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/
mrdata_ is now known as mrdata
jamie has quit [Quit: Terminated with extreme prejudice - dircproxy 1.2.0]
Textmode has quit [Quit: "It was one dev, naked in a room with a carton of cigarettes, a thermos full of coffee and bourbon, and all his summoned angels."]
<nmz787_> macor for making electrical feedthroughs into a vacuum chamber? howso? I was planning on getting some torr-seal epoxy, or maybe trying to shove piano wire through a chunk of Viton or something
<_Sync_> well, actually making good ones vs making makeshift ones
<SpeedEvil> Doens't indium wet macor?
<SpeedEvil> It's not really very expensive
<_Sync_> indium wets some ceramics
<whitequark> that sounds very mechanically flimsy...
chris_99 has joined #homecmos
<SpeedEvil> you preload the fixture adequately so the indium is not doing any clamping
<whitequark> m
jamie has joined #homecmos
Textmode has joined #homecmos
Bike has quit [Read error: No route to host]
Bike has joined #homecmos
Textmode has quit [Quit: "It was one dev, naked in a room with a carton of cigarettes, a thermos full of coffee and bourbon, and all his summoned angels."]
NullMoogleCable has quit [Ping timeout: 245 seconds]
NullMoogle has joined #homecmos
nmz787_i has joined #homecmos
Logxen has quit [Ping timeout: 276 seconds]
<nmz787_> _Sync_: if I got some macor, drilled a hole and shoved a wire through, would I just seal with torr-seal epoxy or is there a better way?
<whitequark> You still need to seal macor to the flange
<whitequark> You could, idk, make a thread on it and fill it with indium
<whitequark> Thread to keep it inside, indium to make it sealed
<whitequark> Not sure if that could work
<nmz787_> don't most feedthroughs use some rubber o-ring too?
<nmz787_> around the flange?
<nmz787_> or can they be metal to metal as well?
<whitequark> um, what?
<whitequark> do you want to seal ceramic to metal using an o-ring?
<whitequark> that won't work, it's too porous to form a seal
<nmz787_> well you said coat it with indium first
<whitequark> probably won't even reach -1
<nmz787_> so then without an o-ring it would be metal to metal
<whitequark> indium wets metal and maybe ceramic
<whitequark> um, what?
<nmz787_> take a block of ceramiic, drill holes, add wires, seal the wires with epoxy, then coat the entire thing in indium? (it sounds like you're saying to do that)
<whitequark> no
<whitequark> I'm assuming you want something like this: http://www.globalspec.com/npapics/94/96270_11092001671_ExhibitPic.jpg
<whitequark> eh, fuck that pic, let me take one instead
<nmz787_> but the other side has a channel for an o-ring
<nmz787_> (it doesn't need to be db9 format)
<nmz787_> and probably would have to be another format for higher voltages (non-control signals)
<whitequark> well, sorta
<nmz787_> yeah
<whitequark> the one above is 40kv
<nmz787_> so I see an o-ring channel there
<whitequark> actually a CF knife edge for a copper gasket
<whitequark> but let's say it's an o-ring
<whitequark> I want to convert it to KF anyway
<nmz787_> so you're saying that the other side is coated in indium?
<nmz787_> the side i cant see?
<whitequark> no
<nmz787_> so it leaks?
<whitequark> no
<whitequark> this one is brazed
<whitequark> both the wire and ceramic to flange
<whitequark> you're saying you'll use epoxy to seal the wire inside
<whitequark> how are you going to attach ceramic to flange?
<whitequark> also epoxy?
<nmz787_> oh, I thought you were saying it would be too porous such that the entire ceramic surface needed coated with indium
<nmz787_> not just the edges
<whitequark> I suppose you could do that as well. cut a thread in macor and the flange, screw it in, seal with epoxy
<nmz787_> so you can braze metal to the ceramic?
<whitequark> sure
<nmz787_> oh
<whitequark> that was what the whole discussion above about :D
<nmz787_> had no idea
<whitequark> the industrial process is some incredibly involved shit with molybdenic acid and sputtering and what the fuck why would you do that
<nmz787_> when you mentioned indium I figured that was a special metal that was specially compatible with ceramic
<whitequark> indium would make an ok seal, I think
<whitequark> I mean, I only mentioned it because someone else did earlier
<whitequark> indium is not mechanically strong
<whitequark> brazing is very strong.
<whitequark> with brazing, the ceramic will fail before the joint
<whitequark> indium is more like a kind of gasket
<whitequark> with a fluid-solid transition
Logxen has joined #homecmos
<nmz787_> would any old brazing setup work? something I could get from craigslist or auction or local hardware store?
