#milkymist IRC log for Monday, 2013-01-14

wpwrakazonenberg: btw, with your chip disassembly procedures, could you strip a chip (e.g., an FPGA) to the point where it would still work but where you'd have little enough material between the silicon and the outside that you could obtain meaningful chip temperature measurements17:40
wpwrake.g., to characterize the maximum heat dissipation ?17:41
wpwrak(assuming that this is an input/constraint for synthesis)17:41
azonenbergwpwrak: You can remove the epoxy entirely using HNO3 and leave bond wires intact17:43
azonenbergthen do IR temp measurements17:44
wpwrak(assumption: if you measure through the entire package, things will already be too diffuse to be able to tell if there are dangerous hot spots)17:44
wpwrakso your process is precise enough to do this ?17:44
azonenbergUsing nitric acid, yes17:44
azonenbergthese days i've been mostly doing sulfuric which is less precise but cheaper and easier to get17:44
wpwrakah, i see. it's all in the acid then ;)17:44
azonenbergsulfuric reacts slowly so you have to heat it a lot and just soak it in the tank for like half an hour or so17:48
azonenbergnitric is much faster and more aggressive but if done right you can control it pretty precisely17:48
azonenbergif all you wnat is a bare die then sulfuric is fine17:48
azonenberghttp://siliconpr0n.org/wiki/lib/exe/detail.php?id=tutorial%3Atutorial_on_epoxy_decapsulation&media=image:walkthrough08s.jpg17:48
azonenberghttp://siliconpr0n.org/wiki/lib/exe/detail.php?id=tutorial%3Atutorial_on_epoxy_decapsulation&media=image:walkthrough07.jpg17:49
azonenbergthat's what a chip done using nitric looks like if prepared skillfully17:49
wpwrakthat looks pretty great17:50
wpwrakpoint-etching into the package adds a nice touch, too ;-)17:51
azonenbergWe drilled it first to a little above the die and then put the acid in the well18:08
azonenbergmakes for a much cleaner decap and less risk of damaging the leadframe18:08
wpwraknice :)18:10
azonenbergoh, and a lot faster too18:10
azonenbergless stuff to etch18:10
wpwrakdoes the acid heat up a lot ?18:11
azonenbergYou have to heat it or it wont etch at any appreciable rate18:11
azonenbergnormally you run the decap at 150C ish18:11
wpwrakah :)18:11
azonenbergred fuming nitric can decap at room temp if you let it sit overnight18:11
azonenberglower concentrations absolutely must be heated, as with sulfuric18:11
wpwrakonce had some fun with HCl+H2O2. made a very peroxide-rich mix. the pcb was etched in mere seconds, but the process wasn't exactly running at room temperature :)18:12
azonenberglo,l18:20
azonenbergI use 1 part HCl to 6 parts 3% H2O218:21
azonenbergat ~80C for aggressive etch or slghtly above room temp for controllable18:21
wpwraki don't have any friendly heating equipment, so i prefer processes that work at room temperature. and that ...erm, self-heating experiment was with a good quantity of 35% peroxide :)18:23
azonenberglol20:45
azonenberg35%? Ok, lol20:45
azonenbergRCA-2 is 1 part conc HCl, one part 30% H2O220:45
azonenberg... and 6 parts water to slow down the reaction to sane rates :P20:45
wpwrakyeah, that was roughly my mix, just with a less cowardly quantity of water :)20:56
wpwraki think there was also a bit of 5% peroxide in there, so it wasn't *that* nasty. but still. definitely not a process that would need 20 minutes for anything.20:58
azonenbergLol21:00
azonenbergI note also that RCA-2 is meant for *stripping* metal (when used at 85C that is)21:00
azonenbergNot for controlled etching21:00
azonenbergIt works reasonably well for PCBs at a less elevated temperature21:01
wpwrak"Solution will bubble vigorously after 1­2 minutes, indicating that it is ready for use."21:01
wpwrakat least they didn't write "impatiently" :)21:01
azonenbergfor wet etching IC-level features I normally use ten parts water to one part of RCA-221:02
wpwraks/12/1-2/21:02
azonenbergetch rate of that is a few hundred nm per minute i believe21:02
wpwrakmore than sufficient for such fine structures21:03
azonenbergYeah, and more importantly somewhat controllable21:03
azonenberglol21:03
Fallenoumwalle I fixed my toolchain with correct CSR numbers and now all MMU unit tests are passing perfectly OK :)21:29
Fallenouthanks for the pointer !21:29
GitHub192[lm32-binutils-mmu] fallen pushed 1 new commit to master: http://git.io/HiQKDg21:45
GitHub192lm32-binutils-mmu/master 0dcd80f Yann Sionneau: Update lm32 toolchain with new refactored lm32 MMU21:45
Fallenoutoolchain is now updated on github as well :)21:46
--- Tue Jan 15 201300:00

Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!