| azonenberg_work | WOOT | 02:19 |
|---|---|---|
| azonenberg_work | Need to do some more testing but i think i just got this hardmask process worked out | 02:19 |
| azonenberg_work | the trick is to use a ton of hcl and very little h2o2 | 02:19 |
| adc | noice | 02:20 |
| adc | what was the pi u sent me | 02:20 |
| azonenberg_work | adc: Which pic? | 02:21 |
| azonenberg_work | The projct website has several | 02:21 |
| azonenberg_work | And i was having trouble with unetched Cr in trenches | 02:22 |
| azonenberg_work | I finally figured out an etch process that can remove it | 02:22 |
| azonenberg_work | Now i just need to see if my photoresist will survive it | 02:22 |
| CIA-67 | homecmos r139 | trunk/lithography-tests/labnotes/azonenberg_labnotes.txt | Today's lab notes | 03:30 |
| azonenberg_work | Getting there - some undercut, it will need tweaking | 03:31 |
| azonenberg_work | But in half an hour i expect to have results from a deep etch | 03:31 |
| azonenberg_work | Might need a little more Cr and a little less Cu | 03:34 |
| azonenberg_work | But it's the right mix | 03:34 |
| azonenberg_work | looks like Cr etches ~5nm/min while Si etches 2600nm/min | 03:36 |
| azonenberg_work | So that would mean my adhesion layer is >5nm thick, more like 50 | 03:36 |
| azonenberg_work | I'll need around a 190 minute etch to go through wafer, which means about a micron of Cr - very feasible to do with an evaporator | 03:37 |
| azonenberg_work | Then coat with a thin Cu layer to protect against oxidation | 03:37 |
| azonenberg_work | and actually, more like 500nm of Cr | 03:37 |
| azonenberg_work | since i am going both ways | 03:37 |
| azonenberg_work | But i think this is the process i will end up using | 03:37 |
| CIA-67 | homecmos r140 | trunk/lithography-tests/labnotes/azonenberg_labnotes.txt | The rest of today's lab notes | 04:54 |
| azonenberg | Die J3 @ 100x http://i.imgur.com/ttSY7.jpg | 05:09 |
| azonenberg | and 400x http://i.imgur.com/eQdeM.jpg | 05:09 |
| azonenberg | There was some undercut of the mask in a few spots, and the Cr hardmask was too thin so it got eaten through | 05:09 |
| azonenberg | And the etch was far from aligned to the crystal axes | 05:09 |
| azonenberg | But for a first attempt i dont think its bad at all | 05:10 |
| azonenberg | But we're a way out from having a working CPU | 05:16 |
| azonenberg | I have done full 8-bit designs on FPGAs (and am working on a 32-bit now) | 05:17 |
| azonenberg | In terms of home fab so far we've done basic 2D copper interconnect wiring and i just got the beginnings of a silicon etch process working | 05:17 |
| azonenberg | transistors are still further down the roadmap | 05:17 |
| flowr | how many Mhz can someone get up to, with a homebrew CPU of transistors? | 05:24 |
| azonenberg | flowr: First you have to have specs for the transistors (switching speeds etc) | 05:26 |
| azonenberg | Then you have to do the gate-level design and find critical paths | 05:26 |
| azonenberg | then you can compute the maximum delay along that path and that gives you the max frequency | 05:26 |
| flowr | critical paths = bottle neck? | 05:27 |
| azonenberg | flowr: yeah, pretty much | 05:28 |
| azonenberg | biggest delay between a pair of flipflops | 05:28 |
| azonenberg | including gates and wiring | 05:28 |
| azonenberg | My guess is, >10 MHz using discrete transistors is not happening | 05:29 |
| azonenberg | even thats high | 05:29 |
| flowr | have you done any homebrew computers this way? | 05:29 |
| azonenberg | if you integrate it you may have a chance | 05:29 |
| azonenberg | I've used FPGAs | 05:29 |
| azonenberg | and am now working from the other end up on home chip fab | 05:29 |
| azonenberg | But we dont even have functional transistors yet | 05:29 |
| flowr | home chip fab? You mean, making your own ICs? | 05:30 |
| azonenberg | flowr: Yeah, thats what the channel is about lol | 05:30 |
| flowr | but I mean, small ones :P | 05:30 |
