Residual Gas analyzer Questions

[Tom Dressel]

I was wondering if anyone could discuss the pros and cons of using a RGA in a fusor. What little I know about RGA's is gleaned from O'Hanlon's guide to vacuum technology. It seems to me that a RGA would be a nice toy to add to the fusion play room. One could get a reading on the D2 pressure in the fusor, some can be used as a total pressure gauge and a helium leak detector. It would probably mean being stuck with using a turbo pump rather than a dirty old dif pump.

Would a RGA have enough sensitivity to detect the gaseous products of fusion, and be yet another way of documenting that fusion is actually taking place?

[plasmann]

I would think it would be a great tool to have to monitor fusion reactions. You could see your D2 and He3 background background partial pressures before during and after operation. I don't know of any con's of having one installed. You could operate with a diff pump, but it would probably get contaminated. That could be greatly reduced by having a LN2 cold trap on your pump inlet.
As to the sensitivity, I would imagine that if there was a reasonable fusion rate, it would be possible to get meaningful data.

[guest]

What you have to remember is that a spectacular fusion yield for the likes of us is around 10^5 reactions/second. This ain't much compared with molar quantities (6 X 10^23). Even an hour's solid run won't leave you with a lot to detect. You wil need to compare these numbers with the sensitivity of the RGA in question. The 3H and 3He products of the D-D reaction also come out with considerable energy (~1Mev), and are more likely to end up embedded in the chamber walls than running about loose.

[Richard Hull]

I have a bit of knowledge regarding the RGA. Tom Ligon used a brand new SRS unit for the last three years.

The RGA is not supposed to even be turned on until the pressure is 10e-5 torr!!! You will ruin the filament and gunk the unit up in minutes operating at 10e-3 torr or micron levels we work at.

The SRS engineers cringe at the thought of operating at even 10e-4 torr!!! Classically, the device is used to inspect deep vacuums for contaminants and identify same.

The sensitivity, especially with the electron multiplier option can approach 10e-14 torr pressures of residual gas pressures. This is incredibly sensitive, but again, we could not run an RGA in our fusors due to the high pressures and fouling of the device.

The fusors operated by some IEC fusor researchers often runs in the 10e-5 torr range which will allow the RGA to work well.

Our simple devices demand the high presures.

Richard Hull

[Tom Dressel]

So you are saying that the only way to use an RGA in a fusor is to have a differentially pumped system. That is, pump the fusor gasses through a throttle into a lower pressure area where thr RGA do its work. Which is probably unreasonably complex.

[Richard Hull]

Yes, only a differentially pumped operation would work and save the RGA from taking a beating.

However , NO, you can go ahead and use an RGA at micron levels just fine. Just measure its lifetime to expensive repair in hours.

I was bummed to learn this myself about three year ago in a talk with the folks at SRS. I was contemplating bitting the bullet and spending the then $3500, but now $4500.

Richard Hull

[Tom Dressel]

I talked to Al Dieken last night, along with a lot of NIM components, he has an RGA, complete with tube, software, and electronics, except the RF module. Would this item still be considered essentially worthless wihtout the RF module?

[plasmann]

Yep, you can't do anything at all without the RF module, you need it to drive the quadrapoles at the right frequency.
I have the same problem. I have the controller and the differentially pumped RGA head, but I'm missing the RF head. If anyone knows where I can find a Inficon IQ 200 RF head, that would be of great help.

[Richard Hull]

The two messages about one KEY missing component is common in the surplus biz.

It is normal for many companies and even government agencies to offer the units for sale at a fixed price as a working item in a special offering with all components cables, etc. If they fail to sell as working units, they will sometimes disgard or trash a key component and palletize the rest for a simple surplus auction by the skid.

In other cases, especially government auctions, the cables which are copper go to a metals auction and the simple electronics to a palletized electronics auction warehouse and large boxed or mechanical attachment hardware items often go to office equipment auctions within the same system. Some few pieces may go to hazmat disposal centers.

So a complicated item in good working condition with lots of attachments and components can be broken up and sent to many auctions based on a decision made by the property disposal manager.

More often, the stuff just gets palletized based on shape and offered at the same auction facility. When the auction is over, a bidder might win pallet numbers 12, 29 and 34 while key components to make some of the items work were on pallets 15 and 16. The bidder often doesn't know or doesn't care either.

