room temperature bonding
[23:14] we can do room temperature bonding uh also annealing is at moderate
[23:16] uh also annealing is at moderate temperatures around 350 to 400 degrees
[23:19] temperatures around 350 to 400 degrees celsius in the end
[23:21] celsius in the end and uh i do assume that with this
[23:24] and uh i do assume that with this technology uh also in the silicon
[23:26] technology uh also in the silicon photonics world or in the peak world you
[23:28] photonics world or in the peak world you will get high optical and
[23:30] will get high optical and electrical connection quality
[23:33] electrical connection quality and of course what's very important uh
[23:35] and of course what's very important uh it's not only a technique not only from
[23:38] it's not only a technique not only from a technical perspective but also with
[23:40] a technical perspective but also with our uh bonding equipment which is a dual
[23:43] our uh bonding equipment which is a dual gantry flip chip equipment you can get
[23:46] gantry flip chip equipment you can get to uph of uh up to 000 for for this for
[23:50] to uph of uh up to 000 for for this for these accuracies that's where what we're
[23:52] these accuracies that's where what we're working on right now
[23:55] working on right now thank you very much peter you have been
[23:57] thank you very much peter you have been the first one slider for the online
[23:59] the first one slider for the online industry meetings remember when i say
[24:01] industry meetings remember when i say one slide sorry please please please but
[24:04] one slide sorry please please please but we have questions for you the first one
[24:05] we have questions for you the first one is coming from size in jenna ulrike
[24:09] is coming from size in jenna ulrike boeing is with us from science the
[24:11] boeing is with us from science the leading manufacturer of optics and
[24:12] leading manufacturer of optics and optical instrumentation
[24:15] optical instrumentation good
[24:16] good afternoon thanks for uh organizing this
[24:20] afternoon thanks for uh organizing this is actually quite uh quite fantastic so
[24:22] is actually quite uh quite fantastic so now my my only question is actually
[24:24] now my my only question is actually about the um
[24:28] if you could maybe specify a bit more um
[24:31] if you could maybe specify a bit more um what the losses are approximately uh for
[24:33] what the losses are approximately uh for these um bonding steps because right now
[24:36] these um bonding steps because right now he said that it's quite supreme it's
[24:38] he said that it's quite supreme it's quite nice but uh could you just uh
[24:41] quite nice but uh could you just uh provide us with some numbers uh possible
[24:43] provide us with some numbers uh possible as well that would be great and i'm a
[24:45] as well that would be great and i'm a lot over over cochrane by the way not
[24:48] lot over over cochrane by the way not enough sorry okay
[24:50] enough sorry okay anyway
[24:51] anyway but losses you mean in terms of yield
[24:54] but losses you mean in terms of yield right yeah yeah yeah
[24:56] right yeah yeah yeah okay i um um
[24:59] okay i um um i'm not sure i'm i'm not the project
[25:01] i'm not sure i'm i'm not the project manager manager uh for these for this
[25:04] manager manager uh for these for this platform and i'm
[25:06] platform and i'm i'm not the process expert uh
[25:09] i'm not the process expert uh i do assume we can have a yield up to
[25:12] i do assume we can have a yield up to maybe 80 right now it's a it's a quite
[25:15] maybe 80 right now it's a it's a quite tricky process uh of course we are
[25:18] tricky process uh of course we are targeting a lot higher yields but that's
[25:20] targeting a lot higher yields but that's not only up to us in in this case
[25:22] not only up to us in in this case because the
[25:23] because the for example the source wafer it's a
[25:26] for example the source wafer it's a diced wafer and it's a wafer that's
[25:29] diced wafer and it's a wafer that's that's framed so you have a tape and
[25:32] that's framed so you have a tape and then you have a standard ejection
[25:33] then you have a standard ejection process so this uh usually introduces
[25:36] process so this uh usually introduces particles when you check this wafer you
[25:38] particles when you check this wafer you then pick it up flip it and flip to bond
[25:41] then pick it up flip it and flip to bond it on onto the target wafer so there is
[25:43] it on onto the target wafer so there is a bit of
[25:44] a bit of yield loss involved since you will
[25:46] yield loss involved since you will generate some particles and as you might
[25:49] generate some particles and as you might know even a
[25:50] know even a quite small particle between those
[25:52] quite small particle between those surfaces will corrupt the
[25:54] surfaces will corrupt the interconnection at some point
[25:57] interconnection at some point we have a working process
[25:58] we have a working process um
[26:00] um that that is currently running at a
[26:02] that that is currently running at a quite big customer uh from from our side
[26:06] quite big customer uh from from our side and frankly speaking i don't know how
[26:09] and frankly speaking i don't know how good the yield is
[26:10] good the yield is at this stage there is
[26:12] at this stage there is an official product available already
[26:14] an official product available already now that's produced with this machine
[26:16] now that's produced with this machine but i don't i'm not even sure if we even
[26:19] but i don't i'm not even sure if we even have those yield numbers
[26:20] have those yield numbers thank you peter and thank you enrique i
[26:23] thank you peter and thank you enrique i go from peter utter ballister back to
[26:25] go from peter utter ballister back to peter the dollar peter the dollar
[26:27] peter the dollar peter the dollar there's a question from you coming from
[26:30] there's a question from you coming from photonect ix and i'm going to try
[26:34] photonect ix and i'm going to try is sushant kumar with us today oh yep
[26:37] is sushant kumar with us today oh yep i'm here please tell us what's on your
[26:39] i'm here please tell us what's on your mind oh uh yeah so uh thank you very
[26:42] mind oh uh yeah so uh thank you very much for organizing this it's really
[26:44] much for organizing this it's really great uh so i just wanted to ask when
[26:45] great uh so i just wanted to ask when you're connecting fiber arrays to chip
[26:48] you're connecting fiber arrays to chip um
[26:49] um does that cause any issues in terms of
[26:52] does that cause any issues in terms of the uniformity of the fiber array and
[26:54] the uniformity of the fiber array and the presence of or maybe the need for
[26:57] the presence of or maybe the need for the
[26:58] the groove block on the fiber rsi itself in
[27:01] groove block on the fiber rsi itself in addition to whatever passive alignment
[27:03] addition to whatever passive alignment you need to do on the chip site and if
[27:06] you need to do on the chip site and if that has any issue um
[27:08] that has any issue um any implications with regards to
[27:10] any implications with regards to performance or cost or something like
[27:12] performance or cost or something like that
[27:13] that yeah sure uh we use a
[27:16] yeah sure uh we use a glass v-groove arrays with fibers in
[27:19] glass v-groove arrays with fibers in them to attach to the to the photonic
[27:21] them to attach to the to the photonic chip
[27:22] chip um
[27:24] um usually the the manufacturers of those
[27:26] usually the the manufacturers of those fiber arrays can make them with a with
[27:28] fiber arrays can make them with a with where the core
[27:30] where the core tolerance is less than one micron and
[27:32] tolerance is less than one micron and that is sufficient for us
[27:34] that is sufficient for us um because we have we use a grading
[27:37] um because we have we use a grading couplers to couple light out of our
[27:38] couplers to couple light out of our silicon photonics
[27:40] silicon photonics so that is a mode size which is
[27:42] so that is a mode size which is very similar to that of a standard
[27:44] very similar to that of a standard single mode fiber so there is some we
[27:46] single mode fiber so there is some we can allow some uh
[27:48] can allow some uh misalignment a micro misalignment is not
[27:51] misalignment a micro misalignment is not not not
[27:53] not not really affecting insertion osteologic
[27:54] really affecting insertion osteologic stem
[27:56] stem well the other and what is important is
[27:59] well the other and what is important is um uh
[28:00] um uh if you attach fibers to a
[28:04] if you attach fibers to a tsv chip
[28:06] tsv chip the tsv chip tsv chips are usually
[28:09] the tsv chip tsv chips are usually thinned down to about 100 microns
[28:11] thinned down to about 100 microns usually they are not completely flat
[28:14] usually they are not completely flat so if you have edge coupling
[28:16] so if you have edge coupling uh the the warpage of that of that pick
[28:19] uh the the warpage of that of that pick may misalign with respect to the
[28:23] may misalign with respect to the to the pretty uniform um alignment of
[28:25] to the pretty uniform um alignment of the fibers in the v-groove so um warpage
[28:29] the fibers in the v-groove so um warpage of the silicon photonics
[28:32] of the silicon photonics tint
[28:34] tint pick
[28:35] pick may be a challenge
[28:36] may be a challenge when you do edge coupling
[28:39] when you do edge coupling peter the doubler i would like to
[28:41] peter the doubler i would like to address quickly with you the topic of
[28:42] address quickly with you the topic of testing we have here you have told us a
[28:45] testing we have here you have told us a bit about the testing approach for
[28:46] bit about the testing approach for silicon photonics is there already a
[28:49] silicon photonics is there already a standard probe to do the wafer level
[28:53] standard probe to do the wafer level testing of silicon photonics
[28:55] testing of silicon photonics um well we have uh
[28:57] um well we have uh as i mentioned in my presentation tokyo
[28:59] as i mentioned in my presentation tokyo electron has a we have collaborated with
[29:02] electron has a we have collaborated with tokyo electron in japan to to build an
[29:04] tokyo electron in japan to to build an optical
[29:05] optical prober based on their standard probing
[29:07] prober based on their standard probing systems
[29:08] systems the difference is that the the
[29:09] the difference is that the the electrical probe card is replaced with a
[29:12] electrical probe card is replaced with a fiber array
[29:13] fiber array and there is software uh running on the
[29:16] and there is software uh running on the machine that is doing the active
[29:18] machine that is doing the active alignment
[29:19] alignment of the the fibers doing the the test so
[29:22] of the the fibers doing the the test so that is one platform and i believe there
[29:24] that is one platform and i believe there are some other ones in the market as
[29:25] are some other ones in the market as well
[29:26] well that have that offer commercial
[29:29] that have that offer commercial optical
[29:30] optical sword capability yes there are actually
[29:33] sword capability yes there are actually tobias nause from jen optic good
[29:36] tobias nause from jen optic good afternoon tavias good afternoon hello
[29:38] afternoon tavias good afternoon hello jose thank you for having me here
[29:41] jose thank you for having me here um
[29:41] um yeah i had actually um one comment on on
[29:46] yeah i had actually um one comment on on jose's
[29:48] jose's announcement and yeah
[29:49] announcement and yeah so we uh for me and optic we we do have
[29:52] so we uh for me and optic we we do have an a kind of standard probe for
[29:57] measuring
[29:58] measuring optical and electrical
[30:01] optical and electrical picks in parallel and with standard
[30:04] picks in parallel and with standard equipment so that's called our ufo probe
[30:07] equipment so that's called our ufo probe and on the contrast to the active
[30:09] and on the contrast to the active alignment we
[30:12] go the path for passive alignment so
[30:15] go the path for passive alignment so that it runs really on uh standard
[30:18] that it runs really on uh standard equipment so
[30:19] equipment so tokyo electron probe architect probe
[30:22] tokyo electron probe architect probe and
[30:23] and we are also going to
[30:25] we are also going to make it compatible to
[30:27] make it compatible to the tester approach that you can use it
[30:29] the tester approach that you can use it with
[30:30] with high throughput testers
[30:33] high throughput testers and
[30:34] and my second second thing was uh also
[30:37] my second second thing was uh also related to that the question to to the
[30:39] related to that the question to to the your um you showed that there um
[30:43] your um you showed that there um that there is an uh best
[30:46] that there is an uh best built-in safe test uh in in your devices
[30:50] built-in safe test uh in in your devices and i i was wondering if there's really
[30:53] and i i was wondering if there's really a built-in
[30:54] a built-in self-test how much uh how big is the
[30:56] self-test how much uh how big is the need for
[30:57] need for actual wafer level test optical and
[31:00] actual wafer level test optical and electrical
[31:02] um well the
[31:04] um well the the the test strategy
[31:07] the the test strategy has many steps so you so there's a when
[31:10] has many steps so you so there's a when you have the photonic waveforms
[31:11] you have the photonic waveforms manufacturer there's a way for
[31:12] manufacturer there's a way for acceptance test at the foundry so there
[31:16] acceptance test at the foundry so there they test test structures in the scribe
[31:18] they test test structures in the scribe grid and they are both electrical and
[31:20] grid and they are both electrical and optical so that is done with the prober
[31:22] optical so that is done with the prober that i just showed a picture of
[31:25] that i just showed a picture of consigned to our
[31:27] consigned to our foundry
[31:28] foundry um
[31:29] um and then when when that is done you do
[31:31] and then when when that is done you do optical sort and during optical sort you
[31:33] optical sort and during optical sort you test every chip that is on the wafer
[31:36] test every chip that is on the wafer optically
[31:37] optically um and then you have uh
[31:40] um and then you have uh depending on which flow you have with
[31:42] depending on which flow you have with tsv's or without dsvs
[31:45] tsv's or without dsvs you have then also a test
[31:47] you have then also a test uh when the electronic chip is attached
[31:50] uh when the electronic chip is attached to the
[31:51] to the due to the photonic dye or and when the
[31:54] due to the photonic dye or and when the the light source is attached to the
[31:55] the light source is attached to the photonic dye when you have all three of
[31:57] photonic dye when you have all three of them combined you can
[32:01] them combined you can contact the whole chipset assembly and
[32:04] contact the whole chipset assembly and have the whole thing operate you can
[32:05] have the whole thing operate you can switch the light source on you can
[32:07] switch the light source on you can switch the
[32:08] switch the electronic ic on the electronic ic can
[32:11] electronic ic on the electronic ic can drive the modulators receive for the
[32:14] drive the modulators receive for the current from from the the photodiodes
[32:16] current from from the the photodiodes and so you can make a whole loop
[32:18] and so you can make a whole loop uh having the light source all modulate
[32:20] uh having the light source all modulate the light
[32:22] the light with a bit pattern then you can receive
[32:24] with a bit pattern then you can receive the bits of optical bit pattern and then
[32:28] the bits of optical bit pattern and then we have in our electronic die
[32:30] we have in our electronic die like the bit error rate tester with the
[32:32] like the bit error rate tester with the error checker so you
[32:34] error checker so you compare with what you send out with what
[32:36] compare with what you send out with what you receive and then from that you can
[32:37] you receive and then from that you can determine the bit error rate and at the
[32:39] determine the bit error rate and at the same time you can turn the the light
[32:41] same time you can turn the the light source down
[32:42] source down as you do that and you see then how
[32:44] as you do that and you see then how sensitive
[32:45] sensitive when when at which laser power your um
[32:48] when when at which laser power your um bit errors start to
[32:50] bit errors start to to to kick in and so that gives you a
[32:52] to to kick in and so that gives you a quality of your transmitter and your
[32:54] quality of your transmitter and your receiver link
[32:55] receiver link and that's a very
[32:57] and that's a very low cost because you don't need
[32:58] low cost because you don't need high-speed
[33:00] high-speed uh interconnect to that everything is
[33:02] uh interconnect to that everything is all the high speed is contained on the
[33:04] all the high speed is contained on the on the chipset
[33:06] on the chipset okay
[33:08] okay so that's
[33:15] peter we have many people who want to
[33:17] peter we have many people who want to ask you questions uh remember that there
[33:20] ask you questions uh remember that there are many important companies i want to
[33:22] are many important companies i want to work with you we have for example femto
[33:24] work with you we have for example femto printed minecraft domain we have
[33:25] printed minecraft domain we have companies like freedom photonics like
[33:27] companies like freedom photonics like ebg everyone here that you want to do
[33:30] ebg everyone here that you want to do something for them post in the chat your
[33:31] something for them post in the chat your room for cooperation and use the chat to
[33:33] room for cooperation and use the chat to interact with each other as much as
[33:35] interact with each other as much as possible but i want to keep the floor to
[33:36] possible but i want to keep the floor to a very close very dear friend of mine
[33:39] a very close very dear friend of mine mar lucovic from five reality thank you
[33:43] mar lucovic from five reality thank you very much for supporting me in this new
[33:45] very much for supporting me in this new job what's on your mind
[33:48] job what's on your mind mark you're muted
[33:51] sorry about that
[33:53] sorry about that uh yeah a question for uh peter at cisco
[33:56] uh yeah a question for uh peter at cisco uh i i seem to remember some advertising
[33:58] uh i i seem to remember some advertising my memory is as good as it used to be uh
[34:00] my memory is as good as it used to be uh remember of some advertising on
[34:03] remember of some advertising on you having discussion about cpl and i
[34:05] you having discussion about cpl and i can't remember your title slide
[34:07] can't remember your title slide but i was i thought i'd get you to talk
[34:09] but i was i thought i'd get you to talk about cpo and osf pxd what's the current
[34:13] about cpo and osf pxd what's the current thinking at cisco right now
[34:15] thinking at cisco right now um we believe that plugable modules will
[34:18] um we believe that plugable modules will always be there if the electrical
[34:21] always be there if the electrical connections work and especially you
[34:23] connections work and especially you might have heard the discussions on 200
[34:25] might have heard the discussions on 200 gigabit lambda if the
[34:28] gigabit lambda if the electrical interfaces work
[34:30] electrical interfaces work a
[34:30] a a pluggable module like the oc
[34:33] a pluggable module like the oc the osfpxd is definitely
[34:36] the osfpxd is definitely if you can get that to work will be used
[34:38] if you can get that to work will be used for sure co-packaged optics is way more
[34:41] for sure co-packaged optics is way more specialized it is very interesting
[34:44] specialized it is very interesting it potentially allows you to reduce the
[34:47] it potentially allows you to reduce the the power dissipation because that's
[34:48] the power dissipation because that's what it's all about if the power doesn't
[34:50] what it's all about if the power doesn't go down compared with plug on modules go
[34:52] go down compared with plug on modules go packaged optics makes no sense
[34:55] packaged optics makes no sense so um
[34:56] so um but copack stoptex is much more
[34:58] but copack stoptex is much more specialized and um it's um it is not as
[35:03] specialized and um it's um it is not as um universal as a pluggable module but
[35:05] um universal as a pluggable module but that's an interesting point because the
[35:07] that's an interesting point because the power needs to go down and as the the
[35:09] power needs to go down and as the the terabits per
[35:11] terabits per ethernet switch are increasing uh
[35:14] ethernet switch are increasing uh we're really hitting uh issues there
[35:16] we're really hitting uh issues there with pluggable modules
[35:18] with pluggable modules um before we go to the next speaker for
[35:21] um before we go to the next speaker for you peter the doubler and also from you
[35:24] you peter the doubler and also from you peter interval itself from bessie there
[35:26] peter interval itself from bessie there is a question
[35:28] is a question from siena young frederick gaines
[35:32] from siena young frederick gaines john frederick thank you very much for
[35:34] john frederick thank you very much for being with us today what's on your mind
[35:40] jim federicane could you please amuse
[35:43] jim federicane could you please amuse your microphone
[35:45] your microphone and tell us what's on your mind
[35:47] and tell us what's on your mind i will i will read his question because
[35:48] i will i will read his question because he texted it to me he said how small can
[35:52] he texted it to me he said how small can two electrical contacts be aligned
[35:56] two electrical contacts be aligned from the dye to wafer accounting for all
[35:59] from the dye to wafer accounting for all sorts of errors so how is how close can
[36:02] sorts of errors so how is how close can do to electrical contact be aligned
[36:08] peter the doubler
[36:10] peter the doubler oh i think this is a question for
[36:12] oh i think this is a question for pete
[36:15] because he makes a machine that can
[36:16] because he makes a machine that can align it so
[36:18] align it so i i mean we make the machine that can
[36:19] i i mean we make the machine that can align it and uh it then depends also on
[36:22] align it and uh it then depends also on the um like the the
[36:25] the um like the the our customers that supply the material
[36:27] our customers that supply the material that we do need to bond together what i
[36:29] that we do need to bond together what i can tell you is that our machine can
[36:31] can tell you is that our machine can with
[36:32] with quite good material can reach uh
[36:35] quite good material can reach uh alignment accuracies of even below 200
[36:37] alignment accuracies of even below 200 nanometers three sigma that's
[36:40] nanometers three sigma that's that's taken into account uh die sizes