| flowr | or is that what you mean? :O | 05:31 |
| azonenberg | flowr: Small as in low gate count? Yes | 05:31 |
| azonenberg | Small as in 22nm? No, our chips are huge dimension wise :p | 05:31 |
| flowr | that's what I thought :P | 05:31 |
| azonenberg | most of my test dies are a few mm across and the components are in a ~0.5mm disk in the middle | 05:31 |
| azonenberg | But i havent actually fabricated anything more complex than test patterns in copper wiring | 05:31 |
| azonenberg | and, most recently, the beginnings of a MEMS device | 05:32 |
| azonenberg | hoping for a simple working MEMS unit by end of this calendar year and logic gates some time in 2012 | 05:32 |
| flowr | what does one have to do, to get past 20Mhz, if one is soldering one's own gates on a circuit board? | 05:34 |
| flowr | Do you need pretty small transistors to get past 20Mhz? | 05:35 |
| azonenberg | flowr: No, you need them close together | 05:35 |
| azonenberg | I mean, getting a single gate that fast is easy | 05:35 |
| azonenberg | getting a string of gates end to end that fast is harder | 05:35 |
| flowr | is it because the copper wire between them? | 05:36 |
| flowr | of* | 05:36 |
| azonenberg | flowr: Routing delay is significant, yes | 05:36 |
| azonenberg | especially in FPGA designs due to the less efficient routing it can be >50% of your total path delay | 05:36 |
| azonenberg | If you have fast gates and long wires (like a breadboarded design) it will be even worse | 05:36 |
| azonenberg | Making a working cpu is entirely feasible | 05:36 |
| azonenberg | making it fast is hard | 05:36 |
| azonenberg | I havent really looked into what transistor speeds etc would be in a homebrew fab process yet | 05:38 |
| azonenberg | its so far out that i have more pressing things to worry about | 05:38 |
| azonenberg | Like fully etching all of my chromium layers http://colossus.cs.rpi.edu/~azonenberg/images/homecmos/2011-09-20/die_i4_009.jpg | 05:39 |
| azonenberg | this is an electron microscope shot of one of my chips taken last week | 05:39 |
| azonenberg | light gray is copper, dark is silicon | 05:39 |
| azonenberg | medium is chromium (which is under the copper and should be completely removed in the trenches) | 05:39 |
| flowr | does the chromium conduct? | 05:40 |
| azonenberg | Slightly, yes | 05:41 |
| azonenberg | But i wasnt using it as a wire | 05:41 |
| azonenberg | The chrome is there because copper won't stick to things well by itself | 05:41 |
| azonenberg | it sticks to Cr and Cr sticks to a lot | 05:41 |
| azonenberg | So a sandwich of the two is common | 05:41 |
| flowr | how did you build this? | 05:41 |
| azonenberg | I also was using the sandwich as an etch mask for a future processing step | 05:42 |
| azonenberg | Cant talk too much, its 2 am here and i need sleep, but heres a brief overview | 05:42 |
| flowr | k, maybe you should sleep :P | 05:42 |
| azonenberg | start with blank <110> silicon wafer, score and snap into ~3mm pieces | 05:42 |
| azonenberg | deposit Cr + Cu layers in a vacuum evaporator on campus (building one of my own is on the to-do list but is a few months out) | 05:43 |
| azonenberg | go back home, spin coat in photoresist | 05:43 |
| azonenberg | expose, develop, etch in HCl with a few drops of 3% H2O2 added | 05:43 |
| azonenberg | the Cr residue in this pic was the result of using too much H2O2 | 05:43 |
| azonenberg | the Cu etched fine but the Cr formed CrO2 which is difficult to remove | 05:43 |
| azonenberg | and protected the underlying Cr from further etching | 05:44 |
| Action: azonenberg AFKs | 05:44 | |
| flowr | night | 05:44 |
| azonenberg | lab notes are in the google code repo in the channel topic | 05:44 |
| azonenberg | if you want all of the gory details | 05:44 |
| azonenberg | just look at recent changes and open up azonenberg_labnotes.txrt | 05:45 |
| B0101 | hi Azonenberg | 12:21 |
| azonenberg | hi B0101 | 12:22 |