By the time we see it at a hamfest or surplus store it makes us just cry that a $100,000 item is complete in every way except for $1,710.94 worth of special cables with oddball and special connectors.

Oh well. The best we can hope for is to locate the parts we need or cobble something up to allow operation. This only rarely works out.

Good luck to both Tom and Mike.

Richard Hull

[SteveHansen]

The RF part of the RGA has to be present and tuned to the exact probe (sensor) that is being used. This is reasonably tricky.

On the pressure range over which an RGA is usable, the probe has to be immersed in a high vacuum: no more than 10-4 Torr for a Faraday cup equipped probe and less than 10-5 Torr for one with an electron multiplier. The reason for high vacuum is primarily to provide a long enough mean free path so that the ions can proceed relatively unimpeded along the length of the quadrupole analyzer.

Because a large industrial use of RGAs is in PVD (sputtering) systems that operate around 2-5 mTorr, there are some RGAs that will work at this pressure without differential pumping. Resolution and sensitivity suffer a bit but it's good enough for basic process monitoring. These units have very short analyzers.

Steve

[Tom Dressel]

Mike:

You may be in luck.

I was at Global Machenery on Friday, they had many Herraeus Leibold (Inficon I think) RGA parts including the RF/DC modules. No cables though.

Give them a call @ 612-302-9469 and talk to Marv Renolds or Russ Renolds.

Tom Dressel

[guest]

ok the last place i worked at used them alot and yes if below 10-5 you will destroy it some systems come with a pin hole cover so that you can ion pump the rga and not the whole system this keeps the rga at a lower pressure while the tank itself is at a higher pressure.

the only problem is the slow pumpdown times involved with ion pumps and the only way to speed them is to use a bake in on the pump itself. (i have had to do this before on the SEM i worked on when we replaced the lab6 filiment on it and needed it up fast (the filiment area needed to be at 10-8 while the chamber only needed about 10-5 for operation.

pat

[DaveC]

I had a VG 300 MAss Spec for years in my lab. All told we paid about $45 - 50K for everything...

With Thoriated -Iridium filaments ( the only way to go at high pressures) you COULD operate at 10^ -4 Torr. But as has already been mentioned, sensitivity and resolution suffer. The Channeltron or Electron Multipliers operate at up to 3500 VDC and thus require pressures below 10^-6 T for good life. The Faraday cup and electrometer limit current detection to about 10^-9 amperes, which is roughly the minimum pressure limit in Torr.

Several years ago, there was an article in one of the monthly Scientific Instruments mags, that described how to implement the RF drive using a PC and a DAQ card. The frequencies are typically in 1 - 3 MHz range, with a DC component and up to about 1300 volts or so.

The RF unit connects to the Quadrupole rods with a pair of matched Coax Cables... RG 58 or 59, size, I think. The connectors are usually MHV or HV BNC types. But nothing really exotic. They do have to be exactly the same length, so the phase of the RF is precisely the same on the rod pairs. Usually, you set everything up, get the ionizer going- 30 to 70 eV, set a reasonable extraction voltage of 5 - 20 volts, and twiddle the frequency around the base value till the output peaks for some standard gas... like nitrogen.

Sorting out the rest of the cables, without a schematic could be a nightmare.

Ion Pumps work nicely for these instruments, but Turbos ARE faster. Even the Diff pump will work, but you do need to be careful about back streaming ( use an LN2) trap), and you need the low vapor pressure diff pump oils (Santo Vac 5 (Poly Phenyl ether), or the DC-7or something) . A molecular sieve on the foreline will help keep roughing pump oils out of the system..but cuts the pump rate in about half.

As to sensitivity....at a typical minimum detectability of about 10^-14 Torr, which is set by the Amperes per Torr figure for the analyzer, the multiplier absolute current gain and the electrometer minimum current, this is about 10^ - 9 moles. This is still a pretty big number. At the target production of 10^5 Neutrons /sec, the fusor is producing He ions at a current equivalent of 10^-13 amperes, just about the threshold detection limit. But most of these will not go into the RGA since it will have to be differentially pumped about 3 -4 orders of magnitude below its operating pressure, and only a small percentage of the He ions will actually get into the sample port.

So RGA will tell lots about what else is in your system, but we will need to get into the 10 ^9 N/sec range to see He... or run for a few days continuous.

Dave Cooper