Positive Pressure Leak Checks

Every fusor and fusion system seems to need a vacuum. This area is for detailed discussion of vacuum systems, materials, gauging, etc. related to fusor or fusion research.
Post Reply
User avatar
Jason C Wells
Posts: 113
Joined: Sun Oct 20, 2013 7:11 pm
Real name: Jason C Wells

Positive Pressure Leak Checks

Post by Jason C Wells »

Is there utility in testing vacuum systems with internal positive pressure and soap bubbles?

Would such a test require 15 psig or could it be done with 1 psig?

There could be component testing versus full-up system testing.

I had mentioned "certifying" components prior to use in a previous message. Positive pressure testing certain components like tubing with swageloks is certainly feasible. But is it useful? I can imagine that a number of components might show good with internal positive pressure, but leak with external pressure.

KF style flanges do not capture the o-ring well under internal pressure. I would expect them to fail a positive pressure test. I could easily machine a backup ring that would allow such a test though.

Rough pump shaft seals I think are questionable to test. The lip seal probably only seals with differential pressure in the correct direction.

Sealed pipe threads would be a good candidate for positive pressure testing. Compression fittings like swageloks could be internally tested.

Glass pieces such as my custom blown glass accelerator column present a specific danger to the user under positive pressure. Glass shards in the eyes anyone?

Your thoughts?

Regards,
Jason C. Wells
John Futter
Posts: 1848
Joined: Wed Apr 21, 2004 10:29 pm
Real name: John Futter
Contact:

Re: Positive Pressure Leak Checks

Post by John Futter »

Jason
yes 15 PSI will approximate what is happening under vacuum.
the higher the pressure the better of course up to the safe limit for the fittings as this will make the method more sensitive.
There are leak test bubble solutions that the HVAC industry uses for this purpose as the bubbles take a long time to burst (ie many hours).
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Positive Pressure Leak Checks

Post by Dennis P Brown »

I agree with you - a glass accelerator tube should be considered a lethal danger under pressure. That said, unless it is very poorly constructed, 15 PSI should never posse a threat. However, when in doubt or when ever going above 15 PSI either place metal window screen around it before placing under pressure or remove it and cap off that access port for a pressure check of the system. That all said, using a pressure check on a system using water/soap is rather useless unless it has a rather large leakage; a bubble check is all but pointless for very small leaks that keep a system from going from 10^-5 torr to 10^-6 torr. For those leaks, alcohol or "Dust Off" spray with a good HV gauge running will find a leak fast and safely.
User avatar
Jim Kovalchick
Posts: 717
Joined: Wed Apr 13, 2011 8:00 pm
Real name:

Re: Positive Pressure Leak Checks

Post by Jim Kovalchick »

Generally, testing leakage in the reverse direction should help identify leaks, but one should be mindful that sometimes the direction of the differential pressure makes a difference. The most obvious example is the bell jar. Vacuum tends to enhance the seal. Pressurizing the system gives ridiculous results. A seal or valve is designed for forward pressure isn't always good for reverse.

Please be smart about using pressure for testing. Not all components can hold pressure even though they can sustain vacuum. Think about each part of your system first.
David Kunkle
Posts: 284
Joined: Thu Apr 17, 2014 12:43 pm
Real name: David Kunkle

Re: Positive Pressure Leak Checks

Post by David Kunkle »

FWIW, I had good results pressurizing my chamber to find leaks. Hooked up the foreline to my shop compressor with some Menards fittings. If I remember, I used about 25 psi. Never had an issue with blowing out the ion gauge or the 6 inch viewport. I did this because I was suspicious of the welds- not so much joints and fittings. Sure enough, watered-down dish soap found them quickly. My leaks were in the Torr range- except for one that took some minutes to blow minute bubbles that took a close watch to see. Didn't get any false positive leaks on a 10" ISO joint and several NW25 joints. At least for larger leaks and welds, I think pressurizing is a cheap and easy technique. Other than that, I'd agree with what everyone else said about reverse pressurizing certain types of joints and miniscule leaks.
If your experiment needs statistics, you ought to have done a better experiment.

Ernest Rutherford
User avatar
Jason C Wells
Posts: 113
Joined: Sun Oct 20, 2013 7:11 pm
Real name: Jason C Wells

Re: Positive Pressure Leak Checks

Post by Jason C Wells »

How's this for state of the art do-it-yourself? I was mostly curious about the 1/8 NPT thread which is sealed with loctite. Everything looks airtight which adds one more bit of certainty to the uncertain business of leak hunting. Not shown is an air tank left over from my autocrossing days that is pressed up to 10 psi using a bicycle pump.

