MFC's

For the design and construction details of ion guns, necessary for more advanced designs and lower vacuums.
Post Reply
Starfire
Posts: 1482
Joined: Wed Oct 24, 2001 2:14 pm
Real name:

MFC's

Post by Starfire »

I have obtained two Sierra side-trak Mass Flow Controllers, one 0-20 SCCM O2 and one 0-50 SCCM n2.

I intend to use one of these to remotely control the Hydrogen feed gas to an Ion gun. I do not have experience of these devices, but need to calibrate them for use. My requirement is for very small rates of flow into the Ion Gun vaccum.

Does anyone have experience to convert from O2 or N2, to H2 gas equivalance and how best to calibrate flow rates? How to determine actual delivery gas mass per unit time?
Roberto Ferrari
Posts: 364
Joined: Tue Mar 02, 2004 12:21 pm
Real name: Roberto Ferrari
Location: Argentina
Contact:

Re: MFC's

Post by Roberto Ferrari »

Hi John,
I would program it for -say- 1 sccm of o2 or N2, placing at the output a calibrated flowmeter.
If you haven't access to an electronic one, try those bubble flow meters common in Gas Chromatography GC). You will get close to 10%.
Having a correction factor, I would suppose they are linear along the range.
A cheap and alternative solution would be a restrictor. Also used in many GC, they are SS frits, with different porosities, in order to give a known flow of a specific gas under a determinate pressure.
They are small cylinders, approx. 1/4 inch OD, 1/2 inch long, sealed with o-rings.
Nitrogen Specific Detectors in GC usually employ a 3 or 4 sccm H2 restrictor.
Roberto
Starfire
Posts: 1482
Joined: Wed Oct 24, 2001 2:14 pm
Real name:

Re: MFC's

Post by Starfire »

Thanks Roberto, I don't have access to a calibrated device but was thinking of discharging bubbles into a water filled volume calibrated pipette and timeing with a stopwatch.
My problem is that I need a variable gas flow rate and ideally remotely operated to control range. I don't intend to be close to the reactor when firing - just in case it works
I have needle valves but again they need calibrated also and need manually adjusted for a change of gas feed parameters. Get up close again

The Mass Flow controllers seem to be the way to go, if I can establish accuracy in measurement and control.
Roberto Ferrari
Posts: 364
Joined: Tue Mar 02, 2004 12:21 pm
Real name: Roberto Ferrari
Location: Argentina
Contact:

Re: MFC's

Post by Roberto Ferrari »

John,
Yes, I understand.
So go to a preliminar calibration with bubbles. Make your own calibrated pipette.
Do you have your chamber and vacuum system ready?
You can calibrate the mass flow controller as a leak into the system, if you are able to measure and record the pressure rise.
Assuming yor system is thight, you can calculate the flow based in pressure rise.
Roberto
DaveC
Posts: 2346
Joined: Sat Jun 30, 2001 1:13 am
Real name:

Re: MFC's

Post by DaveC »

John - other than "very small" volumes, you didn't give an estimate.... but the bubble approach is good. I have used it to add very small amounts of gases to a much larger flow of coolant gas. A mm diameter bubble has about 0.5 nano-liters volume, and represents a reasonably small bubble. With water as the "indicator" you need to consider what degree of gas absorption might occur.

Also, you may find that a tandem needle valves will work a bit more linearly.in setting the flow.

Finally, there are some quite low range floating ball flowmeters, which, even if their ranges are too large for you purpose, make rather nice, and usually, inexpensive needle valves.

Dave Cooper
Starfire
Posts: 1482
Joined: Wed Oct 24, 2001 2:14 pm
Real name:

Re: MFC's

Post by Starfire »

The Hydogen stream is the Ion source feed - but what vol/sec to use? I dont really know yet - had not fully considered this aspect of the gun design until now. Perhaps mass is the better approach than vol. What is the mass of 1 cc H2 STP then too consider the vacuum? I suspect that 1ccm per miniute is a lot more than I need working at a micron. Should I deduct the mass of the stripped electron in the calcs? Should I use silicon oil for the bubble calibration? - I am going have to take time and rethink this.

The MFC's are maximum of 20 cc/min O2 and 50cc/min N2 but will throttle well back via the 5 volt control ( 5v = max O/P ) They are sophisticated devices and contain both thermal mass flow meter and control valve with 'integeral' PID. - and a snip at $10 each. Tks Dave - any help is appreciated!

{BTW did you get the BF3 on ebay? :}
DaveC
Posts: 2346
Joined: Sat Jun 30, 2001 1:13 am
Real name:

Re: MFC's

Post by DaveC »

No.. I didn't even know there one anything on ebay. I don't check very regularly.

If you remember your Chemistry, a gram molecular volume has a weight equal to the molecular weight in grams, and a volume of 22.4 liters, or 22,400 cc. So for H2 ~ 2 grams wt for 22,400 cc. 1 cc would have a weight at STP of 2/22,400 or around 90 micro grams.

Dave Cooper
Starfire
Posts: 1482
Joined: Wed Oct 24, 2001 2:14 pm
Real name:

Re: MFC's

Post by Starfire »

Ahh! -- the bang should not be too big then - Tks Dave

It is a few years since I played with mols, Dave. { and too busy to look up just now. I have the notes somewhere - the 22.4 rings a bell :>} assuming a bar SP 1 cc -- then @ 1 micron { milli-Torr } 1 cc should be approx = 10 picograms ------ next, E=M - etc. for this delivery rate.
SteveHansen
Posts: 294
Joined: Sun Aug 05, 2001 8:37 pm
Real name:

Re: MFC's

Post by SteveHansen »

The control signal to a thermal MFC is 0-5 volts and the flow rate is proportional to the input signal with 5 volts being the full scale flow. Since the MFC is a thermal device it is gas sensitive since different gases have different thermal properties. However, for all practical purposes oxygen is the same as nitrogen. But, hydrogen, helium and deuterium have very different calibration characteristics from the heavier gases and the relationship to nitrogen is non-linear.

For very low flow rates, the MFC may not be able to control very well. Depending upon manufacturer you may find control range specs from 2-100 percent of full scale, 10-90 percent, etc.

Standard accuracy is 1% of full scale which means that at lower flows the absolute accuracy (% of setpoint) gets worse.

For some more info check http://www.belljar.net/71leaks.pdf
Starfire
Posts: 1482
Joined: Wed Oct 24, 2001 2:14 pm
Real name:

Re: MFC's

Post by Starfire »

Steve - many thanks - this is a fantastic article and most valued - my respects to the author.
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: MFC's

Post by Richard Hull »

For MFCs of high flow rate, (most often encountered surplus from old, torn down fab lines), you might not be able to get a slow enough DC controlled flow rate. Electronically "bumping" an MFC is possible rather than hoping the thing responds well or linearly to a .003 volt dc signal. Where there is a will, there is usually a way.

The great thing about a formal MFC is that they are made to come out of the chute metering gases and are not gas kludges.

Some MFCs can be had very inexpensively, as mentioned above, from old fab lines at surplus, but you have to watch out for what gas they passed in a former life and check for possible massive residues which might have reacted with internal components and lines.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
Post Reply

Return to “Ion Gun Design and Construction (& FAQs)”