<nmz787>
it also says you need to do some glass to metal flange work
<pie_>
id get at least one quality gauge for calibration but thats just me
<nmz787>
(it comes in glass, sealed, with wire leads sticking out... so it is self-contained in a glass-sealed bulb to keep under vac until installation)
<pie_>
also something something never trust a single vacuum gauge
<pie_>
but $$$ i guess
<Noxz>
I got custom borosilcate glass apparatus blown before
<nmz787>
Noxz: you have to watch ebay for a while... but I guess the market changes too
<nmz787>
I haven't got one yet
<nmz787>
other than what came with the FIB
<Noxz>
could easily send the dude some gauges and tell him to make it interface in such and such way
<Noxz>
nmz787, that's the point though, I hate scouting
<nmz787>
hmm, yeah...
<Noxz>
like.. I have _some_ amount of money
<Noxz>
just pay the man
<nmz787>
it is a pain to figure out which controller for a given pump, and then the challenge of getting that controller
<Noxz>
looks like current pump+controller is a minimum of $800 + shipping on ebay
<nmz787>
another thing though to consider, if you can find a nice set, a company told me they rebuild them for like $900 or $1000
<Noxz>
and he is giving me pump, controller, hose lines, adapers, gauges, roughing pump, etc
<nmz787>
saw them at the M&M (microscopy and microanalysis) conference's vendor area
<Noxz>
maybe if it ever dies I will hit up someone that repairs
<Noxz>
where was that, btw?
<nmz787>
(I really only went to that conf for the vendor area, didn't even attend the talks... I did however upload all the PDFs of the talks that they gave me on CD)
<Noxz>
also, that was in portland? near enough that if around there again I may be able to check it out
<nmz787>
I wish the local shopping mall was like that vendor area
<nmz787>
I'd def hang out at the mall
<Noxz>
do they like to let in kids (30 year olds) that build their own SEM and Mass Specs? ;)
<nmz787>
I paid for a ticket
<nmz787>
I think I was still technically a student that year
<nmz787>
so I got in pretty cheap
<Noxz>
ah
<nmz787>
it also included a dinner at a restaurant
<nmz787>
and I think we got a drink or two there too
<Noxz>
doh, happened 4 days ago
<Noxz>
in St Louis
<Noxz>
wait, ended yesterday
<nmz787>
the one I went to was in 2015, and here in Portland OR
<Noxz>
I was in St Louis 3 mo ago
<Noxz>
to get my puppt
<Noxz>
puppy
<Noxz>
ah, so you are relatively close to me
<Noxz>
(I'm in Seattle, currently)
<nmz787>
yeah, you mentioned yesterday
<Noxz>
orly?
<nmz787>
I am just west of Portland
<Noxz>
my days merge
<Noxz>
I've been to portland once
<Noxz>
for PuppetConf
<Noxz>
how does molecular beam epitaxy condense on the substrate and not all up inside the chamber?
<Noxz>
a port/slight and low effusion/dispersal?
<Noxz>
does location ofthe vacuum port matter?
<nmz787>
my guess is that the port is situated behind the sample, and inline with the source (also called the target in sputtering jargon)
<nmz787>
and then the beam allows/enables growth/attachment
<nmz787>
as the gas//vapor/plasma passes
<Noxz>
I guess I was thinking along the lines of: once vapor forms, doesnt hte vacuum suck it all up?
<pie_>
if you have low enough vacuum its just molecules bouncing around
<pie_>
(unless im misunderstandig something in the context)
<SpeedEvil>
And if you have clean enough surfaces, they don't bounce.
<SpeedEvil>
they stick
<pie_>
oh
<Noxz>
hrrmmm
<pie_>
i didnt know that but i was thikking about beam coherence
<Noxz>
like traction/friction makes them stick?
<Noxz>
the thought is, sure, vapros eject from a port/orrifice, and propell towards your target.. but why don't they curve to the vacuum
<SpeedEvil>
Noxz: no, molecular forces
<Noxz>
because it's more of 'bouncing' and sticking, and not so much curving?
<Noxz>
ah
<SpeedEvil>
there is no oxide coating/... and the atoms simply form part of the surface.
<Noxz>
that makes more sense
<Noxz>
yeah, it'as basically what you are trying to do with doping/coating
<Noxz>
just happens to be done on the sidewalls of your vessel
<Noxz>
all over, really
<pie_>
i fart in your general direction
<nmz787>
pie_: low enough? ideally I think you'd want as high a vacuum as possible to begin, so anything other than the new stuff will be used to grow
<pie_>
*low enough presure
<pie_>
xD
<nmz787>
maybe after clearing out you could close some restrictor valve?
<nmz787>
between the chamber and the pump
<Noxz>
you mean to seal off the container at a certain vacuum and stop sucking?
<Noxz>
yeah, I doubt that would work as well though
<Noxz>
because eventually you'll increase pressure
<pie_>
for high vacuum theres usually multiple pump stages
<Noxz>
from stuff starting to bounce
<pie_>
well more like usually 2 stages
<Noxz>
but the roughing pump stays on, from what I have read
<Noxz>
exhaust of turbo feeds into roughing
<nmz787>
yeah pretty much, otherwise like you said, you'd suck it all out (unless your beam assist consumes it all fast enough)
<pie_>
te problem isnt sucking it out, its stuff leaking back in from the outside and degassing
<pie_>
mainly degassing actually i think
<Noxz>
PTFE + vacuum grease for the win?
<nmz787>
once you get up to high vac, you're just fighting leaks at that point, and molecules that aren't bouncing toward your sample but colliding with the ones going in that direction
<nmz787>
(molecules meaning the ones you want to grow with)
pie_ has quit [Read error: Connection reset by peer]
pie_ has joined #homecmos
pie__ has joined #homecmos
pie_ has quit [Remote host closed the connection]
<SpeedEvil>
High vacuum means there are functionally no bounces - the mean free path greatly exceeds the chamber dimension
<nmz787>
I would guess in some scenarios though, your process could be efficient enough to maintain high vac, if the beam of supply atoms/molecules was direct enough or something?
<nmz787>
like how an ion getter works?
<nmz787>
I mean, it consumes things... so if your epitaxy was good enough to be like that?
<pie__>
SpeedEvil, by bounces i meant off the walls
<SpeedEvil>
ah
<pie__>
ok youre right it is a bit ambiguous
<pie__>
its completely ambiguous xD
pie__ has quit [Remote host closed the connection]
pie__ has joined #homecmos
<pie__>
<pie__> its completely ambiguous xD
<pie__>
<pie__> though i would have said collisions for gas-gas collisions, which is only disambiguated in my own head