[36:43] that's taken into account uh die sizes of up to 20 by 20 millimeters so even at
[36:46] of up to 20 by 20 millimeters so even at the corners we we can reach this so over
[36:48] the corners we we can reach this so over the whole dye surface area we can reach
[36:52] the whole dye surface area we can reach around 200 nanometers three sigma yeah
[36:55] around 200 nanometers three sigma yeah yes
[36:57] yes now with us so jim frederick did i voice
[36:59] now with us so jim frederick did i voice your question properly maybe you want to
[37:01] your question properly maybe you want to ask yourself yes very well and and that
[37:04] ask yourself yes very well and and that answers so no worries for electrical
[37:06] answers so no worries for electrical contacts for high speeds for example
[37:09] thank you
[37:12] thank you peter the traveler peter on that fight
[37:14] peter the traveler peter on that fight sir companies like size like like an
[37:17] sir companies like size like like an optic thank you very much for this first
[37:19] optic thank you very much for this first q a of the first online industry meeting
[37:22] q a of the first online industry meeting we're gonna continue now with the next
[37:23] we're gonna continue now with the next speaker but what a way what a way to
[37:26] speaker but what a way what a way to start we continue now we continue to the
[37:28] start we continue now we continue to the next topic in the agenda we want to
[37:31] next topic in the agenda we want to understand how to assist from this
[37:34] understand how to assist from this global industry network how to assist
[37:37] global industry network how to assist the darpa initiative when it comes to
[37:39] the darpa initiative when it comes to photonics packaging and tests and we
[37:40] photonics packaging and tests and we couldn't find anyone better because
[37:43] couldn't find anyone better because there is not than gordon killer tom
[37:46] there is not than gordon killer tom houseken a very dear person all of you
[37:48] houseken a very dear person all of you know found gordon killer and pushing
[37:50] know found gordon killer and pushing here to be the voice and to help us
[37:53] here to be the voice and to help us connect gordon thank you very much for
[37:56] connect gordon thank you very much for being with us today the floor and the
[37:58] being with us today the floor and the attention of everyone goes to darpa
[38:02] attention of everyone goes to darpa great uh good morning good evening
[38:05] great uh good morning good evening wherever you are hi everybody thanks for
[38:07] wherever you are hi everybody thanks for joining
[38:08] joining um let me share my screen
[38:11] share
[38:13] share all right
[38:15] all right okay so trying to stick to uh tom's
[38:17] okay so trying to stick to uh tom's recommendation of
[38:19] recommendation of of going with only hold on i think
[38:22] of going with only hold on i think only uh
[38:24] only uh six minutes i i couldn't
[38:26] six minutes i i couldn't only do one slide but uh
[38:28] only do one slide but uh i thought i'd kind of share
[38:31] i thought i'd kind of share the perspective of darpa here we have a
[38:33] the perspective of darpa here we have a number of programs in optical
[38:35] number of programs in optical microsystems which is all about
[38:37] microsystems which is all about packaging things very small very
[38:39] packaging things very small very accurately
[38:40] accurately and
[38:41] and we do a lot of work on sort of wafer
[38:43] we do a lot of work on sort of wafer scale photonics but ultimately you need
[38:45] scale photonics but ultimately you need to put a lot more in the package
[38:48] to put a lot more in the package just a reminder darpa is a
[38:50] just a reminder darpa is a us government agency that's really
[38:52] us government agency that's really charged with doing a lot of
[38:54] charged with doing a lot of development of innovation we've been
[38:56] development of innovation we've been around for
[38:57] around for quite some time
[38:59] quite some time and been working in microelectronics and
[39:02] and been working in microelectronics and a lot of system level things like gps
[39:04] a lot of system level things like gps and autonomous navigation
[39:06] and autonomous navigation and
[39:07] and ai machine learning
[39:09] ai machine learning but i'm going to talk just about
[39:11] but i'm going to talk just about integrated photonics and optical
[39:13] integrated photonics and optical microsystems i think today we're going
[39:15] microsystems i think today we're going to hear a whole lot about co-package
[39:17] to hear a whole lot about co-package optics
[39:18] optics and there's obviously other things that
[39:20] and there's obviously other things that are coming on board
[39:22] are coming on board in terms of of digital compute
[39:25] in terms of of digital compute there's optical interconnect sure and as
[39:27] there's optical interconnect sure and as optics moves
[39:29] optics moves closer to the package closer to the xpu
[39:33] closer to the package closer to the xpu we we need better packaging but there's
[39:35] we we need better packaging but there's also concepts like optical accelerators
[39:38] also concepts like optical accelerators where we might be able to put photonics
[39:40] where we might be able to put photonics in the package
[39:41] in the package and maybe not use fibers at all that
[39:43] and maybe not use fibers at all that would be very exciting and maybe drive
[39:46] would be very exciting and maybe drive volumes much higher than they are today
[39:49] volumes much higher than they are today another
[39:50] another direction we're going is what i kind of
[39:52] direction we're going is what i kind of call optical front end
[39:55] call optical front end this is using photonics for things like
[39:57] this is using photonics for things like ranging like lidar imaging using it for
[40:01] ranging like lidar imaging using it for laser column and free space optical
[40:03] laser column and free space optical communications obviously fibercom as
[40:05] communications obviously fibercom as well
[40:07] well optical displays and ar vr and optical
[40:11] optical displays and ar vr and optical packaging here actually has a lot of
[40:13] packaging here actually has a lot of similarities with some of the other
[40:14] similarities with some of the other stuff happening in compute
[40:17] stuff happening in compute third quadrant that i show you here is
[40:19] third quadrant that i show you here is is micro sensors at darpa we in
[40:21] is micro sensors at darpa we in particular have a lot of work going on
[40:23] particular have a lot of work going on in the areas of positioning and
[40:25] in the areas of positioning and navigation and timing so very
[40:29] navigation and timing so very sort of small chip scale
[40:31] sort of small chip scale clocks magnetometers gyros
[40:34] clocks magnetometers gyros accelerometers and also doing chem and
[40:35] accelerometers and also doing chem and biosensing and well we might be
[40:38] biosensing and well we might be worried about very precise sensors
[40:42] worried about very precise sensors a lot of this as a lot of the reduction
[40:44] a lot of this as a lot of the reduction in size is very applicable to wearables
[40:47] in size is very applicable to wearables which is another
[40:48] which is another huge market for photonics that i think
[40:50] huge market for photonics that i think is is going to be exploding pretty soon
[40:53] is is going to be exploding pretty soon and then finally
[40:55] and then finally i lump a lot of things here into this
[40:57] i lump a lot of things here into this emerging compute
[40:58] emerging compute where uh we're doing things in analog
[41:01] where uh we're doing things in analog domain signal processing microwave
[41:04] domain signal processing microwave generation and a lot of quantum systems
[41:06] generation and a lot of quantum systems quantum information processing and um
[41:09] quantum information processing and um quantum sensors
[41:11] quantum sensors uh
[41:13] uh i thought i would
[41:14] i thought i would without digging real deep into this just
[41:17] without digging real deep into this just kind of give you a highlight of
[41:20] kind of give you a highlight of my day i get to work on
[41:22] my day i get to work on many different programs with a huge
[41:24] many different programs with a huge number of
[41:26] number of uh fantastic technical groups many of
[41:28] uh fantastic technical groups many of which i actually saw on
[41:31] which i actually saw on online today
[41:33] online today the pypes program is one of our big
[41:35] the pypes program is one of our big investments in essentially co-packaged
[41:38] investments in essentially co-packaged optics
[41:39] optics i could talk your ear off about this but
[41:42] i could talk your ear off about this but we're seeing a lot of the same
[41:43] we're seeing a lot of the same challenges here
[41:45] challenges here you know looking at two and a half d and
[41:46] you know looking at two and a half d and 3d
[41:47] 3d electronic photonic packaging
[41:50] electronic photonic packaging fiber attach
[41:52] fiber attach high quality
[41:54] high quality laser development comb lasers
[41:56] laser development comb lasers the chips program is a two and a half
[41:59] the chips program is a two and a half and 3d
[42:00] and 3d integration program helping to drive
[42:04] integration program helping to drive chiplets which may be digital maybe rf
[42:07] chiplets which may be digital maybe rf maybe
[42:09] maybe optical as well
[42:12] optical as well the lumos program
[42:14] the lumos program is about bringing different photonics
[42:16] is about bringing different photonics together and this is essentially
[42:19] together and this is essentially enabling the world of heterogeneous
[42:20] enabling the world of heterogeneous photonics and i could talk about this
[42:23] photonics and i could talk about this quite a bit uh we are seeing teams doing
[42:26] quite a bit uh we are seeing teams doing microtransfer printing
[42:27] microtransfer printing doing um
[42:29] doing um heterogeneous laser integration hybrid
[42:31] heterogeneous laser integration hybrid laser integration maybe the most
[42:33] laser integration maybe the most important or interesting that's
[42:36] important or interesting that's i think going to change the industry
[42:38] i think going to change the industry is heteroepitaxy
[42:41] is heteroepitaxy and i noticed we've got some some folks
[42:43] and i noticed we've got some some folks from suny and aim photonics here
[42:45] from suny and aim photonics here and they've been working at with some of
[42:48] and they've been working at with some of their partners
[42:49] their partners growing three five quantum dot lasers on
[42:51] growing three five quantum dot lasers on 300 millimeter
[42:53] 300 millimeter silicon
[42:54] silicon and integrating thousands of devices per
[42:58] and integrating thousands of devices per reticle
[42:59] reticle in a
[43:00] in a full cmos
[43:01] full cmos fab
[43:02] fab that's really exciting it gets rid of
[43:04] that's really exciting it gets rid of some of the packaging challenges
[43:05] some of the packaging challenges actually
[43:07] actually moab is a lidar program
[43:11] moab is a lidar program chip scale phased arrays for
[43:12] chip scale phased arrays for non-mechanical beam steering griffin is
[43:15] non-mechanical beam steering griffin is about rf
[43:16] about rf uh very very quiet microwave oscillators
[43:19] uh very very quiet microwave oscillators based on photonics again an application
[43:21] based on photonics again an application where you might have photonics
[43:23] where you might have photonics inside the package with with no fibers
[43:25] inside the package with with no fibers to or from
[43:27] to or from and photonics is an enabler which goes
[43:30] and photonics is an enabler which goes beyond what you can do today
[43:31] beyond what you can do today and then dodos which is about
[43:34] and then dodos which is about optical frequency synthesizers on a chip
[43:37] optical frequency synthesizers on a chip it really pushes combs so i think the
[43:40] it really pushes combs so i think the last
[43:41] last slide i wanted to
[43:42] slide i wanted to without pointing out any one particular
[43:46] without pointing out any one particular team or technology just kind of give you
[43:48] team or technology just kind of give you a sense of that
[43:49] a sense of that the different types of packaging that
[43:51] the different types of packaging that we're seeing under pipes there's a lot
[43:53] we're seeing under pipes there's a lot of co-packaged optics where again
[43:55] of co-packaged optics where again there's two and a half in 3d micro
[43:57] there's two and a half in 3d micro electronic integration
[43:59] electronic integration um power is a very big driver here we're
[44:01] um power is a very big driver here we're going
[44:02] going to uh one pika joule per bit including
[44:04] to uh one pika joule per bit including all the electronics which means you have
[44:06] all the electronics which means you have to push it very close
[44:08] to push it very close to the electronics
[44:10] to the electronics many many fibers in some cases we've
[44:12] many many fibers in some cases we've seen um
[44:14] seen um teams that have 15 000 bumps
[44:17] teams that have 15 000 bumps per pick
[44:19] per pick teams that have more than 100 fibers
[44:22] teams that have more than 100 fibers attached to a single chiplet
[44:27] we're seeing integration 3d integration
[44:29] we're seeing integration 3d integration with
[44:29] with digital analog and photonics all
[44:31] digital analog and photonics all together in the same package
[44:35] again i mentioned all the different ways
[44:36] again i mentioned all the different ways people are attaching
[44:38] people are attaching lasers modulators and so on to different
[44:40] lasers modulators and so on to different materials this might be silicon silicon
[44:42] materials this might be silicon silicon nitride
[44:43] nitride uh thin film lithium niabate
[44:45] uh thin film lithium niabate uh
[44:46] uh moab program is a kind of a nice shot of
[44:49] moab program is a kind of a nice shot of a full reticle size pick
[44:52] a full reticle size pick that has in fact many tens of thousands
[44:55] that has in fact many tens of thousands of components on it
[44:57] of components on it it has
[44:58] it has many different
[45:01] many different asics attached to it uh which again
[45:04] asics attached to it uh which again there's thousands of bumps per asic
[45:08] there's thousands of bumps per asic and then
[45:09] and then light comes off of this chip and steers
[45:11] light comes off of this chip and steers in free space in two dimensions to do
[45:13] in free space in two dimensions to do fmcw lidar
[45:16] fmcw lidar griffin we're seeing a lot of work with
[45:18] griffin we're seeing a lot of work with stabilized micro cavities for ultra
[45:21] stabilized micro cavities for ultra narrow line width in particular um some
[45:24] narrow line width in particular um some results that are getting us
[45:25] results that are getting us lasers that are sub hertz line width
[45:29] lasers that are sub hertz line width using semiconductor lasers
[45:31] using semiconductor lasers and then finally dodos is kind of a
[45:33] and then finally dodos is kind of a intuitive force in complexity where
[45:35] intuitive force in complexity where there are
[45:36] there are there's tunable lasers
[45:38] there's tunable lasers narrow line with lasers that are pumping
[45:42] narrow line with lasers that are pumping silica wedge resonators silicon nitride
[45:44] silica wedge resonators silicon nitride resonators non-linear photonics to do
[45:48] resonators non-linear photonics to do frequency conversion and then
[45:50] frequency conversion and then high-speed electronics all in the same
[45:52] high-speed electronics all in the same package
[45:53] package so essentially i guess what i'd like to
[45:55] so essentially i guess what i'd like to say is there's there's a whole lot of
[45:57] say is there's there's a whole lot of photonics packaging that needs to come
[45:59] photonics packaging that needs to come about to enable all of this
[46:03] about to enable all of this and so
[46:04] and so we at darpa have a new initiative
[46:07] we at darpa have a new initiative this was just announced a couple weeks
[46:08] this was just announced a couple weeks ago
[46:10] ago and it is
[46:12] and it is called ngmm next generation
[46:14] called ngmm next generation microelectronics manufacturing
[46:16] microelectronics manufacturing this is
[46:18] this is a solicitation to the community all of
[46:21] a solicitation to the community all of you folks should have a look if you
[46:22] you folks should have a look if you think you can can
[46:24] think you can can weigh in
[46:25] weigh in and we are looking for innovation in
[46:28] and we are looking for innovation in advanced packaging
[46:30] advanced packaging we have uh investment money set aside to
[46:32] we have uh investment money set aside to do this we don't have
[46:34] do this we don't have as specific goals as sometimes we do
[46:36] as specific goals as sometimes we do because we're trying to do a phase zero
[46:38] because we're trying to do a phase zero and get a round table going where there
[46:40] and get a round table going where there are a number of voices that are willing
[46:42] are a number of voices that are willing to weigh in
[46:43] to weigh in what we'd like to do is help to stand up
[46:45] what we'd like to do is help to stand up a national facility to do early
[46:47] a national facility to do early prototyping and bring some of these
[46:50] prototyping and bring some of these needs together how do you do
[46:52] needs together how do you do 3d fine pitch packaging how do you do
[46:54] 3d fine pitch packaging how do you do how do you bring photonics and mams and
[46:56] how do you bring photonics and mams and rf to this
[46:58] rf to this and how do you develop some of the new
[46:59] and how do you develop some of the new packaging technologies to do prototyping
[47:02] packaging technologies to do prototyping um anyway ng m
[47:05] um anyway ng m it is uh out there i think the
[47:08] it is uh out there i think the proposals are due on september 21st
[47:12] proposals are due on september 21st with that
[47:12] with that let me stop and pass it back to jose
[47:15] let me stop and pass it back to jose thank you very much for a great
[47:17] thank you very much for a great presentation it is it means a lot for us
[47:19] presentation it is it means a lot for us to have
[47:20] to have darpa
[47:22] darpa announcing a new initiatives and room
[47:25] announcing a new initiatives and room for cooperation because this is what is
[47:27] for cooperation because this is what is all about gordon i have a difficult
[47:29] all about gordon i have a difficult question for you but i have to ask
[47:31] question for you but i have to ask because i represent a global global
[47:33] because i represent a global global industry network and you presented a
[47:35] industry network and you presented a national
[47:36] national national initiative
[47:38] national initiative is there room for companies outside
[47:41] is there room for companies outside united states and how we can how you can
[47:43] united states and how we can how you can help me connect the industry globally
[47:46] help me connect the industry globally yeah it's a great question so
[47:49] yeah it's a great question so i actually work
[47:50] i actually work very closely with many of the um
[47:53] very closely with many of the um many of the folks online here are who
[47:55] many of the folks online here are who are not from north america uh it becomes
[47:58] are not from north america uh it becomes always a bit of a delicate balance
[48:00] always a bit of a delicate balance because darpa's is funded uh
[48:03] because darpa's is funded uh through the us through tax dollars and
[48:06] through the us through tax dollars and so
[48:07] so tends to be a preference to build up
[48:09] tends to be a preference to build up capability
[48:11] capability domestically however everybody
[48:13] domestically however everybody acknowledges that it's a global
[48:15] acknowledges that it's a global marketplace and global innovation is
[48:18] marketplace and global innovation is important to embrace and to pursue
[48:21] important to embrace and to pursue and so you know
[48:22] and so you know wherever possible we like to work with
[48:24] wherever possible we like to work with all partners
[48:26] all partners and typically when we have a
[48:27] and typically when we have a solicitation it's it's made clear
[48:30] solicitation it's it's made clear uh
[48:31] uh what we're allowed to do within that one
[48:33] what we're allowed to do within that one and so
[48:34] and so this ngmm is not my program but i think
[48:37] this ngmm is not my program but i think it's
[48:38] it's um
[48:39] um spelled out in in that to see who's
[48:41] spelled out in in that to see who's who's allowed to participate
[48:44] who's allowed to participate there are many people with questions are
[48:46] there are many people with questions are known for cooperation in the room i'm
[48:48] known for cooperation in the room i'm gonna i'm gonna bring you first to
[48:50] gonna i'm gonna bring you first to something that we're gonna push from
[48:52] something that we're gonna push from optica in the coming years which is the
[48:54] optica in the coming years which is the micro optics revolution we believe that
[48:56] micro optics revolution we believe that many companies are really making a big
[48:58] many companies are really making a big push for having different wafer level
[49:01] push for having different wafer level optics manufacturing from different
[49:03] optics manufacturing from different processes and that's how the relation
[49:05] processes and that's how the relation revolutionized the industry i'm going to
[49:07] revolutionized the industry i'm going to present you to one company that maybe
[49:09] present you to one company that maybe you haven't heard from but they are
[49:10] you haven't heard from but they are becoming and growing really fast the
[49:12] becoming and growing really fast the company is fento print and the person is
[49:15] company is fento print and the person is my very dear friend rolando ferrini
[49:18] my very dear friend rolando ferrini rolando thank you very much being with
[49:20] rolando thank you very much being with us it means the world to me that you're
[49:21] us it means the world to me that you're here on the first online industry
[49:22] here on the first online industry meeting i would like if you could tell
[49:24] meeting i would like if you could tell us quickly
[49:25] us quickly what fentoprin does what is the
[49:27] what fentoprin does what is the manufacturing process of micro optics
[49:29] manufacturing process of micro optics aventa print and how you can address the
[49:31] aventa print and how you can address the packaging or maybe these sensors that
[49:34] packaging or maybe these sensors that they were putting on the table from
[49:35] they were putting on the table from darpa
[49:37] darpa thank you jose
[49:38] thank you jose it's a pleasure to be here
[49:40] it's a pleasure to be here so fanto print we have a technology
[49:43] so fanto print we have a technology platform and we do 3d printing or better
[49:46] platform and we do 3d printing or better writing into glass so we produce we
