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/
<Noxz> Looks like the coil of the solenoid for the coolant line to the turbo is blown, or the valve itself may be stuck.. @ 24v it pulls half an amp, but it certainly looks burnt on the side, but I can easily blow through the valve without any power, and with power still can (maybe it is normally open, not closed? though that doesnt seem right)
<Noxz> oh, Si actually has a higher electron affinity than Cu
<Noxz> slightly
<Noxz> just read up how a quarts microbalance works for detecting layer size of depositions, really neat
<Noxz> old technology, but still neat
<Noxz> OOoo, DOI: 10.1063/1.3292510
pie_ has joined #homecmos
<Noxz> main downside to that paper is it really couldn't handle more than 500 C
<Noxz> in part by the thermocouple and the Macor which is stable to 600 C
pie_ has quit [Ping timeout: 248 seconds]
pie_ has joined #homecmos
pie_ has quit [Ping timeout: 240 seconds]
pie_ has joined #homecmos
python476 has joined #homecmos
pie__ has joined #homecmos
pie_ has quit [Ping timeout: 240 seconds]
pie___ has joined #homecmos
pie__ has quit [Ping timeout: 260 seconds]
python476 has quit [Ping timeout: 246 seconds]
python47` has joined #homecmos
pie___ has quit [Ping timeout: 246 seconds]
python47` has quit [Remote host closed the connection]
pie_ has joined #homecmos
pie_ has quit [Ping timeout: 240 seconds]
pie_ has joined #homecmos
<Noxz> neat, so people just hack out the quartz part of a crystal/opscilator
<sync> yeah, although the oscillator circuits used by qcbs are not too difficult
<Noxz> yeah, openQCMB has crystals for liek $15 USD.. or $1.50 or even less for one you hack out yourself....
<Noxz> would likely hack some out to start with
<Noxz> but the idea of it being more or less "built in" to the shutter of a knudsen cell to detect the current rate of growth is interesting
<Noxz> applied science and another kid basically explained how the balances/controllers work, and it's basically just a frequency counter that detects deltas
<Noxz> some calibration required per crystal, but yeah
<Noxz> crystal type
<Noxz> super interesting
<Noxz> that paper was also a decent read
<Noxz> but I have a feeling (I know it) that I would need more than 1000 C, so finding the right heating wire is needed
<Noxz> nichrome/knathal can get past that, not too much though
<sync> the problem is the high capacitive loading due to the long coax but it is not too difficult
<sync> nah, the sensor directly over the cell is not what you want
<sync> it will not give you an accurate reading ime
<Noxz> AlN has a MP of 2200 C
<sync> we spent quite a while to get a good correlation between the sensor rate and the actual deposition rate
<Noxz> and decomposes at 2517, heh
<SpeedEvil> SiC is commonly available as heating elements
<Noxz> hrmm
<SpeedEvil> you don't need to get close to 2200C IIRC to make AlN
<Noxz> yeah, I saw a few diff kinds
<Noxz> well, there are other methods, but that is one of them
<SpeedEvil> perhaps
<sync> knudsen cells are not the right way to go imho
<sync> there are huge issues with them
<Noxz> LiF@~900 C is really what I am loooking at
<Noxz> since it isnt soluble in anything
<sync> like not being able to shut them off usually
<Noxz> hence the shutters?
<sync> if you want any meaningful capacity
<sync> no
<sync> well, yes, but no
<Noxz> can't you just disconnect the power to the heating element?
<sync> yes
<Noxz> is that so difficult?
<sync> but then you have a cracked crucible and the material in your chamber
<sync> the problem is that due to the construction they will cool from the aperture
<sync> which encloses the liquid
<Noxz> oh, right
<Noxz> sort of why you dont use a coil condensor in chemistry for refluxing
<sync> we have so many issues with them, we just use them for stuff that sublimates
<sync> or where we know that the ceramic can take the stresses
<Noxz> ceramic crucible?
<sync> yes
<Noxz> are carbon crucibles good or no?
<sync> you can use graphite
<sync> but you need to figure out if your material goes into solution in it or the other way round
<sync> our cells are heated with carbon bands
<sync> but direct heating or electron beam evaps are the way to go
<Noxz> I'm trying to make an electron gun here ;)
<Noxz> what is the difference w/ direct heating?
<sync> find me something to evaporate carbon with
<Noxz> I am not sure I am following what you are saying..
<Noxz> I have heard of mixing carbon with material to get it to heat faster
<Noxz> say, in a microwave
<Noxz> for heat xfer
<sync> no, you cannot use direct heating to evap some materials as their melting temperatuer is just too high
<Noxz> hrm
<sync> the electron gun is the easiest part
<Noxz> well, tungsten cathodes are simple
<Noxz> but I want something more controllable later(earlier) on
<sync> for electron beam heating it is all you will ever need
<Noxz> yeah, and it may be really useful for that early on
<sync> idk, our sem also just uses a tungsten filament, works well enough
<Noxz> one of the initial goals is to make a SEM capable of writing a design to a PMMA photoresist
<Noxz> so, I need to easily turn the electron beam on/off
<Noxz> which apparently is not too easy with the tungsten heated wire
<Noxz> the idea of electron beam evaptoration is nice, I wasnt really thinking about it before, not sure why not
<Noxz> anywho, getting more hype with turbo in hand
<sync> you can just defocus the beam enough to not expose your resist
<sync> or turn the hv on and off
<Noxz> heh, sure, but I've put too much thought and reading papers into this whole photodiode thing ;)
<sync> sure, but I would just go the established way
<Noxz> meh
pie_ has quit [Ping timeout: 264 seconds]
pie_ has joined #homecmos
pie_ has quit [Read error: Connection reset by peer]
pie_ has joined #homecmos
<Noxz> does electron beam deposition work in most circumstances?wiki says some oxides dont work well
<sync> in my experience it does work very well
<Noxz> is sputtering faster than electron beam deposition?
<Noxz> seems a heavy molecule like argonw ould be more effective at knocking stuff off than a small electron
<sync> it all depends
<sync> ebeam is commonly used for aluminizing plastic sheets
<sync> at quite high pressures and high rates
<Noxz> hrm, no cheap Tungsten filaments on ebay currently, atleast none I found
<Noxz> make my own?
pie_ has quit [Ping timeout: 264 seconds]
pie_ has joined #homecmos
pie_ has quit [Ping timeout: 240 seconds]
pie_ has joined #homecmos
<nmz787> I thought not wanting to turn the beam off was why they have beam blankers
<Noxz> blankers = shutters?
<Noxz> the articles I read on using photodiode based electron guns stated one of the benefits was the ability to easily pulse
<Noxz> I think the hard part is the ability to pulse, with, say, Tungsten
<Noxz> vs just turning on/off a source 3.5v UV LED that shines onto the photodiode? easy
<Noxz> about to go through the woods with the pupper.. bbl
<nmz787> blanker is an electromagnet that pulls the beam off to the side of the mid-column, onto just a plate or something
<Noxz> ah, right, okay... but what is the reason they need that? because they cant pulse/turn on/off their electron gun easily, right?
<Noxz> thermionic emission seems like a joke to me
<Noxz> because of that
<nmz787> uh, I guess
<nmz787> peoplelike things to be in steady state I guess