I made some overpressure rings for my KF flanges to prevent o-ring blowout under internal pressure.
A 10 psig positive pressure leak test of a KF50 cross, blanks, soldered passthrough cribbed from an broken thermocouple gauge, and pipe thread. Yes, that is a mop bucket.
A 10 psig positive pressure leak test of a KF50 cross, blanks, soldered passthrough cribbed from an broken thermocouple gauge, and pipe thread. Yes, that is a mop bucket.
User avatar
Nick Peskosky
Posts: 135
Joined: Wed Aug 08, 2012 9:02 pm
Real name: Nicholas Peskosky
Location: San Diego, CA

Re: Positive Pressure Leak Checks

Post by Nick Peskosky »

Make sure you bake all that stuff out once you're done testing or you will have trapped water vapor out gassing for days. Good on you for trying something off the beaten path (positive pressure testing) but submerging vacuum equipment in water is generally frowned upon in industry due to the rust pitting and moisture you will inevitably trap along the flange seems.
Nick Peskosky
NPeskosky@gmail.com

"The whole of science is nothing more than the refinement of everyday thinking." - Albert Einstein
User avatar
Jason C Wells
Posts: 113
Joined: Sun Oct 20, 2013 7:11 pm
Real name: Jason C Wells

Re: Positive Pressure Leak Checks

Post by Jason C Wells »

Thanks for the comment about the seams. I'll see about that before I re-install the parts.
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Positive Pressure Leak Checks

Post by Dennis P Brown »

Testing high vac components under water?! Very bad idea but a great idea when a car tire is exhibiting a very slow leak - especially with some over pressure in th tire. So, when your car tire leaks, use that method for tiny, slow leaks - for a vacuum system, stick to standard methods. Seriously, any leak big enough to bubble under water could generally be heard when the system is under vacuum.
User avatar
Jason C Wells
Posts: 113
Joined: Sun Oct 20, 2013 7:11 pm
Real name: Jason C Wells

Re: Positive Pressure Leak Checks

Post by Jason C Wells »

OK. So my big idea turns out to be useless at best and a blunder at worst. It doesn't appear to have made matters worse. I'm still pulling only 5 microns and still looking for my problem.

Thanks for the comments. They make a difference!

Regards,
Jason C. Wells
John Futter
Posts: 1848
Joined: Wed Apr 21, 2004 10:29 pm
Real name: John Futter
Contact:

Re: Positive Pressure Leak Checks

Post by John Futter »

Jason
if you have a big leak / suspected leak positive pressure testing is valid
All of our ultra high fittings @ work are ultrasonically treated in hot water with detergent then rinsed and put in a hot air oven at above 100 degrees Celsius
and while still hot out of the oven are assembled and pumped on. And we expect high ten to the minus eight millibar after 24 hours of pumping.

strange fittings that are leaking at above our leak detectors minimum vacuum are positive pressure tested at several atmospheres with the aircon leak detector fluid it can take 2-10 hours to see the small tell tale bubbles ( they take days to burst)
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Positive Pressure Leak Checks

Post by Dennis P Brown »

May I suggest that your "positive" pressure check with a soap solution isn't a bad idea and in fact, is a good idea that many people and professionals use! I am sorry I wasn't clear. Also, wasn't being critical of the "under water test" but didn't think it was a good idea for the average person to try (for many reasons.)
In any case, if a positive pressure/soap procedure fails to detect a significant leak, are you certain it is a leak? Have you tested just your vacuum system dead ended on the gauge?
User avatar
Jason C Wells
Posts: 113
Joined: Sun Oct 20, 2013 7:11 pm
Real name: Jason C Wells

Re: Positive Pressure Leak Checks

Post by Jason C Wells »

Dust Off is pretty cool. Two questions on your technique.

- Do you hold the can upside down to wet the joint under investigation?
- How do you get at things that are upside down?

As it turns out, I was unable to find any leaks with Dust Off. I was unsure of my technique so I loosed a couple fittings and tried again. I was able to identify the leaky fittings. After I re-tightened the connections, I was again unable to find leaks.

I was unable to force a small leak in an Ultratorr fitting. It was all or nothing with the Ultratorrs. Does that jive with your experiences?

Thanks,
Jason
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Positive Pressure Leak Checks

Post by Dennis P Brown »