[49:49] writing into glass so we produce we might micromanufacture glass components
[49:52] might micromanufacture glass components that combine several functions we've
[49:54] that combine several functions we've heard a lot about co-packaging we heard
[49:56] heard a lot about co-packaging we heard a lot about miniaturization and
[49:57] a lot about miniaturization and combination of several functions so what
[49:59] combination of several functions so what we can do at fentoprint we can shape
[50:02] we can do at fentoprint we can shape micro shape the surf the the the bulk of
[50:04] micro shape the surf the the the bulk of the glass in three in three dimensions
[50:07] the glass in three in three dimensions and produce micro optical components or
[50:09] and produce micro optical components or systems this can be a combination of
[50:11] systems this can be a combination of micro optics
[50:13] micro optics micro lenses micro mirrors but we can
[50:16] micro lenses micro mirrors but we can combine also with interesting features
[50:17] combine also with interesting features we heard about the v grooves we can
[50:19] we heard about the v grooves we can combine these micro optical elements
[50:21] combine these micro optical elements with v grooves but also with
[50:23] with v grooves but also with 2d for rules for fiber arrays with a
[50:26] 2d for rules for fiber arrays with a precision that goes
[50:27] precision that goes below one micro we are today at the
[50:29] below one micro we are today at the precision and diameters and positioning
[50:31] precision and diameters and positioning of these holes about half a micron which
[50:34] of these holes about half a micron which is which is compatible what we heard a
[50:36] is which is compatible what we heard a few minutes ago we can combine with
[50:38] few minutes ago we can combine with waveguides and we can write in three
[50:40] waveguides and we can write in three dimensions into glass so we can offer
[50:42] dimensions into glass so we can offer really a full solution combining several
[50:45] really a full solution combining several functionalities monolithically in one
[50:48] functionalities monolithically in one single component going up to micro
[50:50] single component going up to micro mechanical features that could be added
[50:53] mechanical features that could be added so this is the entire set that can be uh
[50:56] so this is the entire set that can be uh put into a component the glass component
[50:59] put into a component the glass component that can be actually welded or combined
[51:02] that can be actually welded or combined or packaged with uh with uh with the
[51:06] or packaged with uh with uh with the electronics components are talking about
[51:08] electronics components are talking about and we can produce this chip by chip but
[51:10] and we can produce this chip by chip but also at wafer scale we routinely work
[51:12] also at wafer scale we routinely work today at eight inches already
[51:14] today at eight inches already grand as you can see in the presentation
[51:16] grand as you can see in the presentation from gordon he spoke about the moab the
[51:19] from gordon he spoke about the moab the mob initiative that is looking for a
[51:21] mob initiative that is looking for a free space coupling into a photonic
[51:24] free space coupling into a photonic internet circuit already packaged on a
[51:26] internet circuit already packaged on a cmos internet chip
[51:28] cmos internet chip that's something that you have been
[51:29] that's something that you have been looking at is that they are micro optics
[51:32] looking at is that they are micro optics for doing the freeze coupling between
[51:34] for doing the freeze coupling between a remote light source and the photonic
[51:37] a remote light source and the photonic integrity circuit
[51:38] integrity circuit what i can tell you is that today we're
[51:40] what i can tell you is that today we're working a lot and these are requests
[51:42] working a lot and these are requests from customers in solutions for coupling
[51:46] from customers in solutions for coupling let's say
[51:47] let's say fibers or fibers are raised with with
[51:49] fibers or fibers are raised with with pigs with photonic integrated circuits
[51:52] pigs with photonic integrated circuits and
[51:52] and and
[51:54] and let's say taking the two the two
[51:55] let's say taking the two the two approaches coming from above and then
[51:58] approaches coming from above and then turning the light at 90 degrees or
[52:00] turning the light at 90 degrees or coming from the edge and coupling
[52:02] coming from the edge and coupling directly into the optical chips there
[52:04] directly into the optical chips there are several solutions
[52:06] are several solutions and these are requests that we are
[52:07] and these are requests that we are hanging let's say taking and then trying
[52:10] hanging let's say taking and then trying with the customers to solve today with
[52:12] with the customers to solve today with our technology
[52:14] our technology gordon when it comes to this
[52:16] gordon when it comes to this mob free space coupling challenges could
[52:19] mob free space coupling challenges could you share with us a few more details
[52:21] you share with us a few more details about the the the the key challenge for
[52:24] about the the the the key challenge for the free space coupling into the
[52:25] the free space coupling into the photonic integrity circuit
[52:28] photonic integrity circuit yeah sure so um
[52:30] yeah sure so um as you scale up these arrays you might
[52:32] as you scale up these arrays you might want to do uh
[52:34] want to do uh higher and higher power and so uh
[52:36] higher and higher power and so uh coupling
[52:38] coupling a widely tunable narrow line with uh
[52:40] a widely tunable narrow line with uh chirpable laser into this
[52:43] chirpable laser into this system was was first done with fibers
[52:45] system was was first done with fibers we're now doing on-ship game on-ship
[52:48] we're now doing on-ship game on-ship lasers on-chip amplification to get rid
[52:50] lasers on-chip amplification to get rid of a lot of that
[52:52] of a lot of that free space coupling
[52:54] free space coupling but i would say that you bring up a
[52:56] but i would say that you bring up a really good
[52:58] really good segue i think
[52:59] segue i think that often we're talking about how do
[53:01] that often we're talking about how do you couple from a pick to a fiber but
[53:04] you couple from a pick to a fiber but what we're seeing in our um microsystem
[53:06] what we're seeing in our um microsystem programs is that you need to
[53:08] programs is that you need to uh
[53:10] uh very efficiently couple into many
[53:11] very efficiently couple into many different things into free space
[53:14] different things into free space coupled to
[53:15] coupled to micro cavities coupled to
[53:18] micro cavities coupled to atomic vapor cells
[53:20] atomic vapor cells uh coupled to you know trapped atoms
[53:23] uh coupled to you know trapped atoms um
[53:24] um and
[53:25] and micro optics especially you know 3d
[53:28] micro optics especially you know 3d um additive manufacturing types of
[53:31] um additive manufacturing types of approaches are really compelling to do a
[53:33] approaches are really compelling to do a lot of these different things i would
[53:34] lot of these different things i would say
[53:36] say another topic that i wanted to address
[53:37] another topic that i wanted to address with you gordon because there are many
[53:39] with you gordon because there are many optical corporate members that can help
[53:41] optical corporate members that can help here is the topic of of copaceptics
[53:44] here is the topic of of copaceptics we are going to have many different
[53:46] we are going to have many different meetings and i hereby invite you to our
[53:48] meetings and i hereby invite you to our event in in silicon valley next
[53:51] event in in silicon valley next next month of october but
[53:54] next month of october but one thing that we really see from the
[53:56] one thing that we really see from the package optics approach is the need for
[53:58] package optics approach is the need for new connector with very high level of
[54:00] new connector with very high level of integration so with from optica we
[54:03] integration so with from optica we started scouting technologies that could
[54:04] started scouting technologies that could help the aligning of a large amount of
[54:07] help the aligning of a large amount of fibers and we found out one company that
[54:11] fibers and we found out one company that was not very well known by many of you
[54:13] was not very well known by many of you the company is called it's called
[54:15] the company is called it's called microalign and i have somebody here from
[54:18] microalign and i have somebody here from microalign the person is dario location
[54:20] microalign the person is dario location dario good afternoon i believe in the
[54:23] dario good afternoon i believe in the netherlands good afternoon can you tell
[54:25] netherlands good afternoon can you tell us a bit of microalign and how would you
[54:27] us a bit of microalign and how would you help the challenge from darpa on the
[54:29] help the challenge from darpa on the package especially on those a
[54:32] package especially on those a high level integration of many fibers
[54:35] high level integration of many fibers yeah thank you jose and indeed i'm in
[54:37] yeah thank you jose and indeed i'm in the netherlands
[54:39] the netherlands so um i have one slide
[54:41] so um i have one slide um
[54:42] um i'm not sure how i share that bottom of
[54:45] i'm not sure how i share that bottom of the window there's a green arrow that
[54:47] the window there's a green arrow that says screen click on it and that's very
[54:49] says screen click on it and that's very important select the window that you
[54:50] important select the window that you want to share with the world
[54:52] want to share with the world this one
[54:55] so
[54:56] so you probably can see the slide now
[54:58] you probably can see the slide now crystal clear
[54:59] crystal clear okay what we have done is we developed a
[55:03] okay what we have done is we developed a reconfigurable v-groove
[55:05] reconfigurable v-groove in order to align not the claddings of
[55:08] in order to align not the claddings of the fibers but the cores
[55:11] the fibers but the cores so this way the pitch of the course is
[55:14] so this way the pitch of the course is 125 in the end
[55:16] 125 in the end currently we're not there yet
[55:19] currently we're not there yet and you can move each fiber individually
[55:22] and you can move each fiber individually at the same time
[55:23] at the same time [Music]
[55:24] [Music] and what the idea behind this is that
[55:27] and what the idea behind this is that you solve the
[55:29] you solve the difference between coupling efficiency
[55:31] difference between coupling efficiency between the different channels
[55:33] between the different channels probably most usable for sending
[55:36] probably most usable for sending applications
[55:38] applications so actually what we're looking for now
[55:40] so actually what we're looking for now is the most
[55:42] is the most suitable
[55:44] suitable market or application for
[55:46] market or application for further developing
[55:48] further developing this product
[55:50] this product so maybe
[55:53] so maybe gordon can help us in
[55:55] gordon can help us in defining what area we should look there
[56:00] gordon you can see they have an idea to
[56:02] gordon you can see they have an idea to have a very high
[56:04] have a very high level
[56:05] level alignment of each fiber individually
[56:08] alignment of each fiber individually into an array and that can be done can
[56:10] into an array and that can be done can be used by any kind of sensor or
[56:12] be used by any kind of sensor or anything photonic chip that has many
[56:15] anything photonic chip that has many fiber input outputs can you think about
[56:18] fiber input outputs can you think about an application perhaps related to
[56:20] an application perhaps related to quantum technologies or perhaps related
[56:23] quantum technologies or perhaps related to package optics that could be
[56:24] to package optics that could be benefited by this
[56:26] benefited by this sure so yeah i will i will uh say that
[56:30] sure so yeah i will i will uh say that yes we see a lot of uh
[56:32] yes we see a lot of uh a lot of folks that are having
[56:34] a lot of folks that are having challenges using sort of uh passive or
[56:37] challenges using sort of uh passive or even active alignment of arrays where
[56:39] even active alignment of arrays where you might be only looking at uh coupling
[56:41] you might be only looking at uh coupling of
[56:42] of say the two end fibers and what you get
[56:44] say the two end fibers and what you get in the middle is not nearly uh as good
[56:46] in the middle is not nearly uh as good as what you need
[56:47] as what you need um so this definitely fills uh a need
[56:50] um so this definitely fills uh a need for certain applications where insertion
[56:53] for certain applications where insertion loss is uh
[56:55] loss is uh highly
[56:56] highly important microline is looking for
[56:59] important microline is looking for partners thank you very much microline
[57:00] partners thank you very much microline for being corporate member of optica
[57:02] for being corporate member of optica like many of you are a one thing that i
[57:05] like many of you are a one thing that i would like to ask you before we go to
[57:07] would like to ask you before we go to the next presentation it is about what
[57:11] the next presentation it is about what you present what you put here in your
[57:13] you present what you put here in your slide
[57:14] slide related to ultra narrow language
[57:17] related to ultra narrow language and let me just share your slide again
[57:20] and let me just share your slide again to talk about ultra language and micro
[57:22] to talk about ultra language and micro cavities do you have some kind of
[57:25] cavities do you have some kind of challenges specifications in the pipes
[57:28] challenges specifications in the pipes project you share with us the specs is
[57:30] project you share with us the specs is there any challenges any dream or any
[57:32] there any challenges any dream or any dream specifications about the ultra
[57:34] dream specifications about the ultra narrow language
[57:36] narrow language for a given laser that giveaway
[57:38] for a given laser that giveaway well what i can can tell you about
[57:40] well what i can can tell you about griffin actually is that we're
[57:42] griffin actually is that we're using optical frequency division
[57:44] using optical frequency division techniques
[57:45] techniques um
[57:46] um to create very very narrow line width
[57:49] to create very very narrow line width microwave sources
[57:50] microwave sources right and so um
[57:52] right and so um we all use oscillators every day right
[57:54] we all use oscillators every day right they're in your phone they're in your
[57:55] they're in your phone they're in your computer but the the really high
[57:57] computer but the the really high precision oscillators are used in in
[58:00] precision oscillators are used in in communications
[58:01] communications um like in a in a 5g cell tower and
[58:04] um like in a in a 5g cell tower and they're used in radar systems and other
[58:06] they're used in radar systems and other kinds of you know imaging radar maybe an
[58:09] kinds of you know imaging radar maybe an autonomous vehicle
[58:11] autonomous vehicle it turns out with optical frequency
[58:13] it turns out with optical frequency division
[58:14] division you can make the very uh best microwave
[58:17] you can make the very uh best microwave oscillators on the planet
[58:19] oscillators on the planet better than you can use using
[58:21] better than you can use using traditional techniques
[58:22] traditional techniques griffin is trying to
[58:24] griffin is trying to use some of these advances in in really
[58:27] use some of these advances in in really quiet lasers on chip
[58:29] quiet lasers on chip and then divide that
[58:32] and then divide that narrow line width down to
[58:34] narrow line width down to sort of gigahertz
[58:36] sort of gigahertz ranges so that you've got very very
[58:38] ranges so that you've got very very quiet microwave sources i think this is
[58:40] quiet microwave sources i think this is a uh under appreciated but really
[58:43] a uh under appreciated but really compelling approach for photonics
[58:45] compelling approach for photonics because you could see it in
[58:46] because you could see it in every uh base station in a few years i
[58:49] every uh base station in a few years i think
[58:51] think according there is a person that i want
[58:52] according there is a person that i want you to meet regarding this microwave
[58:55] you to meet regarding this microwave generation technologies and this person
[58:57] generation technologies and this person is guillermo carpintero from today from
[59:00] is guillermo carpintero from today from the university that can lost
[59:03] the university that can lost madrid but very soon from
[59:05] madrid but very soon from i don't know if i can say let's start
[59:08] i don't know if i can say let's start anyway
[59:10] see the the griefing
[59:12] see the the griefing initiative
[59:14] initiative guillermo can you tell us what you do
[59:16] guillermo can you tell us what you do and what kind of room for cooperation
[59:17] and what kind of room for cooperation brings to your mind
[59:20] we are
[59:21] we are actually working on uh developing
[59:24] actually working on uh developing photonic integrated circuits for a
[59:26] photonic integrated circuits for a millimeter terahertz wave generation
[59:29] millimeter terahertz wave generation and actually we have worked with ultra
[59:31] and actually we have worked with ultra narrow line with lasers we are
[59:33] narrow line with lasers we are collaborating with lyonix in that part
[59:36] collaborating with lyonix in that part because they have very high q resonators
[59:40] because they have very high q resonators in silicon nitride which make a very
[59:42] in silicon nitride which make a very narrow line with lasers and what we are
[59:44] narrow line with lasers and what we are actually working is that
[59:46] actually working is that still when you have a very narrow line
[59:48] still when you have a very narrow line with laser you still have to control the
[59:50] with laser you still have to control the drift between the two lasers and we are
[59:52] drift between the two lasers and we are evaluating how that
[59:54] evaluating how that plays off when you have integrated
[59:56] plays off when you have integrated the sources on a single chip compared to
[01:00:00] the sources on a single chip compared to different sources actually on with
[01:00:03] different sources actually on with different gain materials because
[01:00:04] different gain materials because actually when you do
[01:00:07] actually when you do hybrid integration you can choose even
[01:00:09] hybrid integration you can choose even if you have the same gain material and
[01:00:12] if you have the same gain material and you have the same passive substrate or
[01:00:14] you have the same passive substrate or you combine
[01:00:15] you combine the all the gain chips in in
[01:00:18] the all the gain chips in in in a single indian phosphate chip
[01:00:20] in a single indian phosphate chip and on top of that
[01:00:22] and on top of that once we generate a high frequency the
[01:00:24] once we generate a high frequency the problem is how to extract the high
[01:00:26] problem is how to extract the high frequency from the semiconductor because
[01:00:30] frequency from the semiconductor because it usually prefers to stay in a high
[01:00:33] it usually prefers to stay in a high refractive index material
[01:00:34] refractive index material and
[01:00:35] and we are
[01:00:37] we are actually working as well in the
[01:00:39] actually working as well in the connectors that will allow us to extract
[01:00:41] connectors that will allow us to extract the high frequency from the chips
[01:00:44] the high frequency from the chips uh out into connectors that actually
[01:00:47] uh out into connectors that actually have low losses because uh the problem
[01:00:49] have low losses because uh the problem is that in optics it's we work in
[01:00:52] is that in optics it's we work in microwave photons because optics is a
[01:00:55] microwave photons because optics is a low loss waveguide for high frequency
[01:00:57] low loss waveguide for high frequency but as soon as you
[01:00:58] but as soon as you convert it into it back into electrical
[01:01:01] convert it into it back into electrical you always hit the wall that you have a
[01:01:03] you always hit the wall that you have a low high losses
[01:01:05] low high losses in coplanar wavelengths and that's the
[01:01:08] in coplanar wavelengths and that's the problems that we are working on
[01:01:12] thank you jose for the opportunity
[01:01:14] thank you jose for the opportunity and thank you thank you guillermo for
[01:01:17] and thank you thank you guillermo for starting that company and you're going
[01:01:19] starting that company and you're going for upcoming meetings we would like to
[01:01:21] for upcoming meetings we would like to hear more because i have really high
[01:01:23] hear more because i have really high hopes for this thing that you are doing
[01:01:25] hopes for this thing that you are doing gordon you should know that guillermo
[01:01:26] gordon you should know that guillermo has been a one of the leading people in
[01:01:28] has been a one of the leading people in the microwave photonics push that we
[01:01:31] the microwave photonics push that we have from europe and it is great to hack
[01:01:32] have from europe and it is great to hack his support here at optica gordon many
[01:01:35] his support here at optica gordon many people are gonna contact you offline
[01:01:37] people are gonna contact you offline because there are things that they
[01:01:38] because there are things that they shouldn't say publicly it's what they
[01:01:39] shouldn't say publicly it's what they are telling me so we're going to move on
[01:01:41] are telling me so we're going to move on to an expected but please cordon stay
[01:01:42] to an expected but please cordon stay with us because the next topic is very
[01:01:44] with us because the next topic is very interesting for you
[01:01:45] interesting for you everyone when we announced that we were
[01:01:48] everyone when we announced that we were going to have a meeting on photonics
[01:01:50] going to have a meeting on photonics packaging many of you told us you need
[01:01:52] packaging many of you told us you need to bring lidar and if possible you need
[01:01:54] to bring lidar and if possible you need to bring luminar to the table we did our
[01:01:57] to bring luminar to the table we did our best we contacted luminar contacted
[01:02:00] best we contacted luminar contacted jason we talked to them and they told us
[01:02:03] jason we talked to them and they told us if we're gonna talk about photonics
[01:02:04] if we're gonna talk about photonics packaging you need to put freedom
[01:02:07] packaging you need to put freedom photonics on the table or latest
[01:02:09] photonics on the table or latest acquisition to tell us what they are
