| B0101 | Azonenberg: Hi, hows your work on MEMS? | 00:02 |
|---|---|---|
| azonenberg | Going pretty well, I'm going to be doing another few experiments over the weekend with the goal of preparing specimens to cross-section image on the SEM later on | 00:03 |
| azonenberg | i want to measure some of my layer thicknesses | 00:04 |
| azonenberg | My Si wet etch process pretty much works, the only thing left to do is fix the exposure source (better lamp) and use a thinner Ta2O5 layer to reduce overetch | 00:04 |
| azonenberg | if all goes well i'll do another evaporation run in mid-august on half of a <110> wafer coated in ~50nm Ta2O5 | 00:05 |
| azonenberg | put 100-200nm of Cu on that | 00:05 |
| azonenberg | then cleave into dies and start actually building devices | 00:05 |
| B0101 | oh, thats interesting. | 00:06 |
| azonenberg | End of August to early September i should be able to do some high-aspect-ratio tests | 00:06 |
| azonenberg | then soon after that build an actual comb drive | 00:06 |
| azonenberg | all of the other processes are mostly figured out | 00:07 |
| azonenberg | I'll probably do the first version really simple, only a couple of fingers with barely visible motion | 00:07 |
| B0101 | hmm... ok | 00:29 |
| B0101 | Azonenberg: You should make some videos about your experiments | 01:11 |
| azonenberg | B0101: Maybe at some point | 01:11 |
| azonenberg | A lot of the litho has to be done under darkroom conditions so there wouldnt be much to see | 01:11 |
| azonenberg | But if i have some time i can try to set something up | 01:12 |
| B0101 | Do you have an electron microscope? | 01:13 |
| B0101 | Ok, I am just kidding, of course its too expensive | 01:14 |
| azonenberg | B0101: Not *yet* | 01:44 |
| azonenberg | But i can get access to one on campus here and there | 01:44 |
| azonenberg | And I plan to buy one in about 5-10 years when i have the budget | 01:44 |
| azonenberg | It'd cost about as much as a decent car | 01:44 |
| azonenberg | *nice car | 01:44 |
| azonenberg | $20K ish | 01:44 |
| azonenberg | Hmm | 01:51 |
| azonenberg | If I evaporate 80/20 NiCr it should resist KOH nicely | 01:52 |
| azonenberg | without need for a secondary hardmask | 01:52 |
| azonenberg | But i dont know if it will work with tungsten boats | 01:53 |
| bart416 | azonenberg, IR camera... | 05:41 |
| azonenberg | ? | 05:42 |
| bart416 | Isn't the photoresist UV sensitive? | 05:42 |
| azonenberg | Oh | 05:43 |
| azonenberg | You mean for filming stuff | 05:43 |
| azonenberg | Yeah, IR would work if i had one | 05:43 |
| azonenberg | The only IR camera i have is focused for really short range | 05:45 |
| azonenberg | like 6" | 05:45 |
| azonenberg | And isnt easy to refocus (electrical tape holding the lens elements together, etc :P) | 05:45 |
| bart416 | Old CCD webcams are pretty IR sensitive if you remove the filter and replace it with a piece of photo film | 05:50 |
| azonenberg | bart416: Thats actually what i have | 05:50 |
| azonenberg | I have no filter at all | 05:50 |
| azonenberg | so it's sensitive to visible + IR | 05:50 |
| bart416 | Ah | 05:50 |
| azonenberg | But the lens got a bit messed up and kinda hacked | 05:50 |
| bart416 | Then again, I sort of had a lot of time to perfect it | 05:50 |
| azonenberg | so its held together in what happens to be a very short range focus | 05:50 |
| bart416 | We have to do some sort of uhm thesis like test at the end of highschool | 05:51 |
| bart416 | (obviously less complicated) | 05:51 |
| bart416 | Mine was about tracking a hand against a wall | 05:51 |
| bart416 | We solved the skin colour problem by IR, also way less noise heh | 05:51 |
| bart416 | But it depends on a camera to camera basis if it works well | 05:52 |
| bart416 | The cheaper ones are usually easier | 05:52 |
| bart416 | azonenberg, how paranoid are those idiots :| | 06:58 |