The "Dust Off" gas is what one uses - the issue is that one needs to be in the low 10^-3 torr range to use this technique; that is a no go for you since you are in the micron range. A "massive" leak can easily be heard using a toy stethoscope or just getting close to the item. All other leaks - between 10-3 torr and overt "hissing" are the real tough ones. Here, a few methods can be done. First, do examine all 0-ring surfaces and o-rings with a magnifying glass and make certain that there are no scratches or cracks. Reassemble and be careful not to over tighten. Any hose clamps? Check these. Also, reduce number of parts - that is, starting at your fore pump, add just a gauge and see what vacuum you achieve. If in the tens of microns, stop and go no further. You found your problem. If under 10 microns (what ever your pump generally can do - 4 microns? 5 microns? etc), continue. Add components one at a time and check to see if one is the problem (remember, can be more than one so after finding a bad component, don't assume you are home free ... .) Of course, this requires you have proper caps for the components. old oil and similar grim could be an issue, too. If all components check, then consider possible contamination - water is a big one that just does not pump out and will keep even a tight system in the many of microns and act like a real leak. If that is suspected, use 70% alcohol, clean out each part (to remove organics like oil; this isn't necessary but might be worth the effort since everything is already apart), bake out (not the 0-rings!) at 100 C. Then reassemble, put under vacuum and see if that makes things better (one can using heating tape wrap on the system but most don't owe those.) Finally, one can use very high end vacuum grease, wipe on with a lint free cloth and wipe off with a lint free cloth. Reassemble and test. These are common methods.
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Positive Pressure Leak Checks

Post by Richard Hull »

Don't forget, a good high current glow cleaning will heat every part of the Fusion chamber and rid the system of water very well. Just make sure the grid is of a healthy construction (heavy gauge wire.) This demands a supply capable of about 50ma though yet need not be more than 5kv. A separate supply can be fashioned to easily do this. Many ion pump supplies can serve here. when I glow clean or run the fusor for fusion, my fusor's shell exceeds 100 C. That is why it improves markedly after the first run or two as it is warmed up for the HEAS each year. Water is, indeed, the big nasty.

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
User avatar
Jason C Wells
Posts: 113
Joined: Sun Oct 20, 2013 7:11 pm
Real name: Jason C Wells

Re: Positive Pressure Leak Checks

Post by Jason C Wells »

We are running afield of the original topic but I'll follow up here to close this out.

Part of my linear accelerator is very much like an ion pump, though I am using it to produce protons. I went back to the person who made it for me. At his shop is a Varian V-1304 controller for a 16 cell ion pump that has the current and pressure scales shown on the meter. Using some rough order of magnitude comparison of my ion pump to the setup in the shop, we determined that my baratrons are reading incorrectly. (They were contaminated by my aforementioned error. Tried to clean them but now I think they are still dirty.) My ion pump is drawing 40 microamps which should represent a pressure just a bit less than 10e-5 region Torr. This doesn't give me a precise pressure number, but it does tell me that I should be skeptical of my baratrons until I can get a good cal/zero on them.

Here's the meter we used to characterize my ion pumps current/pressure performance.
IMG_20150922_133300016.jpg
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Positive Pressure Leak Checks

Post by Richard Hull »

As I mentioned in another post, ion pump supplies are great for powering a fusor to get the water out as their high voltage and high current will do a great glow cleaning and carry the wall temp of a fusor up to 100C readily.

It is cool that you have used the ion pump to get an idea of your vacuum level. Unfortunately, ion pumps crud up fast at higher pressures than about 10e-4 torr and are usually only turned on in the 10e-5 to 10e-6 torr range to obtain 10e-7 or a little better, with a Ti sub pump finishing to a super vacuum level at or beyond 10e-8 0r -9. I do not even want to contemplate a struggle to obtain 10e-6 torr, much less those lofty super vaccum numbers. Once I drop into the 10e-4 torr range using my diff pump, I am as happy as a clam.

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
User avatar
Jason C Wells
Posts: 113
Joined: Sun Oct 20, 2013 7:11 pm
Real name: Jason C Wells

Re: Positive Pressure Leak Checks

Post by Jason C Wells »

I was warned about running my ion pump at high pressures. I don't turn it on until my no longer trustable baratron is reading microns.

My friend at the university glass shop found a calibration curve for a single cell ion pump in air. That curve puts me at 8e-6 Torr for an indicated current of 42 uA in air.

Today I am pulling 30 uA in hydrogen. Even if my pressure is incorrect by an order of magnitude, I am still go for launch. I'm waiting for darkness now. My alpha "detector" is zinc-sulfide with a thin foil screen to filter out keV protons.
An ion pump installed in a linear accelerator column. Also, a small cockroft-walton ladder multiplier, microammeter, and battery pack.
An ion pump installed in a linear accelerator column. Also, a small cockroft-walton ladder multiplier, microammeter, and battery pack.
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Positive Pressure Leak Checks

Post by Dennis P Brown »

Measuring your accelerator pressure via indirect methods just isn't a good idea; very inexpensive real ion gauges are available all the time on e-bay. Get one. These units are also great for detecting small leaks - when the "Dust Off" method is used, these gauges jump when a leaking section is sprayed. These units work from 10^-4 torr down to 10^-9 torr and are calibrated. Without a good handle on your real pressure you are not going to easily solve your issues. The reason I say this is that an accelerator just has major issues due to ion collisions/gas charging when operated in the 10^-5 torr range and you really need to know the actual tube pressure.
Post Reply

Return to “Vacuum Technology (& FAQs)”