[01:02:11] acquisition to tell us what they are gonna do for packaging but so tell us
[01:02:13] gonna do for packaging but so tell us how others can help them do it better so
[01:02:15] how others can help them do it better so we brought gordon morrison from freedom
[01:02:19] we brought gordon morrison from freedom photonics a luminar company to the room
[01:02:21] photonics a luminar company to the room gordon thank you thank you so much for
[01:02:24] gordon thank you thank you so much for being with us today the floor and the
[01:02:26] being with us today the floor and the attention of everyone goes to the
[01:02:28] attention of everyone goes to the packaging of fncw lidar the floor is
[01:02:30] packaging of fncw lidar the floor is yours
[01:02:32] yours okay well thank you for that uh
[01:02:34] okay well thank you for that uh enthusiastic uh introduction i've really
[01:02:36] enthusiastic uh introduction i've really appreciated uh your your energy in
[01:02:38] appreciated uh your your energy in hosting this let me just pull up uh my
[01:02:41] hosting this let me just pull up uh my slides here
[01:02:51] and then you select the window that you
[01:02:53] and then you select the window that you want to share
[01:02:56] you go to slide show mode presentation
[01:02:58] you go to slide show mode presentation mode yeah
[01:03:01] there we go
[01:03:02] there we go perfect
[01:03:04] perfect okay
[01:03:05] okay so uh
[01:03:06] so uh yeah um my name is gordon morrison uh
[01:03:09] yeah um my name is gordon morrison uh i'm vice president of engineering at
[01:03:11] i'm vice president of engineering at freedom photonics uh and as was
[01:03:13] freedom photonics uh and as was mentioned we were we were acquired by
[01:03:15] mentioned we were we were acquired by luminar
[01:03:17] luminar about five months ago
[01:03:20] about five months ago we are a manufacturer of photonic
[01:03:22] we are a manufacturer of photonic integrated circuits so
[01:03:25] integrated circuits so we've been doing that for for you know a
[01:03:27] we've been doing that for for you know a good 15 or more than 15 years now and
[01:03:32] good 15 or more than 15 years now and dialed lasers picks photo photo
[01:03:34] dialed lasers picks photo photo detectors
[01:03:36] detectors for applications that include lidar but
[01:03:38] for applications that include lidar but also
[01:03:40] also quantum sensing remote sensing
[01:03:43] quantum sensing remote sensing and lidar of course and also telecom
[01:03:45] and lidar of course and also telecom applications so we've been working in
[01:03:47] applications so we've been working in all those fields and have been dealing
[01:03:48] all those fields and have been dealing with
[01:03:49] with packaging problems that frankly i think
[01:03:51] packaging problems that frankly i think are common
[01:03:53] are common uh quite often between all those
[01:03:54] uh quite often between all those different application areas
[01:03:56] different application areas now of course one of our heaviest
[01:03:58] now of course one of our heaviest focuses is in the uh is in the lidar
[01:04:00] focuses is in the uh is in the lidar space
[01:04:04] so uh i with that in mind i've kind of
[01:04:06] so uh i with that in mind i've kind of put together a slide focusing on lidar
[01:04:09] put together a slide focusing on lidar because as i mentioned many of the
[01:04:10] because as i mentioned many of the problems uh in remote sensing in quantum
[01:04:13] problems uh in remote sensing in quantum sensing
[01:04:15] sensing in telecom
[01:04:17] in telecom they're the same problems in in lidar uh
[01:04:20] they're the same problems in in lidar uh but actually i would say lidar pulls all
[01:04:22] but actually i would say lidar pulls all of those problems into one place uh
[01:04:24] of those problems into one place uh makes it a very complex problem
[01:04:26] makes it a very complex problem so uh what are the things that go into a
[01:04:28] so uh what are the things that go into a into a lidar transceiver uh which is
[01:04:31] into a lidar transceiver uh which is sending out the light and receiving it
[01:04:32] sending out the light and receiving it as well well
[01:04:34] as well well you've got a laser chip on carrier uh of
[01:04:36] you've got a laser chip on carrier uh of some kind maybe the chip goes directly
[01:04:38] some kind maybe the chip goes directly on silicon maybe it doesn't but you've
[01:04:40] on silicon maybe it doesn't but you've got this this laser chip that has to be
[01:04:42] got this this laser chip that has to be connected to the outside world both
[01:04:43] connected to the outside world both thermally and electrically
[01:04:45] thermally and electrically um we need extreme cleanliness here too
[01:04:48] um we need extreme cleanliness here too these are these lasers um are not your
[01:04:50] these are these lasers um are not your standard telecom diode laser putting out
[01:04:52] standard telecom diode laser putting out a few milliwatts of optical power we're
[01:04:54] a few milliwatts of optical power we're talking watts of power
[01:04:56] talking watts of power and as such if there's any contamination
[01:04:59] and as such if there's any contamination anywhere uh you know you can you can run
[01:05:02] anywhere uh you know you can you can run into reliability issues much more easily
[01:05:04] into reliability issues much more easily if there isn't really really good
[01:05:05] if there isn't really really good cleanliness
[01:05:06] cleanliness uh we've got uh high dynamic range
[01:05:09] uh we've got uh high dynamic range necessary here so
[01:05:11] necessary here so you know if you're looking at a stop
[01:05:13] you know if you're looking at a stop sign real close to your your your
[01:05:15] sign real close to your your your transceiver um it's putting in a a whole
[01:05:18] transceiver um it's putting in a a whole lot of power whereas if you're looking
[01:05:19] lot of power whereas if you're looking to black dog 300 meters away on a road
[01:05:22] to black dog 300 meters away on a road you're talking about a lot less photons
[01:05:24] you're talking about a lot less photons and if you're using the same unit you
[01:05:26] and if you're using the same unit you have to be really clever to make sure
[01:05:28] have to be really clever to make sure that when you're looking at the black
[01:05:29] that when you're looking at the black dog you can see it and when you're
[01:05:30] dog you can see it and when you're looking at the nearby stop sign you're
[01:05:32] looking at the nearby stop sign you're not
[01:05:33] not blowing a hole and leaving a crater
[01:05:35] blowing a hole and leaving a crater where your photodiodes used to be
[01:05:37] where your photodiodes used to be uh we also need to integrate with
[01:05:39] uh we also need to integrate with electronics as
[01:05:41] electronics as many people have already alluded to
[01:05:43] many people have already alluded to this is a common problem to many other
[01:05:45] this is a common problem to many other areas but in our case we need to avoid
[01:05:48] areas but in our case we need to avoid parasitic inductance and capacitance
[01:05:50] parasitic inductance and capacitance when we're pulsing that laser with high
[01:05:51] when we're pulsing that laser with high powers you know laser trimming of
[01:05:53] powers you know laser trimming of passive components to get uh to get the
[01:05:55] passive components to get uh to get the signals just right is is not a
[01:05:56] signals just right is is not a cost-effective solution so we've got our
[01:05:59] cost-effective solution so we've got our own asic
[01:06:02] own asic design house of course uh black forest
[01:06:04] design house of course uh black forest was was was also acquired by by luminar
[01:06:07] was was was also acquired by by luminar uh and we're working together to
[01:06:08] uh and we're working together to integrate all that into into one tiny
[01:06:10] integrate all that into into one tiny little package
[01:06:12] little package optical coupling of course is also a
[01:06:14] optical coupling of course is also a problem uh there's there's optical
[01:06:16] problem uh there's there's optical coupling to fiber involved there's
[01:06:18] coupling to fiber involved there's optical coupling to free space of course
[01:06:20] optical coupling to free space of course involved
[01:06:21] involved and all those alignments uh are really
[01:06:23] and all those alignments uh are really really critical and for all sorts of
[01:06:25] really critical and for all sorts of lidar companies and different different
[01:06:27] lidar companies and different different companies have different approaches of
[01:06:28] companies have different approaches of course
[01:06:30] course there is a fundamental problem of how do
[01:06:32] there is a fundamental problem of how do you get light out of a tiny little
[01:06:34] you get light out of a tiny little laser diode where you're talking about a
[01:06:36] laser diode where you're talking about a one micron kind of diameter waveguide um
[01:06:40] one micron kind of diameter waveguide um hit something 300 meters away and then
[01:06:42] hit something 300 meters away and then couple it back into a very very small
[01:06:44] couple it back into a very very small space possibly another wave guide
[01:06:46] space possibly another wave guide um we've got micro mechanical type stuff
[01:06:50] um we've got micro mechanical type stuff and in our case we're using uh using
[01:06:52] and in our case we're using uh using rotating mirrors but there's other
[01:06:53] rotating mirrors but there's other approaches as well
[01:06:55] approaches as well uh so there's there's a lot of vibration
[01:06:57] uh so there's there's a lot of vibration considerations in there as well that
[01:06:59] considerations in there as well that that come to mind and then finally
[01:07:01] that come to mind and then finally there's the thermal management of all
[01:07:02] there's the thermal management of all this um
[01:07:03] this um the the environment you're working in is
[01:07:05] the the environment you're working in is brutal you're you're talking about way
[01:07:07] brutal you're you're talking about way below zero freezing temperatures and
[01:07:10] below zero freezing temperatures and going up near the boiling water
[01:07:11] going up near the boiling water temperatures as well and you've got this
[01:07:13] temperatures as well and you've got this uh
[01:07:14] uh high power laser that has to has to
[01:07:17] high power laser that has to has to perform across all those uh all those
[01:07:19] perform across all those uh all those temperatures so
[01:07:20] temperatures so if there's anybody out there who can
[01:07:21] if there's anybody out there who can make a better tech than what's available
[01:07:23] make a better tech than what's available right now i'd love to hear about it but
[01:07:25] right now i'd love to hear about it but we're doing everything we can with
[01:07:26] we're doing everything we can with existing technology to to
[01:07:29] existing technology to to optimize and minimize power consumption
[01:07:31] optimize and minimize power consumption of course
[01:07:32] of course and these problems that i'm i'm listing
[01:07:35] and these problems that i'm i'm listing here i would argue are are common in all
[01:07:38] here i would argue are are common in all cases uh to at least one example of
[01:07:41] cases uh to at least one example of atomic sensing of remote sensing uh and
[01:07:45] atomic sensing of remote sensing uh and in telecom which are all areas that
[01:07:46] in telecom which are all areas that we've been playing in
[01:07:50] another another area that's that's kind
[01:07:52] another another area that's that's kind of come up we found over time is uh
[01:07:55] of come up we found over time is uh that's a that's a problem and which
[01:07:57] that's a that's a problem and which we've made our best attempt to to find a
[01:07:59] we've made our best attempt to to find a solution for is reliability testing um
[01:08:03] solution for is reliability testing um the the world of technology as you're
[01:08:04] the the world of technology as you're all aware is one that moves really fast
[01:08:07] all aware is one that moves really fast uh you can't go perfecting uh a complex
[01:08:11] uh you can't go perfecting uh a complex system and then afterwards start
[01:08:12] system and then afterwards start reliability testing for a year to see
[01:08:14] reliability testing for a year to see whether or not it truly works or if it's
[01:08:16] whether or not it truly works or if it's going to fail for some reason
[01:08:17] going to fail for some reason um so generally we have to be kind of
[01:08:21] um so generally we have to be kind of you know building the airplane while
[01:08:23] you know building the airplane while it's in the air a little bit here so we
[01:08:25] it's in the air a little bit here so we need to get reliability testing started
[01:08:27] need to get reliability testing started on chips that may not actually be the
[01:08:29] on chips that may not actually be the final product it may not have quite the
[01:08:31] final product it may not have quite the right number of contacts it may have a
[01:08:33] right number of contacts it may have a slightly different power consumption and
[01:08:35] slightly different power consumption and how do you do that
[01:08:37] how do you do that without building a new rack every single
[01:08:40] without building a new rack every single time
[01:08:41] time there are there are services out there
[01:08:43] there are there are services out there where you can get chip on carrier lasers
[01:08:44] where you can get chip on carrier lasers uh measured but we're talking about
[01:08:46] uh measured but we're talking about photonic integrated circuits that have
[01:08:48] photonic integrated circuits that have many many contacts they've got
[01:08:50] many many contacts they've got modulators they've got uh soas shutters
[01:08:54] modulators they've got uh soas shutters high power components tunable components
[01:08:57] high power components tunable components and every single one is different so
[01:08:59] and every single one is different so what we've done uh at least for for
[01:09:01] what we've done uh at least for for preliminary reliability tests we've had
[01:09:03] preliminary reliability tests we've had to build our own racks that are very
[01:09:05] to build our own racks that are very very very reconfigurable allowing us to
[01:09:07] very very reconfigurable allowing us to introduce a wide array of multiple
[01:09:10] introduce a wide array of multiple current sources to any one component
[01:09:14] current sources to any one component so each of these racks will do 60
[01:09:15] so each of these racks will do 60 different parts and allows us to test
[01:09:17] different parts and allows us to test modulators shutters
[01:09:19] modulators shutters diodes high power low power lasers uh
[01:09:22] diodes high power low power lasers uh and has automated testing built in that
[01:09:24] and has automated testing built in that way when we've designed a chip
[01:09:26] way when we've designed a chip um even before we we've we've completed
[01:09:29] um even before we we've we've completed the you know the the the wheel between
[01:09:31] the you know the the the wheel between design test uh prototyping and back back
[01:09:35] design test uh prototyping and back back into redesign we've already got these
[01:09:37] into redesign we've already got these things uh running through a liability to
[01:09:39] things uh running through a liability to make sure there's no gotchas in advance
[01:09:41] make sure there's no gotchas in advance i think i think that sort of thing is is
[01:09:43] i think i think that sort of thing is is really useful and i i hope that you know
[01:09:45] really useful and i i hope that you know services like this will be
[01:09:47] services like this will be become more common in the future because
[01:09:49] become more common in the future because it it certainly uh is a requirement that
[01:09:51] it it certainly uh is a requirement that we have
[01:09:54] we have um another area that that we've got in
[01:09:56] um another area that that we've got in our back pocket that i think some people
[01:09:58] our back pocket that i think some people will find interesting is photonic wire
[01:10:00] will find interesting is photonic wire bonding uh we have a tool from vanguard
[01:10:03] bonding uh we have a tool from vanguard where we built our own uh you know
[01:10:05] where we built our own uh you know high-end clean room for for
[01:10:06] high-end clean room for for environmental control
[01:10:08] environmental control uh within our glass 10k manufacturing uh
[01:10:12] uh within our glass 10k manufacturing uh packaging clean room this is a method we
[01:10:14] packaging clean room this is a method we use for for connecting chips to chips
[01:10:17] use for for connecting chips to chips and chips to fiber
[01:10:19] and chips to fiber in particular with complex problems
[01:10:21] in particular with complex problems where we have
[01:10:22] where we have either silicon photonics chips or plcs
[01:10:25] either silicon photonics chips or plcs passive light wave circuits with
[01:10:27] passive light wave circuits with multiple ports needing say an array of
[01:10:29] multiple ports needing say an array of lasers an array of tunable lasers an
[01:10:31] lasers an array of tunable lasers an array of dfb lasers sometimes more than
[01:10:34] array of dfb lasers sometimes more than one array of dfb lasers for instance
[01:10:36] one array of dfb lasers for instance uh that sort of that sort of packaging
[01:10:39] uh that sort of that sort of packaging problem is not easy to solve we've done
[01:10:41] problem is not easy to solve we've done it before with mechanical uh you know
[01:10:44] it before with mechanical uh you know piezoelectric uh six axis tools where
[01:10:47] piezoelectric uh six axis tools where it's a very boutique approach but when
[01:10:50] it's a very boutique approach but when you start talking about you know 16 18
[01:10:52] you start talking about you know 16 18 24 lasers all in a row with all the
[01:10:55] 24 lasers all in a row with all the bending and the different thicknesses of
[01:10:57] bending and the different thicknesses of of the chip itself varying and you need
[01:10:59] of the chip itself varying and you need kind of sub micron alignment it becomes
[01:11:02] kind of sub micron alignment it becomes almost impossible to do that with this
[01:11:05] almost impossible to do that with this system
[01:11:06] system you can package things up and have you
[01:11:08] you can package things up and have you know 10 microns of slop here and there
[01:11:10] know 10 microns of slop here and there no problem it's very easy for us to do
[01:11:12] no problem it's very easy for us to do with our existing manufacturing tools
[01:11:14] with our existing manufacturing tools you come in afterwards and use the uh
[01:11:17] you come in afterwards and use the uh high resolution um
[01:11:19] high resolution um cameron optics to locate the wave guides
[01:11:21] cameron optics to locate the wave guides and you just
[01:11:22] and you just write your waveguide from point a to
[01:11:24] write your waveguide from point a to point b moreover you are
[01:11:27] point b moreover you are also able to
[01:11:28] also able to mode match so instead of designing your
[01:11:30] mode match so instead of designing your dfb laser uh to be not the best in class
[01:11:34] dfb laser uh to be not the best in class dfb laser but to be a good dfp laser
[01:11:37] dfb laser but to be a good dfp laser without with a spot size converter and
[01:11:39] without with a spot size converter and instead of having to put all kinds of
[01:11:40] instead of having to put all kinds of complex spot size converters on on your
[01:11:42] complex spot size converters on on your plc or photonic chip
[01:11:44] plc or photonic chip you can in fact match the the wave guide
[01:11:47] you can in fact match the the wave guide to each of those and adiabatically taper
[01:11:49] to each of those and adiabatically taper across to the next one and you end up
[01:11:51] across to the next one and you end up with a
[01:11:52] with a a nice reproducible optical coupling
[01:11:55] a nice reproducible optical coupling between multiple ports on on two chips
[01:11:58] between multiple ports on on two chips uh so this is a new technology it'll be
[01:12:00] uh so this is a new technology it'll be interesting to see where it goes we're
[01:12:02] interesting to see where it goes we're using it for for prototyping right now
[01:12:04] using it for for prototyping right now i'm certain and confident that it will
[01:12:06] i'm certain and confident that it will become very useful for high-end uh
[01:12:09] become very useful for high-end uh low-to-mid volume uh uh products and i'm
[01:12:12] low-to-mid volume uh uh products and i'm interested to see where over the next
[01:12:14] interested to see where over the next few years that will go for high volume
[01:12:16] few years that will go for high volume products um so this is another area that
[01:12:18] products um so this is another area that we're working hard on developing at
[01:12:20] we're working hard on developing at freedom photonics um the last slide i've
[01:12:23] freedom photonics um the last slide i've got i won't i won't bother reading
[01:12:24] got i won't i won't bother reading through it's basically a summary of the
[01:12:27] through it's basically a summary of the information i already talked about but
[01:12:29] information i already talked about but but common problems across applicant
[01:12:31] but common problems across applicant application areas that we're involved in
[01:12:33] application areas that we're involved in include hybrid integration you know the
[01:12:35] include hybrid integration you know the coupling of light to and from different
[01:12:37] coupling of light to and from different waveguides and into free space
[01:12:40] waveguides and into free space combining and demuxing light sources
[01:12:43] combining and demuxing light sources within a package of course the thermal
[01:12:45] within a package of course the thermal power management problems co-packaging
[01:12:48] power management problems co-packaging compact control electronics and
[01:12:50] compact control electronics and calibration electronics
[01:12:52] calibration electronics all these different areas require that
[01:12:54] all these different areas require that and of course then the reliability of
[01:12:56] and of course then the reliability of these complex packages afterwards and
[01:12:59] these complex packages afterwards and all of that and this is the last point i
[01:13:00] all of that and this is the last point i want to make needs to be done
[01:13:02] want to make needs to be done at high volume and low cost
[01:13:05] at high volume and low cost with automated assembly tests visual
[01:13:07] with automated assembly tests visual inspection is really important pass fail
[01:13:09] inspection is really important pass fail criteria throughout and full
[01:13:11] criteria throughout and full traceability of that
[01:13:13] traceability of that so
[01:13:14] so that that's uh my two cents on
[01:13:16] that that's uh my two cents on on packaging for optoelectronics
[01:13:22] thank you so much i mean when we asked
[01:13:24] thank you so much i mean when we asked you to come we never thought that you
[01:13:26] you to come we never thought that you would really share with us all the key
[01:13:28] would really share with us all the key room for cooperation with the opera