| azonenberg | Did you see my pics from last night? | 12:22 |
| B0101 | nope | 12:23 |
| azonenberg | http://i.imgur.com/ttSY7.jpg | 12:23 |
| azonenberg | http://i.imgur.com/eQdeM.jpg | 12:23 |
| azonenberg | 100 and 400x | 12:23 |
| azonenberg | the mask was somewhat undercut so the lines got attacked | 12:23 |
| azonenberg | and the edges are rough since i didnt properly align everything to the crystal planes | 12:23 |
| B0101 | whoa | 12:35 |
| azonenberg | B0101: where'd you go? | 12:35 |
| azonenberg | (08:23:19) azonenberg: http://i.imgur.com/ttSY7.jpg | 12:36 |
| azonenberg | (08:23:22) azonenberg: http://i.imgur.com/eQdeM.jpg | 12:36 |
| azonenberg | (08:23:25) azonenberg: 100 and 400x | 12:36 |
| azonenberg | (08:23:36) azonenberg: the mask was somewhat undercut so the lines got attacked | 12:36 |
| azonenberg | (08:23:59) azonenberg: and the edges are rough since i didnt properly align everything to the crystal planes | 12:36 |
| B0101 | azonenberg: my computer crashed | 12:36 |
| azonenberg | :( | 12:36 |
| B0101 | this system is old\ | 12:36 |
| azonenberg | But as a first try i think its pretty good - i need a thicker Cr layer and possibly a shorter HCl etch before the KOH | 12:36 |
| azonenberg | this was my first attempt at a deep KOH etch and it was at least somewhat successful :) | 12:36 |
| B0101 | graet! :) btw, i have IRC logging on, so i can still see you links | 12:37 |
| azonenberg | Wasnt sure if they had actually gone through lol | 12:37 |
| azonenberg | often when a client crashes it freezes for a while and stops getting data | 12:37 |
| azonenberg | then the OS closes the connection | 12:37 |
| B0101 | but I got the link 3 seconds before it crashed so it would have been logged in | 12:38 |
| azonenberg | oh i see | 12:38 |
| azonenberg | Any luck with your squids? | 12:39 |
| B0101 | yes | 12:39 |
| azonenberg | :) | 12:39 |
| azonenberg | how far have you gotten | 12:39 |
| B0101 | i managed to fabricate a simple one | 12:40 |
| azonenberg | :) | 12:40 |
| azonenberg | Out of what? And how small? | 12:40 |
| azonenberg | And have you tested it? | 12:40 |
| CIA-67 | homecmos r141 | wiki/images/ (S7301905_fullres.jpg S7301905_thumb.jpg) | Uploaded two more images to wiki | 12:41 |
| B0101 | Silicon, on a piece of silicon 5 cm | 12:41 |
| B0101 | i tested it, but it stopped working an hour ago | 12:41 |
| azonenberg | So it worked for a while, then stopped? | 12:41 |
| B0101 | yup | 12:41 |
| azonenberg | any idea why? If it was transistor based my first guess would be trace metal contamination | 12:42 |
| B0101 | No ideas yet but I am investigating | 12:42 |
| azonenberg | well make sure to post lab notes at some point so we can see | 12:42 |
| azonenberg | Remember, we dont just want a working process | 12:42 |
| azonenberg | we want to know what went wrong and how you dealt with it | 12:43 |
| azonenberg | it could be very helpful to someone else having a similar problem | 12:43 |
| azonenberg | similar symptoms might indicate a similar problem, etc | 12:43 |
| B0101 | The only thing I know is that it changed from rainbow color to silver and thats all i know | 12:43 |
| azonenberg | Hmm, rainbow color indicates a thin transparent layer (oxide etc) | 12:44 |
| azonenberg | silver would mean either much thicker or none | 12:44 |
| azonenberg | did it get hot in operation at all? | 12:44 |
| B0101 | nope | 12:44 |
| B0101 | I don't know if it is the voltage | 12:46 |
| B0101 | I applied 5V AC to it | 12:46 |
| azonenberg | No idea what they expect | 12:47 |
| azonenberg | this is your research lol | 12:47 |
| B0101 | I'm gonna go back to the lab to investigate further | 12:47 |
| azonenberg | quantum computing is not my area of interest atm | 12:47 |
| azonenberg | And ok | 12:47 |
| B0101 | See you in 18+ hours | 12:47 |
| azonenberg | ok | 12:48 |
| --- Wed Sep 28 2011 | 00:00 | |
Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!
