Topic for #homecmos is now Homebrew CMOS and MEMS foundry design | http://code.google.com/p/homecmos/ | Logs: http://en.qi-hardware.com/homecmos-logs/
<azonenberg> Screw thread of some sort to focus the two cameras to each other
<azonenberg> two scopes*
<azonenberg> then one vertical focuser to focus them onto the mask
<glowplug> 17" LCD and a mini ARM computer for the cam outputs would work well.
<azonenberg> Yeah, you dont need a lot of horsepower
<glowplug> Adjustment is with manual micrometers correct?
<glowplug> If thats true then you need zero horsepower. Haha
<azonenberg> Yes
<azonenberg> The wafer is just on an x-y-theta stage
<azonenberg> for mass production you'd need machine vision
<azonenberg> but this isnt :p
<glowplug> We have scipy for that anyways. =)
<Sync_> open cv ftw :D
<glowplug> That too. =)
<azonenberg> Yes, but that would need more CPU is what i mean
<glowplug> Absolutely.
<glowplug> The developer station would need an upgrade at that point. Haha
<glowplug> I have an idea.
<glowplug> The VESA table mounts for LCD's
<glowplug> We can mount the LCD, the exposer and the optical unit each on a VESA mount.
<glowplug> Then everything can be locked down to a table without any special fabrication.
<glowplug> Place sample. Rough alignment. Swing the optic unit over. Do your final alignment. Swing it back. Swing in LED panel. Hit blastoff. Swing it back. Take sample + enjoy. =)
<azonenberg> oh and the rpi would have to also have GPIOs for the LEDs
<azonenberg> and a timer
<azonenberg> but thats trivial
<glowplug> The mounts would be clamped to the rear of the desk with the sample clamped down to the front.
<glowplug> What do you think?
<Sync_> I think it is not time to think about that
<Sync_> but it seems it is a good idea
<glowplug> Holy shit sync_ thinks I had a good idea!
<glowplug> And it was the damn stupid VESA mount one.
<glowplug> Take what I can get...
<Sync_> actually, I'm not here to diss your ideas ):
<glowplug> I know. =)
<glowplug> You were totally right about the projection litho.
<glowplug> Thats for superhumans.
<glowplug> And right about the furnace too. $1,000 is actually not a bad price.
<glowplug> One thing I still think should be DIY is the optical encoder. I'm still going to give it a shot maybe I can save us ~$200.
<glowplug> Depositing of silicon dioxide is done in the evap?
<azonenberg> Normaly people do PECVD
<azonenberg> we have a couple of opions
<Sync_> you need to sputter that
<azonenberg> options*
<azonenberg> spin-on glass is one
<azonenberg> RF sputtering of a glass target is another
<glowplug> So thats the magnetron.
<azonenberg> Reactive sputtering of Si with Ar + O2 plasma is another
<glowplug> Alright just making sure I wasn't missing something important. Haha
<glowplug> Wow manufacturer yields can typically be below 30%.
<glowplug> This is a depressing business.
<azonenberg> Yes, but thats for leading-edge stuff
<azonenberg> they can be quite a bit higher for older-gen stuff
<azonenberg> And these are for complex devices
<glowplug> God point.
<azonenberg> If we can make one NAND2 gate work on a 10um process from scratch i'll be really happy
<glowplug> We are coming into the party at circa 1990 tech.
<azonenberg> Yeah, if that
<azonenberg> maybe 80s
<glowplug> Haha. That is awesome.
<azonenberg> I just want to make a board one day that has a custom flip-chip ASIC on it
<glowplug> You mean this device here? http://en.wikipedia.org/wiki/File:CMOS_NAND.svg
<azonenberg> even if it could be replaced by a few 7400 chips
<azonenberg> And yeah
<glowplug> So thats four contact pads and four transistors.
<glowplug> I think we can pull that off. =)
<azonenberg> Five contacts - inA, in B, power, ground, out
<azonenberg> This is what your layout would look like roughly (actual photo) http://i.imgur.com/su94J06.jpg
<glowplug> Ahh I see.
<azonenberg> greenish is oxide over undoped Si
<azonenberg> reddish is active area
<azonenberg> yellow is poly for gates
<glowplug> Ahah! That is much better.
<azonenberg> metal has been removed already
<azonenberg> So you'd have a strip on the top going from Vdd to the left side of the PMOS and from Vdd to the right side of the PMOS
<azonenberg> then from Vss to the left side of the NMOS at bottom
<azonenberg> then metal from the middle of the PMOS to the right of the NMOS is your output
<azonenberg> and then the big squares on each piece of poly are where you'd hook up the inputs
<azonenberg> The device in question btw is the RSA SecurID
<azonenberg> its a 1-micron 2-metal process
<glowplug> Haha
<glowplug> No way they only used 1 micron. Thats sweet.