[01:13:30] room for cooperation with the opera optica members
[01:13:31] optica members you really understood what this meeting
[01:13:33] you really understood what this meeting was about and there are many people who
[01:13:35] was about and there are many people who can help you here there are many people
[01:13:36] can help you here there are many people who can help you and i would like to see
[01:13:38] who can help you and i would like to see first of all to ask now everyone you see
[01:13:41] first of all to ask now everyone you see their common problems that they have to
[01:13:44] their common problems that they have to ask everyone who have a way to help the
[01:13:48] ask everyone who have a way to help the freedom photonics to tell us right now
[01:13:49] freedom photonics to tell us right now so we can start the discussion and i'm
[01:13:52] so we can start the discussion and i'm gonna start by going to france and going
[01:13:54] gonna start by going to france and going to france to a company called icon
[01:13:57] to france to a company called icon photonics
[01:13:58] photonics yeah
[01:14:00] yeah are you with us
[01:14:02] are you with us hi everybody
[01:14:04] hi everybody hi look so you see that they are looking
[01:14:06] hi look so you see that they are looking for a potential solution for the
[01:14:08] for a potential solution for the coupling of light to and from weight
[01:14:10] coupling of light to and from weight guides we have discussed in the past the
[01:14:11] guides we have discussed in the past the micro optics the micro optics approach
[01:14:14] micro optics the micro optics approach that you have with the mems for the
[01:14:15] that you have with the mems for the coupling of the fiber to the chip
[01:14:18] coupling of the fiber to the chip and you can see how they are not
[01:14:21] and you can see how they are not stranger to micro optics because they
[01:14:22] stranger to micro optics because they already used to photon polymerization so
[01:14:25] already used to photon polymerization so could you tell us what icon photon is
[01:14:27] could you tell us what icon photon is does and how you could help
[01:14:29] does and how you could help particularly freedom photonics today
[01:14:32] particularly freedom photonics today yes thanks thanks a lot for
[01:14:34] yes thanks thanks a lot for making the connection about that topic
[01:14:37] making the connection about that topic um at icon photonics we're working on
[01:14:41] um at icon photonics we're working on optical waveguides that we manufacture
[01:14:43] optical waveguides that we manufacture on top of the uh of the optoelectronic
[01:14:46] on top of the uh of the optoelectronic chip for diets the lasers the excess for
[01:14:49] chip for diets the lasers the excess for top elimination but also we can connect
[01:14:52] top elimination but also we can connect with peaks
[01:14:53] with peaks on the edge
[01:14:54] on the edge and
[01:14:56] and it's as opposed to the two-photon
[01:14:58] it's as opposed to the two-photon polarization it's a in a one-step
[01:15:00] polarization it's a in a one-step lithography that we fabricate this 3d
[01:15:03] lithography that we fabricate this 3d structure
[01:15:04] structure which is a tapered waveguide and the
[01:15:07] which is a tapered waveguide and the advantage of that process is really to
[01:15:09] advantage of that process is really to make it at the waffle level full wafer
[01:15:12] make it at the waffle level full wafer level
[01:15:13] level with low cost
[01:15:15] with low cost and
[01:15:16] and at the same time with the tapered
[01:15:18] at the same time with the tapered waveguide you can ensure very low loss
[01:15:21] waveguide you can ensure very low loss obviously you don't have the same
[01:15:23] obviously you don't have the same flexibility as a two photon
[01:15:25] flexibility as a two photon polarimization that is really suitable
[01:15:27] polarimization that is really suitable for prototyping for very
[01:15:31] for prototyping for very fancy structure
[01:15:33] fancy structure but when you have a simple and straight
[01:15:36] but when you have a simple and straight objective that could lead to very low
[01:15:39] objective that could lead to very low loss because we can have a less than 0.5
[01:15:41] loss because we can have a less than 0.5 gb coupling losses in some cases
[01:15:44] gb coupling losses in some cases with multimode fiber or single mode or
[01:15:47] with multimode fiber or single mode or multicore fibers
[01:15:52] i think i don't know if i'm going to ask
[01:15:54] i think i don't know if i'm going to ask too much look do you really have one one
[01:15:56] too much look do you really have one one slide that you could share with us to
[01:15:58] slide that you could share with us to make it more visible
[01:16:00] make it more visible otherwise i'm going to give you a few
[01:16:01] otherwise i'm going to give you a few minutes so you to share what it's like
[01:16:02] minutes so you to share what it's like and in the meantime in the meantime
[01:16:04] and in the meantime in the meantime gordon i'm going to go to a company
[01:16:06] gordon i'm going to go to a company which i believe already you know and i'm
[01:16:08] which i believe already you know and i'm going to go to that airport from
[01:16:09] going to go to that airport from aerotech
[01:16:11] aerotech what's on your mind
[01:16:13] what's on your mind yes i just wanted to quickly introduce
[01:16:15] yes i just wanted to quickly introduce myself to the group and just listening
[01:16:16] myself to the group and just listening to some of the challenges
[01:16:18] to some of the challenges that are mentioned in these
[01:16:19] that are mentioned in these presentations it just really hits a lot
[01:16:22] presentations it just really hits a lot of spots with what we work with
[01:16:23] of spots with what we work with customers
[01:16:25] customers uh each day and so
[01:16:27] uh each day and so just a couple here uh one that i saw
[01:16:29] just a couple here uh one that i saw last week was a customer that was
[01:16:31] last week was a customer that was working with
[01:16:32] working with uh landing fibers into a v groove and
[01:16:34] uh landing fibers into a v groove and making sure that as those individual
[01:16:36] making sure that as those individual fibers went into that v groove
[01:16:39] fibers went into that v groove that there was not any skewing of that
[01:16:41] that there was not any skewing of that fiber the fact that it stayed straight
[01:16:44] fiber the fact that it stayed straight and that you were getting optimal signal
[01:16:46] and that you were getting optimal signal into each one of those channels
[01:16:48] into each one of those channels all at the same time it is a challenging
[01:16:51] all at the same time it is a challenging application but it is something that we
[01:16:52] application but it is something that we work with customers on
[01:16:54] work with customers on and then just lastly in terms of the
[01:16:56] and then just lastly in terms of the challenges that freedom photonics was
[01:16:58] challenges that freedom photonics was bringing up
[01:16:59] bringing up in terms of getting that ultra high
[01:17:01] in terms of getting that ultra high resolution of alignment and they were
[01:17:03] resolution of alignment and they were working with uh you know piezo
[01:17:06] working with uh you know piezo stages and how sometimes it's you know
[01:17:08] stages and how sometimes it's you know in his words a boutique solution
[01:17:10] in his words a boutique solution um both from a performance standpoint
[01:17:13] um both from a performance standpoint and a reliability standpoint we've been
[01:17:15] and a reliability standpoint we've been able to get those
[01:17:16] able to get those levels of alignment within using a
[01:17:19] levels of alignment within using a linear servo stage technology in all six
[01:17:22] linear servo stage technology in all six degrees of freedom we have a couple
[01:17:23] degrees of freedom we have a couple different configurations so we commonly
[01:17:25] different configurations so we commonly work with customers
[01:17:26] work with customers in these packaging challenges and really
[01:17:29] in these packaging challenges and really from the start when you're working with
[01:17:31] from the start when you're working with our solutions it's not just a lab
[01:17:33] our solutions it's not just a lab solution
[01:17:34] solution but it's also going to be a fab ready
[01:17:36] but it's also going to be a fab ready and a production writing solution too so
[01:17:38] and a production writing solution too so if you have those challenges
[01:17:40] if you have those challenges myself and our engineering team i have
[01:17:42] myself and our engineering team i have we have some dedicated engineers for
[01:17:43] we have some dedicated engineers for photonics applications
[01:17:45] photonics applications and and we can help with that so
[01:17:47] and and we can help with that so give me a shout
[01:17:49] give me a shout thank you thank you very much for that
[01:17:51] thank you thank you very much for that er when we talk about the seven
[01:17:53] er when we talk about the seven nanometer with six degrees of freedom
[01:17:57] nanometer with six degrees of freedom alignment
[01:17:58] alignment i would like to ask
[01:18:00] i would like to ask er i would like to ask back a luminar
[01:18:03] er i would like to ask back a luminar freedom photonics gordon morrison do you
[01:18:05] freedom photonics gordon morrison do you have some challenges there on the
[01:18:06] have some challenges there on the accuracy that you need
[01:18:09] accuracy that you need okay
[01:18:14] yeah uh by and large with with the
[01:18:17] yeah uh by and large with with the alignment of our
[01:18:18] alignment of our diode laser arrays the
[01:18:21] diode laser arrays the accuracy of alignment has to be better
[01:18:23] accuracy of alignment has to be better than half a micron
[01:18:25] than half a micron in order to have
[01:18:27] in order to have less than 3 db coupling loss
[01:18:30] less than 3 db coupling loss and that's something we can achieve
[01:18:34] and that's something we can achieve through brute force of but coupling one
[01:18:36] through brute force of but coupling one component to another and we've done that
[01:18:37] component to another and we've done that with the rays as well
[01:18:39] with the rays as well what you run into is a problem if you're
[01:18:42] what you run into is a problem if you're talking about larger arrays of diodes
[01:18:43] talking about larger arrays of diodes although they're lithographically
[01:18:45] although they're lithographically defined in their position and so
[01:18:47] defined in their position and so consistent that in that sense
[01:18:49] consistent that in that sense controlling the thickness
[01:18:51] controlling the thickness of the bar
[01:18:52] of the bar uh across that uh length of several
[01:18:55] uh across that uh length of several millimeters can become tricky when
[01:18:57] millimeters can become tricky when you're talking about less than half a
[01:18:59] you're talking about less than half a micron and then depending on what
[01:19:01] micron and then depending on what whatever your bonding approach is for
[01:19:03] whatever your bonding approach is for that bar there's a certain amount of
[01:19:05] that bar there's a certain amount of natural stress and strain that's going
[01:19:06] natural stress and strain that's going on and so uh you tend to end up with a
[01:19:09] on and so uh you tend to end up with a little bit of bowing or curving from one
[01:19:11] little bit of bowing or curving from one end to the other and so if you've got
[01:19:12] end to the other and so if you've got two chips that don't have perfectly
[01:19:14] two chips that don't have perfectly matching curves it's not an easy problem
[01:19:16] matching curves it's not an easy problem to solve
[01:19:17] to solve um this is this is something that that i
[01:19:19] um this is this is something that that i think a lot of different people uh are
[01:19:22] think a lot of different people uh are are working on you can solve it
[01:19:23] are working on you can solve it mechanically by by careful
[01:19:26] mechanically by by careful design of the of the bar and design of
[01:19:28] design of the of the bar and design of the coupling optics or you can
[01:19:30] the coupling optics or you can try other brute force solutions such as
[01:19:32] try other brute force solutions such as photonic wire bonding or i guess also
[01:19:34] photonic wire bonding or i guess also this uh one step lithography that icon
[01:19:36] this uh one step lithography that icon photonics was also talking about
[01:19:39] photonics was also talking about i have many people in the room who would
[01:19:40] i have many people in the room who would like to help you who like to assist you
[01:19:42] like to help you who like to assist you and i will determine everyone at this
[01:19:44] and i will determine everyone at this step that after the meeting i will make
[01:19:45] step that after the meeting i will make the one by one introductions by email so
[01:19:48] the one by one introductions by email so you all can contribute and i believe
[01:19:50] you all can contribute and i believe that john luke has now the slide that
[01:19:52] that john luke has now the slide that you can share with that they can share
[01:19:53] you can share with that they can share with us can look
[01:19:57] can look like indeed i'm ready for that
[01:19:59] can look like indeed i'm ready for that yes
[01:20:01] yes so
[01:20:01] so this is a very important slide because i
[01:20:03] this is a very important slide because i believe that many people here are going
[01:20:04] believe that many people here are going to be benefited by working with you on
[01:20:07] to be benefited by working with you on this
[01:20:08] this okay
[01:20:09] okay so
[01:20:10] so this is actually a slide which is
[01:20:12] this is actually a slide which is showing you a main picture of the
[01:20:15] showing you a main picture of the optical waveguide that we can do
[01:20:16] optical waveguide that we can do vertically this is here the the
[01:20:19] vertically this is here the the waveguide
[01:20:21] waveguide with a tapered shape and some pillows to
[01:20:23] with a tapered shape and some pillows to maintain it but actually this is just an
[01:20:25] maintain it but actually this is just an illustration
[01:20:26] illustration and you can
[01:20:29] and you can do
[01:20:30] do solution in more detail here
[01:20:32] solution in more detail here uh where you have
[01:20:35] uh where you have actually
[01:20:36] actually the chip
[01:20:37] the chip that is embedded on a silicon wafer
[01:20:39] that is embedded on a silicon wafer approach or this is your wafer if we are
[01:20:42] approach or this is your wafer if we are working at wafer level on top of which
[01:20:44] working at wafer level on top of which you have the taper waveguide there and
[01:20:46] you have the taper waveguide there and you can see the same view there or an
[01:20:49] you can see the same view there or an optical
[01:20:50] optical microscopic view it is made in polymer
[01:20:53] microscopic view it is made in polymer so it's transparent of a wide range of
[01:20:56] so it's transparent of a wide range of wavelengths so it's applica
[01:20:59] wavelengths so it's applica can find applications from
[01:21:01] can find applications from 600 nanometers to
[01:21:03] 600 nanometers to more than uh
[01:21:04] more than uh to more than two micron
[01:21:06] to more than two micron wavelengths actually we can go up to
[01:21:08] wavelengths actually we can go up to five micron and then we can have
[01:21:10] five micron and then we can have mechanical fixture for example for your
[01:21:12] mechanical fixture for example for your fiber attachment that is an option
[01:21:15] fiber attachment that is an option so
[01:21:17] so you can also have the
[01:21:19] you can also have the surface input output array here your
[01:21:22] surface input output array here your chip and the structure we've seen here
[01:21:25] chip and the structure we've seen here in an array of 12 or
[01:21:27] in an array of 12 or n times fibers but you can find it as
[01:21:30] n times fibers but you can find it as well this structure is same put on the
[01:21:32] well this structure is same put on the edge of your chip and mechanically
[01:21:34] edge of your chip and mechanically attached to it so you can have the
[01:21:38] attached to it so you can have the the purpose of a mod
[01:21:40] the purpose of a mod mod size converter from your
[01:21:44] mod size converter from your peak
[01:21:45] peak to the fiber array
[01:21:48] to the fiber array so
[01:21:49] so finally in terms of
[01:21:51] finally in terms of of performances you can find some some
[01:21:53] of performances you can find some some examples here if you want capacity
[01:21:55] examples here if you want capacity multi-mode fibers or to signal not fiber
[01:21:58] multi-mode fibers or to signal not fiber with the losses you can you can
[01:22:00] with the losses you can you can [Music]
[01:22:01] [Music] you can target and uh well today we are
[01:22:04] you can target and uh well today we are working for optical transceiver but
[01:22:06] working for optical transceiver but there is also the quantum computing
[01:22:08] there is also the quantum computing because
[01:22:09] because with a mechanical fixture present taper
[01:22:11] with a mechanical fixture present taper you can um in the taper shape uh
[01:22:15] you can um in the taper shape uh compensate for the medialis materials
[01:22:17] compensate for the medialis materials for for the fiber to the chip
[01:22:19] for for the fiber to the chip and then you can
[01:22:21] and then you can make it at room temperature the
[01:22:23] make it at room temperature the alignment
[01:22:24] alignment in a blind mode because you don't need
[01:22:26] in a blind mode because you don't need an active environment you can be only
[01:22:28] an active environment you can be only passive and then you cool it down to
[01:22:30] passive and then you cool it down to cryogenic temperature and that is
[01:22:33] cryogenic temperature and that is giving a high advantage for the quantum
[01:22:36] giving a high advantage for the quantum computing and you can make it not for
[01:22:38] computing and you can make it not for one device but really collectively for a
[01:22:41] one device but really collectively for a raise of
[01:22:42] raise of of uh tens of devices if you if you'd
[01:22:45] of uh tens of devices if you if you'd like
[01:22:47] like so that was very short just to to put
[01:22:49] so that was very short just to to put the focus on it but yes now we got
[01:22:52] the focus on it but yes now we got everyone including gordon to understand
[01:22:54] everyone including gordon to understand and gordon understands now why i think
[01:22:56] and gordon understands now why i think there is a clear room for cooperation
[01:22:59] there is a clear room for cooperation here gordon i would like to only ask you
[01:23:03] here gordon i would like to only ask you that you have a follow-up discussion
[01:23:04] that you have a follow-up discussion with only young luke and with all the
[01:23:07] with only young luke and with all the others that are gonna introduce to you
[01:23:08] others that are gonna introduce to you because with all these challenges many
[01:23:10] because with all these challenges many people are already asking me to see they
[01:23:12] people are already asking me to see they can have a discussion offline with you
[01:23:14] can have a discussion offline with you it is it means the world that you work
[01:23:15] it is it means the world that you work with us on this first online industry
[01:23:17] with us on this first online industry meeting thank you very much and we're
[01:23:20] meeting thank you very much and we're going to continue now with the program
[01:23:22] going to continue now with the program we go from lidar to the world of
[01:23:25] we go from lidar to the world of plugable transceivers and we have one of
[01:23:27] plugable transceivers and we have one of the market leaders in the room we have
[01:23:30] the market leaders in the room we have the company hg
[01:23:33] the company hg genuine h genuine the u.s society of hg
[01:23:36] genuine h genuine the u.s society of hg tech
[01:23:37] tech represented today by jim theodoras the
[01:23:40] represented today by jim theodoras the vp from r d from h.g genuine team thank
[01:23:43] vp from r d from h.g genuine team thank you very much being with us the floor
[01:23:45] you very much being with us the floor and the attention of everyone goes to
[01:23:48] and the attention of everyone goes to you
[01:23:50] you all right thank you jose
[01:23:56] okay i assume you can hear and see
[01:23:58] okay i assume you can hear and see everything okay
[01:24:00] everything okay so all be as jose mentioned i'll be
[01:24:02] so all be as jose mentioned i'll be speaking from the perspective of an
[01:24:04] speaking from the perspective of an optical transceiver manufacturer
[01:24:06] optical transceiver manufacturer um we were humming along pre-covet
[01:24:09] um we were humming along pre-covet around 23 to 24 million transceivers
[01:24:11] around 23 to 24 million transceivers shipped a year
[01:24:13] shipped a year and this post coveted boom really
[01:24:16] and this post coveted boom really caught us and caught me by surprise
[01:24:18] caught us and caught me by surprise we're on track to ship 100 million
[01:24:20] we're on track to ship 100 million transceivers this year
[01:24:22] transceivers this year and i just bring that up to mention the
[01:24:25] and i just bring that up to mention the scale of the problem we're dealing with
[01:24:27] scale of the problem we're dealing with when it comes to photonic
[01:24:29] when it comes to photonic um testing and assembly it's
[01:24:32] um testing and assembly it's it's you know i'm i continue to learn
[01:24:35] it's you know i'm i continue to learn it's been a learning experience for me
[01:24:37] it's been a learning experience for me and uh events like this are critical for
[01:24:39] and uh events like this are critical for bringing together the people to solve
[01:24:41] bringing together the people to solve these kind of
[01:24:42] these kind of problems so i won't go through this in
[01:24:44] problems so i won't go through this in detail in the interest of time but
[01:24:46] detail in the interest of time but suffice to say
[01:24:47] suffice to say no two transceivers are alike you would
[01:24:49] no two transceivers are alike you would think the beings they look the same on
[01:24:51] think the beings they look the same on the outside maybe there's some
[01:24:52] the outside maybe there's some commonality but unfortunately today we
[01:24:55] commonality but unfortunately today we still all the um packaging assembly and
[01:24:59] still all the um packaging assembly and testing solutions are pretty custom to
[01:25:02] testing solutions are pretty custom to each variant now we do try to group
[01:25:04] each variant now we do try to group together like things like dr's
[01:25:07] together like things like dr's fr4s
[01:25:08] fr4s pixel-based but in the end it almost
[01:25:11] pixel-based but in the end it almost becomes a unique solution a unique line
[01:25:14] becomes a unique solution a unique line for each of our transceivers
[01:25:17] for each of our transceivers um no no trans to transceivers are alike
[01:25:20] um no no trans to transceivers are alike so
[01:25:21] so i've shown this slide in the past and i