| azonenberg | lol no idea | 06:58 |
| bart416 | Also, I got my BGA soldering process to work while drunk, lol | 07:54 |
| azonenberg | lol | 07:59 |
| bart416 | That's the good news | 08:01 |
| bart416 | The bad news is that I went through roughly 40 ICs to get it to work well | 08:01 |
| azonenberg | Lol | 08:02 |
| azonenberg | What pitch? | 08:02 |
| azonenberg | and what kind of pcb? | 08:02 |
| bart416 | scrap pieces of dual layer PCB | 08:03 |
| azonenberg | And what kind of chips? | 08:04 |
| bart416 | 0.4mm | 08:04 |
| azonenberg | o_O | 08:04 |
| bart416 | NVRAM | 08:04 |
| azonenberg | and i thought the 0.8s i had were small | 08:04 |
| azonenberg | Grr | 08:06 |
| azonenberg | Four more pins to route | 08:06 |
| bart416 | lol | 08:06 |
| azonenberg | in a sea of wires between an FPGA and a SRAM | 08:06 |
| bart416 | I'm looking into package on package for my ARM Cortex board | 08:07 |
| azonenberg | http://i.imgur.com/fWQ3c.png | 08:08 |
| azonenberg | This design really needs six layers | 08:09 |
| azonenberg | And probably a bga fpga | 08:09 |
| azonenberg | That one little spot right between the fpga and the ram is getting too dense to route anything | 08:09 |
| bart416 | heh I'm first drawing everything on paper >_> | 08:09 |
| azonenberg | i probably wont be able to clock the ram faster than 20-40 MHz given those deltas in trace length | 08:09 |
| azonenberg | And how would you do that for something like this? :P | 08:10 |
| azonenberg | One of the problems is that this fab (expresspcb) cant do really tiny vias | 08:10 |
| azonenberg | Or blind vias | 08:10 |
| bart416 | they might just work azonenberg | 08:10 |
| azonenberg | So i cant route under any power vias | 08:10 |
| azonenberg | And they cant do vias smaller than 660um outer diameter with a 200um hole | 08:11 |
| azonenberg | *wants laser-drilled microvias* | 08:11 |
| bart416 | lol | 08:12 |
| R0b0t1 | Blind via? | 08:12 |
| azonenberg | R0b0t1: A via that doesnt penetrate the board all the way | 08:13 |
| azonenberg | and goes from an outer layer to an inner one but not out the other side | 08:13 |
| azonenberg | buried vias go between two inner layers and dont penetrate either surface | 08:13 |
| R0b0t1 | Ah | 08:13 |
| azonenberg | This fab cannot do either | 08:13 |
| R0b0t1 | I'd never seen those on things I've handled. Are they relatively rare? | 08:14 |
| azonenberg | Relatively | 08:14 |
| R0b0t1 | Well, one you can't see, but... | 08:14 |
| azonenberg | Except on things like GPUs and mobos | 08:14 |
| azonenberg | Where there are so many you'd never see them | 08:14 |
| azonenberg | And cell phones | 08:14 |
| azonenberg | Those things are so dense you wouldnt believe it | 08:14 |
| azonenberg | They basically pile the top and bottom board surface with components | 08:15 |
| azonenberg | laser microvia in every pad | 08:15 |
| azonenberg | and do all the interconnect on inner layers with buried vias :P | 08:15 |
| azonenberg | i'm exaggerating a little bit, but they're DENSE | 08:15 |
| azonenberg | Check out this guy for example http://www.huv.com/gumstix/Gumstix2-big.jpg (gumstix single-board computer) | 08:16 |
| azonenberg | All BGA for the ram, flash, and cpu | 08:16 |
| azonenberg | one tssop | 08:16 |
| azonenberg | looks like a SMT cap array for decoupling (or maybe that's bus termination resistors) | 08:17 |
| azonenberg | lots of 0402 or 0201 decoupling caps | 08:17 |
| azonenberg | vias everywhere | 08:17 |
| azonenberg | In the back http://www.huv.com/gumstix/Gumstix-big.jpg you can see the lack of significant blind vias on this one | 08:18 |
| azonenberg | since the BGA vias go all the way through | 08:18 |
| azonenberg | But they are using a process that can do uber-tiny traces that go between bga pads etc | 08:18 |
| azonenberg | The one i'm using doesnt go below 150 micron trace width | 08:18 |
