whitequark changed the topic of #tinyqma to: design of an open hardware DDS-based QMA with a low-voltage dc/rf stage :: http://irclog.whitequark.org/tinyqma
reitffili has joined #tinyqma
<whitequark> reitffili: o/
<reitffili> hmm?
<reitffili> o/
<whitequark> you're @runefoams right?
<reitffili> yep
<whitequark> so a mass spec is pretty simple to make. you need a high vacuum system, a faraday cup and a quadrupole mass filter
<reitffili> are those things, also simple to make?
<whitequark> weeeeellll
<whitequark> define "simple"
<whitequark> you need access to a machine shop and around $2000, give or take
<whitequark> fortunately, i have access to a machine shop and some cash that i can burn to collaborate with people on making instruments
<reitffili> as simple as the list you described?
<reitffili> so hypothetically what would one need to do if one did not have 2000 dollars or access to a machine shop?
<whitequark> you can get the cost arbitrarily low if you're lucky in dumpster-diving at the nearest uni
<whitequark> machining is a hard requirement though
<reitffili> hardware is actually hard in difficulty?
<whitequark> I wouldn't call it hard, but it's certainly a step above your typical arduino project
<whitequark> you can probably go a long way with a laser cutter from a hackerspace
<whitequark> for the actual mass spectrometer
<whitequark> the envelope of the vacuum system is ideally machined from 316L stainless and TIG welded, which some hackerspaces can also assist with
<whitequark> but like, in worst case it's probably possible to do something using stuff like copper plumbing fittings
<whitequark> you do need a turbopump or a diffusion pump
<reitffili> I had no idea mass spectrometers were so big, for some reason when I think meter I think about things that could fit in a hand?
<whitequark> the actual mass spectrometer can fit in your hand.
<whitequark> the vacuum systems are ... well... traditionally larger.
<whitequark> that said you can make them compact BUT it'll cost you like $10k in used equipment
<whitequark> the EXT70 turbopump alone is $3k used
<whitequark> you can also make one that fits like on a half of your palm using some interesting advances in MEMS vacuum pumps
<whitequark> well, you can't
<whitequark> and I can't
<whitequark> NASA can :p
<reitffili> I guess its in how you use those things to make the spectrometer?
<whitequark> ehhhhh, not really
<whitequark> it's more of an accident overall I think
<whitequark> with these low volume components cost is essentially irrelevant to size
<whitequark> and performance is often more important than cost
<whitequark> until recently no one cared about a small massspec, so it happened to be the size it was convenient to make in to meet a given performance target
<whitequark> i think there's now a market for 'desktop' massspecs but you pay more for them or something
<reitffili> what changed?
<whitequark> I'm not sure exactly. I think they became more widespread in labs
<whitequark> and it's just more annoying to service, deliver, use a larger instrument
<whitequark> like you didn't have really good turbopumps until late 80s
<whitequark> or sensitive detectors
<whitequark> and the larger your rods are, the easier it is to detect the signal
<reitffili> I'm having some fun immature thoughts.
<whitequark> ...
<reitffili> ok not fun.
<reitffili> yeah I was looking at a picture of one in a wiki article.
<reitffili> what are some average performance targets for practical use?
<whitequark> this one is a REAL FANCY one
<reitffili> yep that one.
<whitequark> like wow that ceramic
* whitequark salivates
<reitffili> how do I get the skills to make that?
<whitequark> uh
<whitequark> it's hard.
<whitequark> right now i'm paying someone to develop a 1800°C sintering oven
<whitequark> and there's some fairly new gel coagulation casting process that one can use to make near-net-shape green bodies
<whitequark> I'm going to start with making the insulators from macor or something
<reitffili> how does that answer the question?
<whitequark> i don't know how to get the skills to make that because i can't make that.
<whitequark> yet, at least
<reitffili> you didn't answer the question before it either?
<whitequark> oh
<whitequark> > c:\slave32\miniconda\conda-bld\src_cache
<whitequark> argh
<whitequark> >what are some average performance targets for practical use?
<whitequark> uh hm
<whitequark> well for example the one i want will have a m/z range from 1 to 2000
<whitequark> as for sensitivity it's "whatever comes out"
<whitequark> generally an m/z rnage from 1 to 200 is the more sedate target
<whitequark> but i want to try a novel-ish technique
nathan7 has quit [Quit: Reconnecting]
nathan7 has joined #tinyqma
<reitffili> what is the novel-ish technique, why is it novel, why do you want to do it?
<reitffili> now I'm wondering how this is related to the topic of the room?
reitffili has quit [Quit: Page closed]
nortti has joined #tinyqma
<nortti> :S this seems a "bit" too technical for me
<nortti> seems really neat, tho
nortti has left #tinyqma ["Only in silence the word, only in dark the light, only in dying life: bright the hawk's flight on the empty sky. —The creation of Éa"]
tangibleghost has quit [*.net *.split]
tangibleghost has joined #tinyqma