[01:25:23] i've shown this slide in the past and i immediately was inundated with questions
[01:25:25] immediately was inundated with questions people saying it was too complicated i
[01:25:26] people saying it was too complicated i don't get so let me i'll spend a minute
[01:25:28] don't get so let me i'll spend a minute on this so what i've plotted here is for
[01:25:31] on this so what i've plotted here is for one of our products um
[01:25:33] one of our products um the in red is our production volume from
[01:25:36] the in red is our production volume from zero to where we're at today and then in
[01:25:39] zero to where we're at today and then in blue is the cost coming down
[01:25:42] blue is the cost coming down and as you make more the price comes
[01:25:44] and as you make more the price comes down that's that's everyone understands
[01:25:46] down that's that's everyone understands that basic um economic theory but
[01:25:49] that basic um economic theory but there's different phases as we uh to the
[01:25:52] there's different phases as we uh to the production cycle so i call it the
[01:25:53] production cycle so i call it the economic life cycles of a transceiver
[01:25:56] economic life cycles of a transceiver and so at first there's the make it work
[01:25:59] and so at first there's the make it work then there's the make a producible and
[01:26:01] then there's the make a producible and then there's what i call the make it
[01:26:02] then there's what i call the make it profitable and
[01:26:04] profitable and um
[01:26:06] um as an industry i i think we to get out
[01:26:08] as an industry i i think we to get out my soapbox here i think we focused too
[01:26:10] my soapbox here i think we focused too much on the green and maybe the start of
[01:26:12] much on the green and maybe the start of the yellow there we were so focused on
[01:26:14] the yellow there we were so focused on how do i make this work how to get to
[01:26:16] how do i make this work how to get to the next level how do i get faster
[01:26:18] the next level how do i get faster that we forget most of the life cycle of
[01:26:20] that we forget most of the life cycle of the transceiver so the majority of the
[01:26:22] the transceiver so the majority of the transceiver revenue profits occur very
[01:26:24] transceiver revenue profits occur very late in this cycle it's this blue region
[01:26:26] late in this cycle it's this blue region here that's when you're really
[01:26:29] here that's when you're really benefiting and profiting from all that
[01:26:31] benefiting and profiting from all that hard work that came before it and to put
[01:26:33] hard work that came before it and to put that into perspective on a tr on a
[01:26:36] that into perspective on a tr on a typical transceiver will have three to
[01:26:38] typical transceiver will have three to five design engineers 30 to 50 npies new
[01:26:41] five design engineers 30 to 50 npies new product introduction engineers and then
[01:26:44] product introduction engineers and then three to 500 factory engineers those
[01:26:46] three to 500 factory engineers those aren't the workers those are the
[01:26:47] aren't the workers those are the engineers doing the machines the jigs
[01:26:49] engineers doing the machines the jigs the test fixtures
[01:26:51] the test fixtures um
[01:26:52] um setting up the line designing it and
[01:26:54] setting up the line designing it and that's almost unique per transceiver so
[01:26:57] that's almost unique per transceiver so i've worked at companies before that
[01:26:58] i've worked at companies before that were opposite of this three to 500
[01:27:01] were opposite of this three to 500 design engineers 30 to 50 mpies and
[01:27:04] design engineers 30 to 50 mpies and maybe three to five factory engineers
[01:27:06] maybe three to five factory engineers helping set up a line so
[01:27:07] helping set up a line so um i think as an industry something we
[01:27:09] um i think as an industry something we need to do better at is focusing on the
[01:27:11] need to do better at is focusing on the blue where we actually are making the
[01:27:13] blue where we actually are making the money and again that's why i love coming
[01:27:16] money and again that's why i love coming to these events because it really gets
[01:27:18] to these events because it really gets the whole industry working on that blue
[01:27:20] the whole industry working on that blue section there um and and and helps us be
[01:27:22] section there um and and and helps us be more profitable so
[01:27:24] more profitable so uh so i was asked to talk about what do
[01:27:27] uh so i was asked to talk about what do i need from the people on this call and
[01:27:29] i need from the people on this call and this isn't techy and it probably isn't
[01:27:31] this isn't techy and it probably isn't exotic but it's it's it's the truth
[01:27:34] exotic but it's it's it's the truth here's what i need i need more
[01:27:35] here's what i need i need more cross-functional cooperation which is
[01:27:37] cross-functional cooperation which is why i make time for events like this and
[01:27:39] why i make time for events like this and today
[01:27:40] today uh we've got um friends we've got this
[01:27:43] uh we've got um friends we've got this linear mentality where you know the
[01:27:45] linear mentality where you know the wafer is focused on the best diet they
[01:27:47] wafer is focused on the best diet they can make and delivering known good dye
[01:27:49] can make and delivering known good dye and then they throw it over the fence to
[01:27:51] and then they throw it over the fence to the uh the sub assembly house or maybe
[01:27:53] the uh the sub assembly house or maybe the pick house and they're focused on
[01:27:56] the pick house and they're focused on on yields and and and how good can i
[01:27:59] on yields and and and how good can i make that pick and meeting their spec
[01:28:00] make that pick and meeting their spec they threw it over the wall to me the
[01:28:02] they threw it over the wall to me the transceiver vendor now i'm focused on
[01:28:04] transceiver vendor now i'm focused on meeting msa and ieee specs but
[01:28:07] meeting msa and ieee specs but really we need each stage of this
[01:28:09] really we need each stage of this manufacturing process should contain
[01:28:11] manufacturing process should contain assembly and test features needed by the
[01:28:13] assembly and test features needed by the other stages and they can only do that
[01:28:15] other stages and they can only do that at events like this where we're all
[01:28:17] at events like this where we're all talking and collaborating to where they
[01:28:19] talking and collaborating to where they know what to put on there so i'll just
[01:28:20] know what to put on there so i'll just give one example of each of these um
[01:28:23] give one example of each of these um just
[01:28:24] just just to help clarify what i'm referring
[01:28:26] just to help clarify what i'm referring to for instance the wafer
[01:28:29] to for instance the wafer uh
[01:28:31] uh i there was a uh
[01:28:33] i there was a uh i had to say so there's a wafer um
[01:28:36] i had to say so there's a wafer um features that were being diced off that
[01:28:38] features that were being diced off that i found were being diced off i actually
[01:28:40] i found were being diced off i actually could use those at the next two stages
[01:28:42] could use those at the next two stages and so i i asked the wafer manufacturer
[01:28:45] and so i i asked the wafer manufacturer stop throwing away your test coupons and
[01:28:47] stop throwing away your test coupons and i need the box silicon as a mechanical
[01:28:50] i need the box silicon as a mechanical handle um they weren't aware that that
[01:28:52] handle um they weren't aware that that was a need so simply communicating that
[01:28:54] was a need so simply communicating that be able to make that change um on the
[01:28:57] be able to make that change um on the pick uh
[01:28:58] pick uh we
[01:28:59] we so the pick designers are narrowly
[01:29:01] so the pick designers are narrowly focused on how small can i make it how
[01:29:04] focused on how small can i make it how many wafers can i how many die can i get
[01:29:06] many wafers can i how many die can i get on a wafer and
[01:29:08] on a wafer and yeah maybe if you're doing a tsmc five
[01:29:11] yeah maybe if you're doing a tsmc five nanometer part that makes sense but
[01:29:14] nanometer part that makes sense but uh in some of these pick wafer processes
[01:29:17] uh in some of these pick wafer processes maybe there's value added by not being
[01:29:19] maybe there's value added by not being focused on that so i would argue rather
[01:29:21] focused on that so i would argue rather um
[01:29:23] um uh controversially that it's okay to put
[01:29:26] uh controversially that it's okay to put throwaway assembling test features on
[01:29:28] throwaway assembling test features on that pick for instance if i can add 80
[01:29:30] that pick for instance if i can add 80 cents in electronics that are only used
[01:29:33] cents in electronics that are only used in manufacturing tests and then once
[01:29:35] in manufacturing tests and then once it's in the field they're never even
[01:29:36] it's in the field they're never even powered up and it saves me eight dollars
[01:29:39] powered up and it saves me eight dollars in that blu-ray in manufacturing it's
[01:29:41] in that blu-ray in manufacturing it's worth it and
[01:29:43] worth it and i was glad that peter touched on the
[01:29:45] i was glad that peter touched on the built-in
[01:29:46] built-in self-test features that they put on
[01:29:48] self-test features that they put on cisco puts on their picks but i'd go one
[01:29:50] cisco puts on their picks but i'd go one step further and say if 80 of that die
[01:29:53] step further and say if 80 of that die never got turned on in the field but
[01:29:54] never got turned on in the field but help me in that blue phase it'd be okay
[01:29:57] help me in that blue phase it'd be okay so as long as you're not adding power
[01:29:58] so as long as you're not adding power because i can power it down as long as
[01:30:00] because i can power it down as long as you're not affecting your liability i
[01:30:01] you're not affecting your liability i don't care that picks twice the size
[01:30:03] don't care that picks twice the size again if you save me eight dollars in
[01:30:06] again if you save me eight dollars in assembly but cost me 80 cents in parts
[01:30:08] assembly but cost me 80 cents in parts i'm willing to throw that away in the
[01:30:10] i'm willing to throw that away in the field i don't care and then maybe one
[01:30:12] field i don't care and then maybe one example in the transceiver side is i
[01:30:15] example in the transceiver side is i already have microcontrollers and i
[01:30:17] already have microcontrollers and i already have eight of these d a's um so
[01:30:20] already have eight of these d a's um so i if i can add control loops that make
[01:30:22] i if i can add control loops that make the life of the pic designer easier or
[01:30:24] the life of the pic designer easier or the wafer designer easier
[01:30:26] the wafer designer easier then that's worth it from if that that
[01:30:28] then that's worth it from if that that helps all of us and one example um i had
[01:30:31] helps all of us and one example um i had problems getting a certain part and when
[01:30:33] problems getting a certain part and when i tracked it down i said oh wafer yields
[01:30:35] i tracked it down i said oh wafer yields are really bad and so then i worked with
[01:30:37] are really bad and so then i worked with the wafer company i said why are the
[01:30:38] the wafer company i said why are the yields bad oh we're missing this one
[01:30:40] yields bad oh we're missing this one corner so there was one of the nine
[01:30:42] corner so there was one of the nine corners they tested one corner was being
[01:30:44] corners they tested one corner was being missed
[01:30:45] missed um at the hot corner and yet i was using
[01:30:47] um at the hot corner and yet i was using the part with the text so that didn't
[01:30:49] the part with the text so that didn't apply to me and so i had another part
[01:30:51] apply to me and so i had another part number sign and said give me everything
[01:30:53] number sign and said give me everything you've been feeling i never had an issue
[01:30:55] you've been feeling i never had an issue getting parts and again so this
[01:30:57] getting parts and again so this communication
[01:30:58] communication uh really is a big thing and so again
[01:31:01] uh really is a big thing and so again each stage in this process should have
[01:31:03] each stage in this process should have assembling test features that the other
[01:31:05] assembling test features that the other guy needs not just what you need to
[01:31:07] guy needs not just what you need to deliver your product and i'm going to
[01:31:09] deliver your product and i'm going to move to this next one so i added this
[01:31:12] move to this next one so i added this this weekend because i was asked to talk
[01:31:14] this weekend because i was asked to talk more about photonics packaging assembly
[01:31:16] more about photonics packaging assembly and testing and what do we need and so
[01:31:19] and testing and what do we need and so um i am doing cool stuff like peter at
[01:31:22] um i am doing cool stuff like peter at cisco but they're through
[01:31:24] cisco but they're through several industry partners i can't show
[01:31:26] several industry partners i can't show their work howard this is our internal
[01:31:28] their work howard this is our internal siliconic teams work so i can share this
[01:31:31] siliconic teams work so i can share this without worrying about ndas and so this
[01:31:32] without worrying about ndas and so this is our peter showed his 400 gig gear for
[01:31:35] is our peter showed his 400 gig gear for that was uh through a 3d chiplet
[01:31:38] that was uh through a 3d chiplet technology this is our 2d silicon
[01:31:40] technology this is our 2d silicon photonic thing and so on this for
[01:31:43] photonic thing and so on this for instance we have two alignments we have
[01:31:46] instance we have two alignments we have the a free space alignment from the
[01:31:48] the a free space alignment from the laser to the pick through a mode coupler
[01:31:51] laser to the pick through a mode coupler but then we also have an fau alignment
[01:31:53] but then we also have an fau alignment that's done with the robot automatically
[01:31:55] that's done with the robot automatically so there's two fau blocks that get
[01:31:57] so there's two fau blocks that get aligned both receiver and the
[01:31:58] aligned both receiver and the transmitter we also have sonogram
[01:32:01] transmitter we also have sonogram inspection x-ray inspection and
[01:32:04] inspection x-ray inspection and i've blocked out everything here but
[01:32:05] i've blocked out everything here but those are clear elements because we have
[01:32:07] those are clear elements because we have optical inspections as well so that's an
[01:32:09] optical inspections as well so that's an example of the type of stuff we do i was
[01:32:11] example of the type of stuff we do i was also asked to talk about packaging so
[01:32:13] also asked to talk about packaging so what we saw pic wise and transceivers
[01:32:16] what we saw pic wise and transceivers 100 200 gig tended to be wire bonded out
[01:32:19] 100 200 gig tended to be wire bonded out 400 gig 800 gig tends to be flip chip
[01:32:22] 400 gig 800 gig tends to be flip chip which peter referred to and then as we
[01:32:24] which peter referred to and then as we move to this oso pxd that's going to be
[01:32:26] move to this oso pxd that's going to be 1.6 and 3.2 terabits things will be
[01:32:29] 1.6 and 3.2 terabits things will be moving to 3d chiplets and all that means
[01:32:31] moving to 3d chiplets and all that means is this flip chip but you're stacking in
[01:32:33] is this flip chip but you're stacking in the z dimension vertically to puts on
[01:32:35] the z dimension vertically to puts on top of chipboards on top of chiplets and
[01:32:37] top of chipboards on top of chiplets and so it's kind of flip chip extended to
[01:32:40] so it's kind of flip chip extended to the 3d realm and that will be needed so
[01:32:42] the 3d realm and that will be needed so and then finally jose i was uh jan was i
[01:32:45] and then finally jose i was uh jan was i think jan will talk to us jan was asking
[01:32:47] think jan will talk to us jan was asking well what if these common ppac
[01:32:49] well what if these common ppac capabilities were available from an
[01:32:50] capabilities were available from an industrial base would you use them
[01:32:52] industrial base would you use them instead of internally so the two top
[01:32:55] instead of internally so the two top pick companies currently globally i
[01:32:58] pick companies currently globally i would argue is intel and cisco and they
[01:33:00] would argue is intel and cisco and they have the benefit of having a lot of
[01:33:02] have the benefit of having a lot of internal money uh we were lucky in that
[01:33:05] internal money uh we were lucky in that we built a 100 million dollar factory
[01:33:07] we built a 100 million dollar factory right before the boom and wheat so we
[01:33:09] right before the boom and wheat so we have a lot of this equipment but as an
[01:33:11] have a lot of this equipment but as an industry i think what's holding us back
[01:33:13] industry i think what's holding us back and i'm going to flip back to this blue
[01:33:14] and i'm going to flip back to this blue part here
[01:33:15] part here we need a as an industrial capability to
[01:33:19] we need a as an industrial capability to do this photonics packaging assembly and
[01:33:21] do this photonics packaging assembly and testing
[01:33:22] testing so those companies all those hundred
[01:33:24] so those companies all those hundred companies focused on the green and the
[01:33:26] companies focused on the green and the yellow can turn to somebody for the blue
[01:33:28] yellow can turn to somebody for the blue and still make money um and i think
[01:33:31] and still make money um and i think that's we need to get away from having
[01:33:32] that's we need to get away from having this internal vertical capability to
[01:33:34] this internal vertical capability to where as a whole everyone here can go
[01:33:37] where as a whole everyone here can go and get the stuff built and then i think
[01:33:39] and get the stuff built and then i think you'll see innovation and adoption of
[01:33:41] you'll see innovation and adoption of the photonic pics really take off from
[01:33:44] the photonic pics really take off from it's it's kind of narrowly used today
[01:33:45] it's it's kind of narrowly used today and what's holding us back is we don't
[01:33:47] and what's holding us back is we don't have this common industry-wide
[01:33:50] have this common industry-wide um
[01:33:51] um capability
[01:33:52] capability and that's what i have
[01:33:54] and that's what i have and thank you again jose for inviting me
[01:33:56] and thank you again jose for inviting me this is a great to be a part of this
[01:33:58] this is a great to be a part of this industry-wide discussion on this topic
[01:34:00] industry-wide discussion on this topic it is the first of many jim get used to
[01:34:03] it is the first of many jim get used to this and you're also going to be 104
[01:34:05] this and you're also going to be 104 speakers
[01:34:06] speakers at our event in silicon valley on co
[01:34:10] at our event in silicon valley on co package and flaggable optics in
[01:34:12] package and flaggable optics in cooperation with kobo hosted by dupont
[01:34:14] cooperation with kobo hosted by dupont center 267 of october i can't wait for
[01:34:18] center 267 of october i can't wait for that meeting with such a great agenda
[01:34:20] that meeting with such a great agenda and to find ways for us to work together
[01:34:22] and to find ways for us to work together there are many people who have questions
[01:34:25] there are many people who have questions many people who have comments but i
[01:34:27] many people who have comments but i would like to bring to the room a
[01:34:29] would like to bring to the room a foundry
[01:34:31] foundry have we have global foundries in the
[01:34:32] have we have global foundries in the room and is represented by takako sang
[01:34:37] room and is represented by takako sang takakohiro kawasan nakakosan good
[01:34:40] takakohiro kawasan nakakosan good good evening for you are you with us
[01:34:46] yes hello
[01:34:47] yes hello hello thank you very much for being with
[01:34:49] hello thank you very much for being with us again it's great to hear you i just
[01:34:51] us again it's great to hear you i just would like to for you to say hello to
[01:34:53] would like to for you to say hello to everyone but also to tell us what you do
[01:34:55] everyone but also to tell us what you do at global fundraise and how you are
[01:34:57] at global fundraise and how you are addressing the the need for wafer level
[01:35:00] addressing the the need for wafer level packaging wafer level testing especially
[01:35:02] packaging wafer level testing especially testing and what brings you to this
[01:35:04] testing and what brings you to this meeting
[01:35:05] meeting hi um yeah so i work
[01:35:09] hi um yeah so i work on the v-groove
[01:35:11] on the v-groove at global foundries
[01:35:12] at global foundries um so this is a technology that
[01:35:16] um so this is a technology that we developed
[01:35:19] we developed with ibm a few years ago
[01:35:22] with ibm a few years ago and now we
[01:35:24] and now we have v-grooves that we can
[01:35:27] have v-grooves that we can manufacture on um
[01:35:29] manufacture on um each chiplet
[01:35:32] each chiplet um so yeah we are
[01:35:35] um so yeah we are have
[01:35:36] have obviously people are very well aware of
[01:35:39] obviously people are very well aware of the the challenges
[01:35:41] the the challenges associated with
[01:35:42] associated with um
[01:35:45] um you know testing
[01:35:47] you know testing in wine
[01:35:49] in wine and
[01:35:50] and yeah so
[01:35:52] yeah so on the vigroof manufacturing tacaco is
[01:35:54] on the vigroof manufacturing tacaco is there any any any challenge or room for
[01:35:57] there any any any challenge or room for cooperation that you could maybe see is
[01:35:59] cooperation that you could maybe see is there anything that you would think oh
[01:36:01] there anything that you would think oh my god if i could find a company that
[01:36:03] my god if i could find a company that could help me with this that would be a
[01:36:04] could help me with this that would be a great use of my time
[01:36:08] um
[01:36:10] um yeah i think right now
[01:36:13] yeah i think right now i guess broadly speaking we have
[01:36:16] i guess broadly speaking we have uh
[01:36:17] uh we have
[01:36:18] we have our manufacturing uh
[01:36:21] our manufacturing uh fibers at a certain pitch but
[01:36:24] fibers at a certain pitch but there are there is interest in uh
[01:36:28] there are there is interest in uh making a tighter pitch
[01:36:30] making a tighter pitch um
[01:36:33] as well as uh
[01:36:34] as well as uh you know there's also been uh
[01:36:37] you know there's also been uh been um
[01:36:39] been um some
[01:36:40] some uh research on multi-core fibers as well
[01:36:43] uh research on multi-core fibers as well um so
[01:36:46] um so how we end up packaging if multi-core
