Science/Engineering/Technical forums

You all probably noticed I haven't been posting alot recently.
I've been building a process gas analyser for a client.
The only difference between a process gas analyser and a residual gas analyser is the process gas version samples at atmospheric pressure.
This beastie uses the Extorr 300 RGA head a Leybold turbovac 50 and an industrial PC from Avantek a 12 volt 18 watt 1.6GHZ 2 GB ram 0.5TB HDD 6 serial ports 6 USB ports two independant LAN ports (can be seem bottom right of pic).
Still to come the capillary heater that will go in the milled out diecast box in the pic the PCD of the holes = CF2.75" bolt pattern
capillary = 80 micron x 1 metre this will give 10 to the minus 6 millibar for the RGA. The heater is to stop volatiles from blocking the capillary

So, you're analyzing gas at STP (more or less). Evidently you're not sweating the recombination of ions or chem reactions in the pressure reduction capillary, right? And yes, we miss it when you don't post much.

Doug
Well as the gas sample is coming off a pyrolysis process the capillary is reasonably cold even at 125 degrees C.
And yes it is at atmosphere on the input side.
The Extorr RGA has built in pirani and hot ionisation gauges and its software controls when the ioniser is turned on so there is an unconditional guarentee on the filament. We did not get the electron multiplier on this unit, it is retrofittable though if we feel that we do need it ---just more $$$$

Ah, my bad. I should have realized that myself. I do get some real strange odd things with my mass specs when the pressure is low enough for strange things to exist - never having hit a tank wall to become something more normal. But if your feed is at some high pressure - things have already done all the hitting they need to do to become whatever end product.

In fact, I think I need to build one of these myself. We have an interesting effect in a fusor over about 10 min runtime in batch mode. Neutron output and Q both go down with time, refreshing the gas always brings it right back up. This happens even with a base pressure under 3e-9 mbar, so it's not just that we've boiled some water off the tank walls, and you can do it all day long (I'll put up some plots at some point). Gotta get that turbo driver project going on - we have spare pumps for things like this. Fusor runs at far too high a pressure to have a mass spec running in it under use conditions.

So I suppose it could be that the buildup of reaction products somehow inhibits the neutron reaction in DD (for all I know, net fusion could be going up - it's just neutrons going down, and I don't even have a gamma spec on the thing at present - which I'll fix pronto). It'd be nice to know, at any rate. We have made measurable amounts of T in the fusor, though it took reasonably heroic collection and sensor efforts to prove that.
It was never large as a percentage, of course - that's one large tank of gas with a fusor in the sidearm. Yet, this effect starts almost the instant the fusor is started and is fairly linear (eg it's probably not tank wall temp which doesn't fall that quick when I take out gas and let in fresh - that only takes a couple of seconds).

OK long time no post

Garage and other things getting in the way
Customer is anxious for his Gas analyser so I'm getting it finished
All powered up today with first Mass spectra

Okay a few pics
first is sampling ordinary air

second is the base pressure after 1.5 hours pumping from brand new items

third same pressure but to limit of mass spec head

forth
pic of sparse internals

five
pic of what the customer sees

still to mount Samsung 23" monitor on top and deliver during the next week

I'm pretty pleased with the pressure its getting to --no leaks - and fairly clean

How long does it take to react to sample changes with that long of a capillary?

Jerry
It seems to take around 10 seconds
more than fast enough for the intended application

Just finished putting in the sample pump and mounting the new monitor on top

I suppose its time to post a pic of the beast in the customers premises doing what he wants