| bart416 | Some motherboards and cellphones have up to 8 layers | 08:19 |
| azonenberg | try 12+ | 08:19 |
| bart416 | And sometimes even components embedded in between the layers | 08:19 |
| azonenberg | Yeah, those are rarer | 08:19 |
| azonenberg | But not unheard of | 08:19 |
| bart416 | 12+ is only on very expensive high end products though | 08:21 |
| azonenberg | Yeah | 08:21 |
| bart416 | 8 is normal manufacturing at this point | 08:21 |
| azonenberg | Like the ones that use $20,000 FPGAs in 1900 ball packages? :P | 08:21 |
| bart416 | Yeah, that sort of stuff, lol | 08:21 |
| bart416 | "Oh hey 4 power layers!" | 08:21 |
| azonenberg | Yeah, i wanted to do that | 08:22 |
| bart416 | I'm doing 500 pins on two layers | 08:23 |
| bart416 | It's mathematically possible | 08:23 |
| bart416 | heh | 08:23 |
| azonenberg | 500 total pins, or 500 pins on one chip? | 08:23 |
| bart416 | one chip | 08:23 |
| azonenberg | bga? | 08:23 |
| bart416 | yes | 08:24 |
| azonenberg | But i'd be willing to bet you have big pitch and tiny traces | 08:24 |
| bart416 | Why do you think I needed to perfect 0.4 mm pitch process? | 08:24 |
| azonenberg | I assumed it was for something small lol | 08:25 |
| azonenberg | And http://www.ami.ac.uk/courses/topics/0262_hdi/index.html | 08:25 |
| azonenberg | These things are blurring the line between PCBs and ICs | 08:25 |
| azonenberg | using photolithography and plasma etching for vias | 08:25 |
| azonenberg | resistors and caps embedded in the board | 08:25 |
| bart416 | I wonder if you could layer / layer deposit your own PCB to more than two layers | 08:27 |
| azonenberg | Given a vacuum evaporator and such, i could make them using IC-style materials :P | 08:27 |
| azonenberg | 500nm Cu + 200nm SiO2 | 08:27 |
| azonenberg | per layer | 08:27 |
| azonenberg | Start with a dual layer FR4+30um Cu substrate for power | 08:28 |
| azonenberg | lay down a thick dielectric layer, punch vias | 08:28 |
| azonenberg | fill vias with conductive material and CMP flat | 08:28 |
| azonenberg | Well its 0430 here and i've been up a bit longer than planned... time to get some sleep lol | 08:29 |
| bart416 | lol | 08:31 |
| bart416 | 'nite | 08:31 |
| R0b0t1 | azonenberg, whoah | 09:03 |
| berndj | azonenberg, hi. speaking of vacuum evaporators, do you think FR4 would outgas too much for a nice layer of Cu? at least Cu isn't crazy reactive like Al is, but still... need to have a good free path length | 14:07 |
| berndj | that pacman ghost thing or whatever it was is cute :) | 14:07 |
| nathan7 | berndj: the nyanotechnology? | 14:39 |
| lekernel | given that FR4 smells sometimes, it probably outgases | 15:38 |
| azonenberg | berndj: Hmm | 17:34 |
| azonenberg | I've evaporated Cu before, but on Si and SiO2 | 17:34 |
| azonenberg | and Ta2O5 | 17:34 |
| azonenberg | if anything outgasses it'd be the epoxy | 17:34 |
| azonenberg | the glass wont care | 17:34 |
| azonenberg | lekernel: What you might be smelling is particulates though | 17:35 |
| azonenberg | But yes, it probably does | 17:35 |
| azonenberg | berndj: also, that's a nyan cat | 17:35 |
| berndj | oooh the one that's rainbow-propelled! | 17:37 |
| azonenberg | Yeah | 17:40 |
| azonenberg | Didnt you recognize it? | 17:40 |
| berndj | no, my pop culture fu is rather weak | 17:40 |
| B0101 | Azonenberg: Hi, have you got any updates or discoveries | 23:48 |
| azonenberg | B0101: Running some copper experiments tonight in an hour or so | 23:49 |
| B0101 | hmm for the interconnects? | 23:49 |
| azonenberg | For hardmask, actually | 23:49 |
| azonenberg | I want to see a) how well it survives KOH, and b) design a better etch | 23:50 |
| azonenberg | SC2 is too fast | 23:50 |
| azonenberg | gotta dilute it | 23:50 |
| --- Sun Jul 31 2011 | 00:00 | |
Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!