[01:36:49] how we end up packaging if multi-core fibers uh end up being the direction we
[01:36:52] fibers uh end up being the direction we want to go in as an industry how we
[01:36:55] want to go in as an industry how we package that um
[01:36:57] package that um is an interesting
[01:36:59] is an interesting challenge but we're not we're not there
[01:37:01] challenge but we're not we're not there yet
[01:37:02] yet thank you so much yes
[01:37:04] thank you so much yes please this well if i could comment i've
[01:37:06] please this well if i could comment i've heard multiple times now of reducing the
[01:37:09] heard multiple times now of reducing the pitch of the v grooves and that is so
[01:37:11] pitch of the v grooves and that is so important so i think this the width of
[01:37:14] important so i think this the width of the transceivers hasn't gotten any wider
[01:37:16] the transceivers hasn't gotten any wider and says as but the channel counts are
[01:37:18] and says as but the channel counts are going up so we're now in a situation
[01:37:20] going up so we're now in a situation where when we move to 16 wide
[01:37:23] where when we move to 16 wide buses that quarter micron spacing adds
[01:37:25] buses that quarter micron spacing adds up to where it suddenly becomes
[01:37:28] up to where it suddenly becomes uh too wide and so yeah moving that down
[01:37:31] uh too wide and so yeah moving that down and getting the fibers closer together
[01:37:32] and getting the fibers closer together is going to be very key to moving
[01:37:34] is going to be very key to moving forward to the higher channel counts and
[01:37:36] forward to the higher channel counts and transceivers because they're not getting
[01:37:38] transceivers because they're not getting any wider so our only choice is to get
[01:37:40] any wider so our only choice is to get the fibers
[01:37:41] the fibers pitch reduced so we can fit more
[01:37:43] pitch reduced so we can fit more width-wise
[01:37:46] we have more questions and comments the
[01:37:48] we have more questions and comments the next one is coming from
[01:37:50] next one is coming from ibm ibm in bramant canada alexander
[01:37:54] ibm ibm in bramant canada alexander janta alexander good afternoon good
[01:37:56] janta alexander good afternoon good afternoon for me good morning for you
[01:38:00] afternoon for me good morning for you alexander
[01:38:03] i lost alexander this is the problem of
[01:38:05] i lost alexander this is the problem of doing the meeting live
[01:38:07] doing the meeting live alexander yanta three two one in the
[01:38:10] alexander yanta three two one in the meantime jim
[01:38:12] meantime jim and takatko i would like to introduce to
[01:38:14] and takatko i would like to introduce to some comp to a company that is very
[01:38:16] some comp to a company that is very meaningful for our global industry
[01:38:18] meaningful for our global industry network the company is heidelberg
[01:38:20] network the company is heidelberg instruments
[01:38:21] instruments phillip
[01:38:23] phillip phillip paul are you with us yes hello
[01:38:26] phillip paul are you with us yes hello yes i am
[01:38:27] yes i am you have seen today from takaco that was
[01:38:29] you have seen today from takaco that was talking about the the the gratings and
[01:38:32] talking about the the the gratings and the the v grooves and reducing the pitch
[01:38:33] the the v grooves and reducing the pitch between the big grooves and the need for
[01:38:35] between the big grooves and the need for jim on the packaging the coupling of the
[01:38:38] jim on the packaging the coupling of the light
[01:38:39] light philip can you tell us what hydraulic
[01:38:40] philip can you tell us what hydraulic instruments does especially in the field
[01:38:42] instruments does especially in the field of peak packaging how can you call the
[01:38:44] of peak packaging how can you call the companies around the table today yes so
[01:38:47] companies around the table today yes so um heidelberg instruments are
[01:38:50] um heidelberg instruments are specialists and
[01:38:52] specialists and world leader in masterless lithography
[01:38:55] world leader in masterless lithography so
[01:38:56] so we actually uh um
[01:38:59] we actually uh um basically we're going in the direction
[01:39:02] basically we're going in the direction that many packaging
[01:39:06] packaging applications will um
[01:39:09] packaging applications will um we need that is um that we want to
[01:39:11] we need that is um that we want to dynamically
[01:39:14] dynamically dynamically align
[01:39:16] dynamically align the
[01:39:18] the basically make
[01:39:20] basically make the
[01:39:21] the lithography follow
[01:39:24] lithography follow the wire bonding or maybe the the
[01:39:26] the wire bonding or maybe the the even the flip chip bonding so you might
[01:39:29] even the flip chip bonding so you might uh you might find that um
[01:39:32] uh you might find that um okay we've seen some very impressive
[01:39:34] okay we've seen some very impressive alignment of 200 nanometers
[01:39:38] alignment of 200 nanometers earlier but
[01:39:39] earlier but you always get some dye shift and um
[01:39:42] you always get some dye shift and um heidelberg instruments is addressing
[01:39:44] heidelberg instruments is addressing this uh so from from the other way up
[01:39:47] this uh so from from the other way up that you bond down
[01:39:49] that you bond down first and uh with our um lithography
[01:39:52] first and uh with our um lithography equipment
[01:39:55] basically you'll be able to
[01:39:59] then um match
[01:40:01] then um match the physical location of
[01:40:03] the physical location of of
[01:40:04] of wherever the bonds or wherever the the
[01:40:06] wherever the bonds or wherever the the structures are wounded so it's kind of
[01:40:09] structures are wounded so it's kind of taking taking the reverse approach
[01:40:12] taking taking the reverse approach [Music]
[01:40:14] [Music] thank you very much philip yes yes
[01:40:18] thank you very much philip yes yes jim there's many room for cooperation
[01:40:20] jim there's many room for cooperation here many people want to get in touch
[01:40:21] here many people want to get in touch with you i just want to give you a
[01:40:23] with you i just want to give you a teaser there is a question from the
[01:40:25] teaser there is a question from the youtube we are live streaming the event
[01:40:27] youtube we are live streaming the event in youtube many people are watching you
[01:40:29] in youtube many people are watching you right now so i hope that you all look as
[01:40:30] right now so i hope that you all look as good as you can and you'll look
[01:40:32] good as you can and you'll look fantastic and the question is
[01:40:35] fantastic and the question is coming from
[01:40:38] dr subash arya
[01:40:41] dr subash arya his question is the following which
[01:40:43] his question is the following which modulation format is being used for the
[01:40:46] modulation format is being used for the high data rate that you presented is it
[01:40:49] high data rate that you presented is it pump 4 pam 8
[01:40:51] pump 4 pam 8 pam 16 format
[01:40:54] pam 16 format uh well so the industry was stuck at
[01:40:57] uh well so the industry was stuck at energy for a long time and we made the
[01:40:59] energy for a long time and we made the leap uh around 100 gigabit ethernet is
[01:41:02] leap uh around 100 gigabit ethernet is when the leap was made to pan four
[01:41:04] when the leap was made to pan four because
[01:41:05] because the band was for outpacing the ability
[01:41:07] the band was for outpacing the ability of the lasers to go faster so the leap
[01:41:09] of the lasers to go faster so the leap to pam4 was made around 100 of it
[01:41:12] to pam4 was made around 100 of it ethernet what comes after that i can't
[01:41:14] ethernet what comes after that i can't really say because the industry
[01:41:16] really say because the industry standards bodies are still working on
[01:41:18] standards bodies are still working on does is it still pan four does pam six
[01:41:20] does is it still pan four does pam six makes sense is is there now there's
[01:41:22] makes sense is is there now there's alternatives to pam modulation are being
[01:41:24] alternatives to pam modulation are being discussed but certainly
[01:41:26] discussed but certainly everything was nrz then there was a leap
[01:41:28] everything was nrz then there was a leap to pan for what comes next uh it's
[01:41:30] to pan for what comes next uh it's exciting we'll see what the what the big
[01:41:33] exciting we'll see what the what the big brains of our industry come up with
[01:41:35] brains of our industry come up with jim i can't wait to meet you at silicon
[01:41:37] jim i can't wait to meet you at silicon valley at our event we already have
[01:41:39] valley at our event we already have confirmation from keynote speakers from
[01:41:41] confirmation from keynote speakers from from meta and from google it's gonna be
[01:41:44] from meta and from google it's gonna be quite a fantastic meeting i can't wait
[01:41:47] quite a fantastic meeting i can't wait to have dinner with you there with you
[01:41:49] to have dinner with you there with you and we continue the meeting today
[01:41:52] and we continue the meeting today we go to our last speaker it's 5 40 p.m
[01:41:55] we go to our last speaker it's 5 40 p.m in the netherlands so it is exactly time
[01:41:56] in the netherlands so it is exactly time to go to the last speaker and the last
[01:41:58] to go to the last speaker and the last speaker requires no introduction but i
[01:41:59] speaker requires no introduction but i will try myself we wanted to have the
[01:42:01] will try myself we wanted to have the semiconductor
[01:42:03] semiconductor approach the semiconductor feeling in
[01:42:06] approach the semiconductor feeling in the big packaging and for that we have
[01:42:08] the big packaging and for that we have jan vardalan who has really huge
[01:42:11] jan vardalan who has really huge experience for the semiconductor
[01:42:12] experience for the semiconductor industry and who can help us make this
[01:42:15] industry and who can help us make this very important breach which is key in
[01:42:17] very important breach which is key in the strategy of optical for photonics
[01:42:19] the strategy of optical for photonics connect with semicone
[01:42:22] connect with semicone thank you very much for being with us
[01:42:23] thank you very much for being with us the floor and the attention of everyone
[01:42:25] the floor and the attention of everyone what a beautiful rune right young the
[01:42:27] what a beautiful rune right young the flow and the attention of everyone goes
[01:42:29] flow and the attention of everyone goes to you
[01:42:30] to you thank you
[01:42:31] thank you let me talk to you a little bit
[01:42:34] let me talk to you a little bit first of all i want to thank you for
[01:42:35] first of all i want to thank you for inviting me
[01:42:37] inviting me and i want to talk to you a little bit
[01:42:39] and i want to talk to you a little bit about what we see are the challenges and
[01:42:40] about what we see are the challenges and i'm glad that you placed me kind of at
[01:42:42] i'm glad that you placed me kind of at the end of the discussion because i
[01:42:44] the end of the discussion because i think that we have really
[01:42:45] think that we have really um
[01:42:47] um seen a lot today
[01:42:48] seen a lot today and one of the first things that i want
[01:42:50] and one of the first things that i want to mention
[01:42:51] to mention is that
[01:42:53] is that in our challenges that we see and this
[01:42:55] in our challenges that we see and this is really kind of what we want to what
[01:42:57] is really kind of what we want to what we want to talk about
[01:42:58] we want to talk about is that
[01:42:59] is that the
[01:43:00] the the main challenges that we have seen
[01:43:03] the main challenges that we have seen are
[01:43:04] are it's not like they're totally
[01:43:05] it's not like they're totally technically
[01:43:07] technically insurmountable and i think that's what
[01:43:09] insurmountable and i think that's what our discussion has shown us today but
[01:43:11] our discussion has shown us today but really integrating things are the issue
[01:43:14] really integrating things are the issue and i think we need to work more on our
[01:43:17] and i think we need to work more on our ecosystem i think that's been pointed
[01:43:18] ecosystem i think that's been pointed out today
[01:43:20] out today including design capabilities we really
[01:43:22] including design capabilities we really haven't talked enough about design
[01:43:24] haven't talked enough about design first of all there's no standard pdk for
[01:43:27] first of all there's no standard pdk for the silicon fab simulations or assembly
[01:43:30] the silicon fab simulations or assembly everybody has their own process
[01:43:33] everybody has their own process our light source is really a long and
[01:43:35] our light source is really a long and expensive development cycle we're going
[01:43:37] expensive development cycle we're going to have to have more co-design of the
[01:43:39] to have to have more co-design of the asic and the silicon photonics
[01:43:41] asic and the silicon photonics especially you know and i'm really
[01:43:42] especially you know and i'm really talking about co-packaged optics
[01:43:45] talking about co-packaged optics there's really not any standardization
[01:43:47] there's really not any standardization of our module assembly
[01:43:49] of our module assembly including the
[01:43:51] including the fiber light source attach and test
[01:43:54] fiber light source attach and test and every product has this kind of
[01:43:56] and every product has this kind of custom assembly infrastructure and
[01:43:58] custom assembly infrastructure and process and if we want to lower our cost
[01:44:00] process and if we want to lower our cost we have to have
[01:44:02] we have to have less customization
[01:44:05] less customization and more standardization
[01:44:07] and more standardization and so really what we need is um
[01:44:10] and so really what we need is um we need the readily available standard
[01:44:12] we need the readily available standard components that can be integrated by our
[01:44:15] components that can be integrated by our osats and our oems
[01:44:17] osats and our oems so we very easily need to be able to
[01:44:19] so we very easily need to be able to test and debug
[01:44:21] test and debug we have to have a low cost assembly
[01:44:23] we have to have a low cost assembly process for to get ourselves to the high
[01:44:26] process for to get ourselves to the high volume including the fiber attached
[01:44:28] volume including the fiber attached and that brings me to one of the things
[01:44:30] and that brings me to one of the things that we've been talking to people about
[01:44:32] that we've been talking to people about people have told us that one of the
[01:44:34] people have told us that one of the greatest challenges in the co-packaged
[01:44:36] greatest challenges in the co-packaged optics is the fiber attach how to attach
[01:44:39] optics is the fiber attach how to attach it whether you use epoxy or solder and
[01:44:41] it whether you use epoxy or solder and there's a lot of people exploring new
[01:44:43] there's a lot of people exploring new methods
[01:44:45] methods also another area is integrating the
[01:44:47] also another area is integrating the lasers in the package
[01:44:49] lasers in the package if we integrate the laser in the package
[01:44:51] if we integrate the laser in the package on chip it's going to give us the lowest
[01:44:54] on chip it's going to give us the lowest form factor lowest power lowest cost
[01:44:57] form factor lowest power lowest cost better testability and easier to repair
[01:44:59] better testability and easier to repair but but the temperatures if we put it on
[01:45:02] but but the temperatures if we put it on the chip could degrade
[01:45:04] the chip could degrade we have to figure out how to replace the
[01:45:06] we have to figure out how to replace the lasers
[01:45:08] lasers we were also talking in co-packaged
[01:45:10] we were also talking in co-packaged optics about moving from the pluggable
[01:45:14] optics about moving from the pluggable formats which we are also familiar with
[01:45:16] formats which we are also familiar with they're in very high volume today
[01:45:19] they're in very high volume today to mounting something like a bga either
[01:45:22] to mounting something like a bga either on the board or as a socketed solution
[01:45:25] on the board or as a socketed solution and as we approach that we need to be
[01:45:27] and as we approach that we need to be able to figure out how to replace that
[01:45:30] able to figure out how to replace that to upgrade to do field repair
[01:45:33] to upgrade to do field repair and we have to have some standards to
[01:45:35] and we have to have some standards to lower the cost and the manufacturing
[01:45:36] lower the cost and the manufacturing cycle time so i think that we have a lot
[01:45:39] cycle time so i think that we have a lot of things that we're we're seeing done
[01:45:42] of things that we're we're seeing done today um but we have to continue to work
[01:45:47] today um but we have to continue to work um together and
[01:45:49] um together and a an opportunity like this is fantastic
[01:45:52] a an opportunity like this is fantastic so thank you very much
[01:45:54] so thank you very much thank you very much jan how could you
[01:45:56] thank you very much jan how could you advise us the optica industry team to
[01:46:01] advise us the optica industry team to interact closer with the semiconductor
[01:46:03] interact closer with the semiconductor companies many of the semiconductor
[01:46:05] companies many of the semiconductor companies in the us are really photonic
[01:46:08] companies in the us are really photonic companies like for example intel and
[01:46:10] companies like for example intel and today we have global foundries in the
[01:46:12] today we have global foundries in the room but we like to bring more
[01:46:14] room but we like to bring more more
[01:46:15] more semiconductor players how can we
[01:46:17] semiconductor players how can we convince them to be part of these cheap
[01:46:20] convince them to be part of these cheap acts that are going on in the u.s and
[01:46:21] acts that are going on in the u.s and europe i know that tsmc is doing a lot
[01:46:23] europe i know that tsmc is doing a lot of work in this area they've published
[01:46:25] of work in this area they've published things we need to have them
[01:46:27] things we need to have them um come into this forum as well because
[01:46:29] um come into this forum as well because a lot of people have their their
[01:46:31] a lot of people have their their component you know their their
[01:46:33] component you know their their their devices fabricated there
[01:46:35] their devices fabricated there we need to have our assembly partners
[01:46:38] we need to have our assembly partners amcor ase in this discussion
[01:46:41] amcor ase in this discussion um and we need to have more of these
[01:46:43] um and we need to have more of these forums where we talk about these things
[01:46:45] forums where we talk about these things we also need to have some more of the
[01:46:46] we also need to have some more of the design
[01:46:47] design a community in as well to talk to us
[01:46:50] a community in as well to talk to us about some of the challenges
[01:46:52] about some of the challenges so i think that by combining all these
[01:46:55] so i think that by combining all these parts of the ecosystem
[01:46:57] parts of the ecosystem including the equipment and materials
[01:46:59] including the equipment and materials people we can really move ahead in this
[01:47:02] people we can really move ahead in this area uh quite rapidly
[01:47:04] area uh quite rapidly when it comes to a design and when it
[01:47:06] when it comes to a design and when it comes to the also to the design and test
[01:47:09] comes to the also to the design and test because very often they go together
[01:47:11] because very often they go together there is a company that has been really
[01:47:12] there is a company that has been really setting up the the stone for the supply
[01:47:14] setting up the the stone for the supply chain the company it was acquired by
[01:47:16] chain the company it was acquired by hitachi uh two years ago is in valencia
[01:47:20] hitachi uh two years ago is in valencia and i'm talking of course about dlc
[01:47:22] and i'm talking of course about dlc photonics polios now jalis thank you
[01:47:24] photonics polios now jalis thank you very much being with us today there are
[01:47:26] very much being with us today there are many companies in the room who are
[01:47:27] many companies in the room who are already your customers many companies
[01:47:29] already your customers many companies also your suppliers
[01:47:31] also your suppliers could you tell us a bit what are your
[01:47:35] could you tell us a bit what are your current
[01:47:36] current situation when it comes to offering to
[01:47:38] situation when it comes to offering to your customers not only design but also
[01:47:41] your customers not only design but also testing and how do you do that
[01:47:44] testing and how do you do that police we cannot hear you and we really
[01:47:46] police we cannot hear you and we really really want to
[01:47:48] really want to oh
[01:47:51] [Music]
[01:47:52] [Music] photonics and valencia we are
[01:47:54] photonics and valencia we are actually
[01:47:55] actually doing more and more work towards some
[01:47:57] doing more and more work towards some proposition packaging as well
[01:47:59] proposition packaging as well and we start to use wafer level tools
[01:48:01] and we start to use wafer level tools and
[01:48:02] and for wafers as well and for single chips
[01:48:05] for wafers as well and for single chips as well using
[01:48:07] as well using convenient conventional methods by
[01:48:10] convenient conventional methods by using manual um
[01:48:12] using manual um characterization tools and automated as
[01:48:14] characterization tools and automated as well we closely collaborate here with a
[01:48:16] well we closely collaborate here with a company who is as well attending in here
[01:48:19] company who is as well attending in here icon tech so to develop the tooling for
[01:48:23] icon tech so to develop the tooling for all kinds of assemblies and automated
[01:48:25] all kinds of assemblies and automated testing
[01:48:26] testing and we start offering these services as
[01:48:28] and we start offering these services as well for external
[01:48:31] well for external users and
[01:48:32] users and all the people around here who's meeting
[01:48:35] all the people around here who's meeting us here today
[01:48:38] us here today you have started offering packaging
[01:48:40] you have started offering packaging services to to your customers and that
[01:48:43] services to to your customers and that is based on automation in the
[01:48:45] is based on automation in the presentation in the beginning by by
[01:48:48] presentation in the beginning by by peter he actually talked about the
[01:48:50] peter he actually talked about the testing and characterization and the
[01:48:52] testing and characterization and the need to often
[01:48:54] need to often do very deep testing for many of the
[01:48:56] do very deep testing for many of the structure even including bitter rates