<superkuh> What most surprised me about the above discussion was that it's possible to grow pore free alumina.
<nmz787_> hmm, http://documents.indium.com/qdynamo/download.php?docid=20 "Bonding Non-Metallic Materials using Indium and High Indium Alloys
<whitequark> superkuh: ohhh I have an article just for you
<whitequark> nmz787_: no that kind of brazing won't work
<whitequark> you need 3% Ti in your brazing alloy
<whitequark> and do it in vacuum
<superkuh> Yeah. That's the one I read.
<superkuh> I'm still trying to figure out if the MgO only works for sintering or if there's some way to work it into electrolytic deposition.
<whitequark> ah. yeah, surprised me too.
<whitequark> electrolytic deposition of... alumina? O_o
<superkuh> ER, growth on surface electrolytically.
<superkuh> Sorry.
<whitequark> hrm
<whitequark> what?
<superkuh> I guess I can just use the common words. Anodizing.
chris_99 has quit [Remote host closed the connection]
<whitequark> well, I got that. can you write down the redox half-reaction you expect there?
<superkuh> Not of the top of my head, no. I was hoping to find something in a literature search so I could avoid doing actual thinking.
<whitequark> ok, what /at all/ do you expect there?
<whitequark> alumina is non-conductive
<whitequark> MgO is not something you can deposit electrolytically at all in... water
<whitequark> or Mg for that matter
<superkuh> Well, I suppose that answers that question then.
<whitequark> do you mean like in some kind of molten salt? that will probably attack the ceramic, and definitely attack whatever's you're trying to attach to the ceramic
Logxen has quit [Changing host]
Logxen has joined #homecmos
<superkuh> I just wanted to avoid having to find and buy an alumina tube to fit my conductor (http://superkuh.com/dfp_annotated_anode_hole.png). I figured growing alumina on aluminum would be easier than finding a perfect fitting alumina tube. But then I found out about the problem of pores.
<SpeedEvil> Pores are good.
<SpeedEvil> you can fill them with dye, and make it pretty
<whitequark> growing alumina on aluminium?..
<SpeedEvil> What is that?
<superkuh> Aluminum oxide.
<superkuh> What is always coating aluminum in air.
<superkuh> Er, Aluminium.
<whitequark> both spellings are correct :p
<SpeedEvil> superkuh: yes - I mean - what is the thing you are trying to make.
<superkuh> Anyway, not relevant to homecmos. Just thought the Lucalox was neat and wildly extrapolated.
<whitequark> Lucalox is very neat.
<whitequark> It's halfway between sapphire and small-grain-alumina.
<whitequark> haha you also have a bunch of files like that
<whitequark> oh, yeah, have you seen the plasma window stuff?
<whitequark> you can make a direct non-solid-state vacuum-atmosphere interface
<whitequark> and put e.g. electron or proton beam through it with so little attenuation it is not measurable
<SpeedEvil> superkuh: ah
<superkuh> I've heard of plasma windows but I don't understand how it works.
<whitequark> very simple
<whitequark> pV=nRT
<whitequark> T is large enough for p to rise to 1 atm
<whitequark> at the same time it is mostly confined within the window by the field
<whitequark> and where it is not, an additional pump gets it away
<_Sync_> superkuh: that's old news
<_Sync_> nmz787_: if you use torrseal you might as well just use some kind of plastic
<_Sync_> or just embed the wires in it isolated form some kind of metal body
<whitequark> actually, yes
<nmz787_i> "The only limit to the size of the plasma window are current energy limitations as generating the window consumes around 20 kilowatts per inch (8 kW/cm) in the diameter of a round window.[citation needed]"
<whitequark> I did flanges from acrylic
<whitequark> they're ok if by ok you mean complete shit
<whitequark> they do hold -2 or so though (have not tested lower)
<nmz787_i> I was thinking get a slab of Viton, jam piano wire throug (very stiff) then compress with metal plates on both sides, one of which would be the flange
<_Sync_> it always just depends on what you want to do
<whitequark> gross but might work ok
<nmz787_i> this would be for feeding into a FIB or SEM... so would need between -5 and -8
<nmz787_i> a FIB/TEM guy said torr-seal works fine for his equipment when he's rigged things up before
<nmz787_i> this is my crude drawing of what I'm thinking http://imgur.com/kSYwYzP
<nmz787_i> (no epoxy involved)
nmz787_i has quit [Quit: Leaving.]