<azonenberg> Yes they did
<glowplug> We can totally match that. 8)
<azonenberg> i have full die images at transistor resolution
<azonenberg> of every layer
<glowplug> Haha that is epic.
<glowplug> I haven't searched yet.
<glowplug> Are there example masks for these devices?
<azonenberg> Full masks are rarely published
<azonenberg> a few shops like Alliance have open source standard cell libraries
<azonenberg> with layouts for single gates you can study
<azonenberg> oh, and look up the GDS-II file format
<azonenberg> it's the industry standard for masks
<azonenberg> so whatever laser system we make will need to read gds files
<Sync_> or some gds to gcode thing
<azonenberg> It may not do so directly
<azonenberg> another good file format to use natively is gerber
<azonenberg> you could probably convert gds to gerber
<azonenberg> but if you want to be compatible with IC layout CAD tools we need to eventually be able to read GDS
soul-d has quit [Remote host closed the connection]
<glowplug> Lots of interesting stuff in here. http://vlsiarch.ecen.okstate.edu/flows/
<glowplug> A full license of LayoutEditor is $1500. Ouch.
<azonenberg> Glade is free but not open source
<azonenberg> that's what i've been using
<azonenberg> it is almost certainly not good enough for anything huge, it seems to be mostly a viewer
<azonenberg> but it does exist
<glowplug> That is free. And works. And is very old.
<azonenberg> magic? Yes
<azonenberg> its ancient
<glowplug> Sounds like a perfect match for our tech. Haha
<azonenberg> Lol
<azonenberg> good point :P
<Sync_> magic is also still developed
<Sync_> I know someone who made quite a large asic with it
<glowplug> I'm sold (pun intended). 8)
<azonenberg> Sync_: how large? and what process?
<azonenberg> can it do extractions too?
<azonenberg> or just layout
<Sync_> oh I dunno
<azonenberg> and does it include P&R and synthesis stuff?
<Sync_> no
<Sync_> iirc
<glowplug> It looks like it can do extractions.
<azonenberg> i mean for initial designs we wont be using hdl
<azonenberg> it'll be manual drawing
<glowplug> Timberwolf can be used for place and route.
<glowplug> Version 6.3 is GPL
<Sync_> I guess it's not large for todays standards but it was a video processor
<azonenberg> Nontrivial
<glowplug> video processor!
<Sync_> yeah it was not something really simple
<glowplug> Good enough for me. Haha
<glowplug> This site has everything we need.
<glowplug> Even some things we wont need for years.
<Sync_> azonenberg: iirc it was nmos
<glowplug> Our goal is CMOS devices?
<azonenberg> CMOS would be nice
<azonenberg> mems comes first
<Sync_> I need to get me some wafers from work
<glowplug> By the way my name is Adam. I know Andrew now. Sync_ ?
<B0101> I don't think oxides will grow on Silicon wafers that are put in desiccator bags, will they?
<azonenberg> you'll still have native oxide
<azonenberg> it wont be thick, but it'll be there
<Sync_> iirc 15nm or so at room temp
<azonenberg> Homebrew CMOS and MEMS foundry design | Wiki: http://homecmos.drawersteak.com/wiki/Main_Page | Repository: http://code.google.com/p/homecmos/ | Logs: http://en.qi-hardware.com/homecmos-logs/
azonenberg changed the topic of #homecmos to: Homebrew CMOS and MEMS foundry design | Wiki: http://homecmos.drawersteak.com/wiki/Main_Page | Repository: http://code.google.com/p/homecmos/ | Logs: http://en.qi-hardware.com/homecmos-logs/
xiangfu has joined #homecmos
<Sync_> ah 1.5nm seems to be the equilibrium
<Sync_> so I had at least the right number!
<B0101> http://www.instructables.com/id/Simple-DIY-Vacuum-Chamber-and-Pump/ - I wonder if this DIY Vacuum pump is good enough for RIE
<azonenberg> No
<azonenberg> aspirators are limited to the vapor pressure of water
<azonenberg> and the chamber gets full of water vapor at that pressure
<Sync_> also water everywhere
<Sync_> I actually had a closed loop water aspirator for a certain project
<glowplug> Have you guys seen this?
<B0101> seen
<glowplug> I knew about jeri but I haven't seen these videos yet.
<glowplug> Have we asked if shes interested in helping?
<glowplug> Spin coating with a CPU fan. Holy shit. Haha
<glowplug> And it looks like she has a glass tube in the furnace.