[01:48:59] structure even including bitter rates what kind of testing are you supplying
[01:49:00] what kind of testing are you supplying to your customer is mostly the the iv or
[01:49:03] to your customer is mostly the the iv or do you provide in-depth testing at wi-fi
[01:49:05] do you provide in-depth testing at wi-fi level or a dye level
[01:49:07] level or a dye level indeed we provide both
[01:49:09] indeed we provide both dye level and wafer level testing
[01:49:10] dye level and wafer level testing services right now and
[01:49:12] services right now and we're getting better on tooling so now
[01:49:15] we're getting better on tooling so now we start to offer as well for
[01:49:17] we start to offer as well for transceivers some high-speed uh
[01:49:19] transceivers some high-speed uh characterization system level
[01:49:20] characterization system level characterization as well not only
[01:49:22] characterization as well not only the single component testing so we just
[01:49:26] the single component testing so we just acquired
[01:49:27] acquired tools for up to 110 g
[01:49:30] tools for up to 110 g modulation
[01:49:32] modulation characterization so for transceivers and
[01:49:34] characterization so for transceivers and high speed applications we can easily
[01:49:36] high speed applications we can easily support a lot of people around here
[01:49:39] support a lot of people around here both on wafer level and single die as
[01:49:41] both on wafer level and single die as well peter peter the doubler let me come
[01:49:43] well peter peter the doubler let me come back to you from the beginning to close
[01:49:45] back to you from the beginning to close the circle nicely in the beginning of
[01:49:47] the circle nicely in the beginning of the presentation were you talking about
[01:49:48] the presentation were you talking about testing you were talking about the need
[01:49:50] testing you were talking about the need to the bitter rate testing the bear
[01:49:52] to the bitter rate testing the bear testing for some of your structures a as
[01:49:55] testing for some of your structures a as a as a customer of foundries looks
[01:49:57] a as a customer of foundries looks tender
[01:49:58] tender was always a fabulous company in the
[01:50:00] was always a fabulous company in the origin uh what is currently in your
[01:50:02] origin uh what is currently in your opinion the need for testing of the
[01:50:04] opinion the need for testing of the chips when it comes out of the foundry
[01:50:06] chips when it comes out of the foundry how close are we from the semiconductor
[01:50:08] how close are we from the semiconductor approach and jam vardalan is used to
[01:50:10] approach and jam vardalan is used to that you don't really need to test the
[01:50:11] that you don't really need to test the chips when they arrive from the foundry
[01:50:13] chips when they arrive from the foundry or you always have to test everyone is
[01:50:15] or you always have to test everyone is stealth in the chips when they come from
[01:50:16] stealth in the chips when they come from the foundry but you look at the the test
[01:50:18] the foundry but you look at the the test flow the foundry before they can ship
[01:50:21] flow the foundry before they can ship the wafers to us and that's general for
[01:50:23] the wafers to us and that's general for electronic and photonic wafers they do
[01:50:25] electronic and photonic wafers they do what's called the wafer acceptance test
[01:50:27] what's called the wafer acceptance test they test
[01:50:28] they test some structures on the wafer and based
[01:50:31] some structures on the wafer and based on their test results they
[01:50:33] on their test results they they decide that they can ship the wafer
[01:50:35] they decide that they can ship the wafer to us or not and we have to accept the
[01:50:37] to us or not and we have to accept the wafers when they pass the waiver
[01:50:39] wafers when they pass the waiver acceptance test so these are like kind
[01:50:40] acceptance test so these are like kind of crude tests but after that
[01:50:43] of crude tests but after that you own the wafer
[01:50:45] you own the wafer and then you have to test every chip
[01:50:47] and then you have to test every chip that you have designed that is of course
[01:50:49] that you have designed that is of course your design is your responsibility so
[01:50:52] your design is your responsibility so that if there's yield fallout there then
[01:50:55] that if there's yield fallout there then you have to yeah to deal with it
[01:50:57] you have to yeah to deal with it and um that
[01:50:59] and um that that testing is usually done with both
[01:51:01] that testing is usually done with both opt for photonic wafers
[01:51:04] opt for photonic wafers optically and electrically and there are
[01:51:06] optically and electrically and there are certain test structures we have to test
[01:51:08] certain test structures we have to test if that is okay if these chips are okay
[01:51:11] if that is okay if these chips are okay and then when you uh go to the next step
[01:51:14] and then when you uh go to the next step you assemble a chipset then you can test
[01:51:16] you assemble a chipset then you can test that bit error rate test and all of that
[01:51:19] that bit error rate test and all of that that i just mentioned using the built-in
[01:51:21] that i just mentioned using the built-in self-test if you cannot do built-in
[01:51:23] self-test if you cannot do built-in self-test for instance if you have tx
[01:51:25] self-test for instance if you have tx and rx separately
[01:51:26] and rx separately then you have to
[01:51:28] then you have to yeah
[01:51:29] yeah try to test these things independently
[01:51:31] try to test these things independently you need probably need more
[01:51:33] you need probably need more test equipment which is more expensive
[01:51:35] test equipment which is more expensive and slower
[01:51:36] and slower and or you take the yield hit at the end
[01:51:38] and or you take the yield hit at the end which is not preferred um
[01:51:41] which is not preferred um when when you test at speed at the
[01:51:43] when when you test at speed at the module level but then your yields need
[01:51:45] module level but then your yields need to be very high when you throw away a
[01:51:47] to be very high when you throw away a lot of good
[01:51:48] lot of good a lot of modules
[01:51:50] a lot of modules so that's that's
[01:51:53] let's hear the point of view because you
[01:51:55] let's hear the point of view because you present the point of view as the
[01:51:56] present the point of view as the customer of the founder let's hear the
[01:51:57] customer of the founder let's hear the point of view of the foundry we have one
[01:51:59] point of view of the foundry we have one of the silicon nitride foundries that
[01:52:00] of the silicon nitride foundries that the headset at the base of manufacturing
[01:52:02] the headset at the base of manufacturing silicon nitrate in the world coming all
[01:52:04] silicon nitrate in the world coming all the way from lausanne but now setting up
[01:52:06] the way from lausanne but now setting up offices worldwide and collaborating with
[01:52:08] offices worldwide and collaborating with semiconductor foundries i'm talking of
[01:52:10] semiconductor foundries i'm talking of course about ligentech davide vasanela
[01:52:14] course about ligentech davide vasanela thank you so much gretchen miller for
[01:52:15] thank you so much gretchen miller for being with us today uh davide uh you
[01:52:18] being with us today uh davide uh you just heard the perspective of peter the
[01:52:20] just heard the perspective of peter the double about the testing that is needed
[01:52:21] double about the testing that is needed as a customer of the foundry could you
[01:52:23] as a customer of the foundry could you tell us as a foundry what testing to
[01:52:26] tell us as a foundry what testing to provide and if you could could you tell
[01:52:28] provide and if you could could you tell us a couple of challenges that the rest
[01:52:30] us a couple of challenges that the rest of the companies today could help you
[01:52:31] of the companies today could help you with
[01:52:33] with um well thank you first of all for
[01:52:36] um well thank you first of all for having me here
[01:52:40] let me start by saying that i just
[01:52:42] let me start by saying that i just joined lijntech a few months ago so i'm
[01:52:44] joined lijntech a few months ago so i'm quite new of the
[01:52:47] quite new of the of the whole uh business but
[01:52:50] of the whole uh business but i work here as a test engineer and
[01:52:52] i work here as a test engineer and certainly what we want to do is to
[01:52:54] certainly what we want to do is to establish
[01:52:57] establish a very let's say
[01:52:59] a very let's say uh pipeline for the testing from uh the
[01:53:03] uh pipeline for the testing from uh the uh from our foundries to uh what we can
[01:53:07] uh from our foundries to uh what we can actually ship to the customer so we want
[01:53:09] actually ship to the customer so we want to have to to create um
[01:53:12] to have to to create um a pipeline that we can understand we can
[01:53:14] a pipeline that we can understand we can we can know we know that it's uh
[01:53:17] we can know we know that it's uh reliable and we want to test everything
[01:53:19] reliable and we want to test everything we can of course but
[01:53:22] we can of course but it's always difficult to
[01:53:24] it's always difficult to understand which are the the
[01:53:27] understand which are the the the most important part that we have to
[01:53:29] the most important part that we have to focus on
[01:53:32] thomas hessler from legend tech it is so
[01:53:35] thomas hessler from legend tech it is so great to have you here tomorrow is the
[01:53:37] great to have you here tomorrow is the ceo of vlogging tech david gave a good
[01:53:40] ceo of vlogging tech david gave a good perspective but i would like thomas to
[01:53:42] perspective but i would like thomas to tell us
[01:53:43] tell us about what kind of cooperations we are
[01:53:45] about what kind of cooperations we are very amazed by your cooperation with
[01:53:47] very amazed by your cooperation with x-fab last year what kind of other
[01:53:49] x-fab last year what kind of other cooperations are you looking to bring
[01:53:51] cooperations are you looking to bring the silicon knight photonic platform to
[01:53:54] the silicon knight photonic platform to the semiconductor
[01:53:55] the semiconductor the mindset
[01:54:00] and and the proud optical member thanks
[01:54:03] and and the proud optical member thanks jose um
[01:54:05] jose um since a long time
[01:54:06] since a long time caller well first of all maybe to go a
[01:54:09] caller well first of all maybe to go a little bit back to the first question i
[01:54:10] little bit back to the first question i think one of the key challenges is to
[01:54:12] think one of the key challenges is to link the pcm measurement for wave
[01:54:16] link the pcm measurement for wave acceptance to to the
[01:54:18] acceptance to to the to the yield of the customer device to
[01:54:21] to the yield of the customer device to make that strong link that's one of the
[01:54:23] make that strong link that's one of the key challenge to have that loop closed
[01:54:26] key challenge to have that loop closed um what kind of collaboration i mean um
[01:54:31] um what kind of collaboration i mean um i think that is quite uh was a big step
[01:54:33] i think that is quite uh was a big step with xfab and in terms of mindset i
[01:54:36] with xfab and in terms of mindset i think it starts with the people so we
[01:54:37] think it starts with the people so we have quite
[01:54:39] have quite one of our founders he he comes from the
[01:54:41] one of our founders he he comes from the semiconductor industry
[01:54:42] semiconductor industry i think that where the majority of that
[01:54:44] i think that where the majority of that mindset is
[01:54:46] mindset is started so it's really think industrial
[01:54:48] started so it's really think industrial i have a lot of industrial experience in
[01:54:49] i have a lot of industrial experience in that area
[01:54:51] that area i think it's really to
[01:54:53] i think it's really to not to come from the academic viewpoint
[01:54:56] not to come from the academic viewpoint but having seen it in industry
[01:54:58] but having seen it in industry i don't want to talk too much about
[01:55:01] i don't want to talk too much about upcoming
[01:55:03] upcoming collaboration at the moment please
[01:55:05] collaboration at the moment please understand
[01:55:07] understand i i am very old the one from x5 is
[01:55:10] i i am very old the one from x5 is public i would like to congratulate you
[01:55:13] public i would like to congratulate you because that really has been a stepping
[01:55:15] because that really has been a stepping stone for all the efforts that the
[01:55:16] stone for all the efforts that the photonic companies are doing
[01:55:18] photonic companies are doing in europe and u.s connecting with
[01:55:20] in europe and u.s connecting with semiconductor companies so thank you
[01:55:22] semiconductor companies so thank you very much for that thomas
[01:55:24] very much for that thomas i cannot finish a meeting on photonic
[01:55:25] i cannot finish a meeting on photonic packaging without giving the floor to
[01:55:27] packaging without giving the floor to phi contact i cannot i have more it's in
[01:55:30] phi contact i cannot i have more it's in the room but it's thank you very much
[01:55:32] the room but it's thank you very much for being with us today i would like for
[01:55:34] for being with us today i would like for you you heard about
[01:55:36] you you heard about the needs the needs from companies like
[01:55:38] the needs the needs from companies like like h tech like luminar like the global
[01:55:42] like h tech like luminar like the global foundries
[01:55:44] foundries cisco could you tell us a couple of
[01:55:47] cisco could you tell us a couple of follow-ups a couple of interesting
[01:55:49] follow-ups a couple of interesting mindsets that are now in your mind after
[01:55:51] mindsets that are now in your mind after the last two hours
[01:55:53] the last two hours um yes first of all thank you jose and
[01:55:56] um yes first of all thank you jose and it's nice to see you with the same
[01:55:57] it's nice to see you with the same energy motivation in your new world so
[01:56:00] energy motivation in your new world so it's uh good that you're saying the same
[01:56:02] it's uh good that you're saying the same person
[01:56:03] person so um
[01:56:05] so um i actually need to admit that i have
[01:56:06] i actually need to admit that i have could not follow all the complete
[01:56:08] could not follow all the complete discussions so i may have missed a few
[01:56:10] discussions so i may have missed a few points here and there
[01:56:12] points here and there but what's quite interesting for us and
[01:56:14] but what's quite interesting for us and i think it's the channel photonic trend
[01:56:15] i think it's the channel photonic trend is um
[01:56:16] is um there's
[01:56:17] there's so much going on also in different
[01:56:19] so much going on also in different aspects of the photonics world but
[01:56:21] aspects of the photonics world but there's um wherever we are looking um
[01:56:23] there's um wherever we are looking um there are so many new opportunities new
[01:56:25] there are so many new opportunities new challenges but it's good to see that um
[01:56:27] challenges but it's good to see that um there's so many um yeah great people
[01:56:30] there's so many um yeah great people working on that
[01:56:31] working on that and um
[01:56:32] and um so from our side it's really fascinating
[01:56:35] so from our side it's really fascinating to see um where all of this is this is
[01:56:37] to see um where all of this is this is driving in the future
[01:56:38] driving in the future and um yeah we were happy being
[01:56:41] and um yeah we were happy being a part of that photonic
[01:56:43] a part of that photonic community
[01:56:45] community and yeah the challenges we are seeing is
[01:56:48] and yeah the challenges we are seeing is quite nice so for us this is i mean yeah
[01:56:50] quite nice so for us this is i mean yeah you know of exactly what we're doing
[01:56:52] you know of exactly what we're doing it's um photonics assembly and testing
[01:56:54] it's um photonics assembly and testing and fast some of the trends is really
[01:56:56] and fast some of the trends is really the um
[01:56:58] the um yeah
[01:56:59] yeah higher degree of automation which is for
[01:57:02] higher degree of automation which is for sure um theirs is in some a lot of years
[01:57:04] sure um theirs is in some a lot of years but we are seeing now um with the line
[01:57:07] but we are seeing now um with the line concepts we are pushing um
[01:57:08] concepts we are pushing um yeah
[01:57:10] yeah quite a
[01:57:11] quite a good step into industrial
[01:57:14] good step into industrial yeah
[01:57:15] yeah the highlights for me today was the
[01:57:17] the highlights for me today was the presentation from gordon morrison when
[01:57:19] presentation from gordon morrison when he presented how they are using the
[01:57:22] he presented how they are using the machine from vanguard photonics to do
[01:57:24] machine from vanguard photonics to do the the two photon polymerization and
[01:57:27] the the two photon polymerization and the photonic wire bombs and that is an
[01:57:29] the photonic wire bombs and that is an automated machine that was done in
[01:57:31] automated machine that was done in cooperation with icon tech and that was
[01:57:35] cooperation with icon tech and that was a success story of an introduction made
[01:57:37] a success story of an introduction made in the past and i would like to remind
[01:57:38] in the past and i would like to remind all of you what i want to say with this
[01:57:40] all of you what i want to say with this introductions really drive this business
[01:57:42] introductions really drive this business if you want to get in touch beautiful
[01:57:44] if you want to get in touch beautiful t-shirt mark lukovich
[01:57:47] t-shirt mark lukovich if you want to get in touch with any of
[01:57:48] if you want to get in touch with any of the participants today
[01:57:50] the participants today all you have to do is send me an email
[01:57:52] all you have to do is send me an email we're gonna share with you in advance
[01:57:54] we're gonna share with you in advance all the participants in the coming
[01:57:56] all the participants in the coming meetings for this one after the summer
[01:57:58] meetings for this one after the summer holidays so many people register on the
[01:58:00] holidays so many people register on the last day that we didn't want to share it
[01:58:02] last day that we didn't want to share it without having complete in the next one
[01:58:03] without having complete in the next one you're gonna have it in advance this one
[01:58:04] you're gonna have it in advance this one you're gonna have one day after the
[01:58:06] you're gonna have one day after the meeting 490 people registered you're
[01:58:09] meeting 490 people registered you're going to have the people the name and
[01:58:10] going to have the people the name and linkedin addresses not the emails of
[01:58:12] linkedin addresses not the emails of course if you want to get in touch with
[01:58:13] course if you want to get in touch with any of them or you have to send me an
[01:58:15] any of them or you have to send me an email and we'll make the introduction we
[01:58:16] email and we'll make the introduction we do this for the cooperation let me just
[01:58:18] do this for the cooperation let me just remind in 30 seconds we talk from cisco
[01:58:22] remind in 30 seconds we talk from cisco and share with us the needs for
[01:58:24] and share with us the needs for packaging and test for some of the key
[01:58:27] packaging and test for some of the key driven factors from the luxe data point
[01:58:30] driven factors from the luxe data point of view what's great to have peter the
[01:58:31] of view what's great to have peter the doubler we talked from darpa about the
[01:58:33] doubler we talked from darpa about the key initiatives and we focus on the move
[01:58:36] key initiatives and we focus on the move great from five stars and the chips 3d
[01:58:38] great from five stars and the chips 3d integration then afterwards we have
[01:58:41] integration then afterwards we have freedom photonics aluminar company with
[01:58:43] freedom photonics aluminar company with perhaps the highlight of the event their
[01:58:45] perhaps the highlight of the event their key
[01:58:46] key challenges the key room for cooperation
[01:58:49] challenges the key room for cooperation in the packaging of ncw lighter modules
[01:58:51] in the packaging of ncw lighter modules and we also have hg generic that were
[01:58:54] and we also have hg generic that were looking for partnership and cooperation
[01:58:57] looking for partnership and cooperation for continuing their trend on their
[01:58:59] for continuing their trend on their plugable package transceivers this has
[01:59:02] plugable package transceivers this has been a fantastic meeting
[01:59:04] been a fantastic meeting for all of you who want to have a
[01:59:06] for all of you who want to have a follow-up discussion stay young for
[01:59:08] follow-up discussion stay young for those of you
[01:59:09] those of you following the meeting in youtube we're
[01:59:11] following the meeting in youtube we're gonna finish the live stream now you're
[01:59:13] gonna finish the live stream now you're gonna miss the rest of the meeting thank
[01:59:15] gonna miss the rest of the meeting thank you very much everyone thank you for
[01:59:17] you very much everyone thank you for making the first online industry meeting
[01:59:19] making the first online industry meeting of optica a very very special one
[01:59:25] [Music]
[02:00:06] you

https://www.youtube.com/watch?v=zAjkGhkjppA

Summary

TL;DR — Optica's first online industry meeting focused on photonics packaging, assembly, and testing, bringing together key players from across the supply chain to discuss current challenges and future opportunities. The event highlighted the increasing demand for advanced packaging solutions driven by higher data rates, miniaturization, and new applications like lidar and co-packaged optics.

Key points

• The meeting emphasized the critical need for collaboration between different stages of the supply chain, from chip manufacturers to transceiver vendors, to optimize for cost, performance, and reliability.

• Discussions covered advancements in packaging technologies, including 3D chiplets, wafer-level assembly, and novel bonding techniques like direct bonding, aiming to reduce parasitics and improve integration.

• Specific application areas like lidar and co-packaged optics present unique challenges in thermal management, optical coupling, and high-volume manufacturing.

• The importance of standardization in components, assembly processes, and testing methodologies was stressed to enable wider adoption and innovation in photonics.

• Emerging initiatives from DARPA and industry collaborations aim to foster the development of advanced packaging capabilities and address the growing demands of the photonics industry.

Takeaway — To accelerate photonics innovation and market growth, the industry must prioritize standardization, cross-functional collaboration, and the development of cost-effective, high-volume manufacturing and testing solutions.