<Sync_> quartz
<Sync_> they actually come like that
<glowplug> Which appears to be a grill.
<glowplug> Thats not a commercial furnace is it?
<Sync_> I guess it is
<glowplug> Hmm
<Sync_> haha holy shit her diode is bad
<glowplug> All the devices were pretty bad. But they were made without any litho.
<glowplug> Extremely impressive.
<Sync_> yeah well a mosfet is extremely easy to make
<Sync_> because it just needs a channel
<Sync_> and metal
<B0101> hmmm, to make an electrical field, the potential difference between 2 charged plates is inversely proportional to the distance between my plates, am I right?
<azonenberg> I've talked briefly with jeri
<azonenberg> She was surprised that nobody had copied her setup yet
<glowplug> I'm surprised too.
<azonenberg> BTW, my guess is that one of the reasons for her trouble was the complete lack of attention paid to things like metal contamination
<glowplug> We need to get our asses in gear.
<azonenberg> i bet her stuff was loaded with sodium
<glowplug> I'm sure she knows that too. But the fact is she has a working device.
<azonenberg> Yeah
<azonenberg> But an RCA clean isnt that hard to do
<glowplug> I think she would be interested in collaborating if we can prove we are serious and get a working device as well.
<azonenberg> Maybe
<azonenberg> We can definitely try and pick her brain
<azonenberg> I'll ping her on FB once we get some more serious progress
<glowplug> Sounds like a plan.
<azonenberg> Yeah, geared motors of that level exist
<azonenberg> but they're not necewssarily precise
<azonenberg> i bet the backlash is terrible
<glowplug> Probably better than my diy geargoxes though.
<glowplug> Did you have any gearboxes in mind?
<glowplug> Here are the insides of the Kysan motor. http://www.flickr.com/photos/hoeken/sets/72157612405194866/
sivoais has quit [Ping timeout: 240 seconds]
sivoais has joined #homecmos
superkuh has quit [Quit: DayZ]
xiangfu has quit [Remote host closed the connection]
soul-d has joined #homecmos
superkuh has joined #homecmos
B0101 has quit [Ping timeout: 264 seconds]
B0101 has joined #homecmos
B0101 has quit [Ping timeout: 264 seconds]
B0101 has joined #homecmos
* B0101 seriously never thought that sending a 22 GB backup file over a local network would be difficult
<glowplug> Do you have linux systems or is the transfer CIFS?
<Sync_> azonenberg: my prof said use quartz, it is the easiest and safest
xiangfu has joined #homecmos
B0101 has quit [Remote host closed the connection]
<glowplug> A quartz tube for the furnace?
<glowplug> Some interesting information. First on the gearing for the x,y stage motor. http://www.zincland.com/hypocycloid/
<glowplug> And also this motor controller is interesting. http://www.youtube.com/watch?v=S9PyQBrqR4E
<glowplug> Standard ungeared BLDC motor spinning at 10rpm constant.
xiangfu has quit [Ping timeout: 256 seconds]
<glowplug> If you combine the two you get .0016 r/s. With a .5 pitch leadscrew (standard micrometer) which is .5mm /r or .0008mm per second at the lowest speed. 800nm. =)
<glowplug> This is of course assuming the slowest we can turn a BLDC motor is 10RPM. I can't find any examples of slower but it can probably be done.
<glowplug> At any rate that makes the motor/gear assembly (not including leadscrew, nut, or encoder) about $15. The raw material for the acrylic cycloid gearbox and the cheap hobby outrunner.
<azonenberg> Sync_: i'd use quartz for the actual wafer carrier
<azonenberg> the question is whether alumina insulation on the furnace is dangerous
<Sync_> yeah I think not
<Sync_> if the carrier is quartz everything sould be fine
<azonenberg> I wont even consider trying to clean my existing glassware
<azonenberg> when i move to CMOS i'm buying all new fused quartz
<azonenberg> and never handling it without gloves
<Sync_> the mbe chamber shat itself again :/
<Sync_> at first the thickness monitor died
<Sync_> and now we have a short in the vacuum
<glowplug> So I understand. The quartz carrier for the furnace is a cylinder correct?
<Sync_> no
<glowplug> Ahh that is for doing multiple samples at once. Interesting.
<glowplug> How much does that thing cost?
<Sync_> they should be fairly cheap
<glowplug> I have been researching closed loop control of steppers and finally realized that the Mesa SoftDMC can do it with quadrature encoder input.
<glowplug> And it can do two at the same time.
<glowplug> I was totally wrong about the parallel port latency being too high for th encoders also. It is a severe bottleneck for high rpm operation but at low rpm's it is completely acceptable.