Don Carter

I've got a cylinder of CO2 from years ago when I used to brew my own ...
and I 'm wondering if there's any reason I couldn't use it to leak test AC systems. It may not be as good as Nitrogen, but would it do any harm if it's vacuumed out shortly after testing?

My regulator gauge only goes to 30psi, so it may not be able to output enough pressure, but might be worth a try.

blown 87

Why are you trying to re invent the wheel?
Just charge it and test it with freon if it holds vacuum.

How are you going to find a CO2 leak anyway?

Don Carter

Not trying to reinvent. Leak testing with N works very well and I think is fairly well accepted at many shops. We did it on my car in DFW with the help of John Curry and when we pressurized the system to 100psi, I could actually hear the leak near the compressor. Spraying some Simple Green around the area revealed the leak on a joint of the suction hose.

The vacuum test failed on my system last summer. It held vacuum but at the time had at least two leaks that caused me to lose all the R12 (not cheap) twice!

dr bob

You can use dried air if you want. Darn near anything so long as it has no moisture. Dry nitrogen is commonly used because it's cheap.

The vacuum test is a macro test for leaks and missing parts/connection, and isn't definitive. O-rings seal better under vacuum as the vacuum pulls a loose compression joint a little tighter, generally improvng the seal especially if the o-ring is properly lubricated. The maximum differential pressure under vacuum is a bit under 15 PSID, hardly representative to the 300+ PSIG that the system is can be exposed to. Pressure in the system, in contrast, will push a loose connection apart, and enable leak checking with an ultrasonic detector, or even the soap-and-bubbles method mentioned for macro leaks. FWIW, helium is the weapon of choice for process systems that need to be really tight. For large systems, sometimes a blend of helium and nitrogen. If your auto AC system is helium tight (and it won't everr be with the single-face mechanical compressor shaft seal) you can rest assured that no big freon molecules will EVER get out.

Don Carter

bob,

Thanks for the info. Just didn't want to use a gas that would absorb into the oil in the system or have some adverse reaction in the system.

dprantl

I'm not so sure about using CO2 gas as a leak detector for an A/C system. I seem to remember something about CO2 being able to react with certain refrigerant oils creating an acidic substance, but not sure about that though...

Dan
'91 928GT S/C 475hp/460lb.ft

blown 87

Most shops use freon to test, I know of no tester that is in common use at shops that would test for a CO2 leak, even my old freon leak detector will alert to a leak as small as 1/100 of a gram in a year, or so the maker of it says.

I guess what I am trying to say that charging with some other gas seems like taking the long way around a problem.

When you are paying some one 80-120 a hour it just makes no sense to me to waste a hour to save 30 bucks worth of freon.

Carry on.

dr bob

Greg makes a great point. You really need to test with a substance you can detect when there is an actual leak. No (read: functionally none) auto repair places will have ultrasonic leak detection capability, and few will have specific gas detection capability beyond freons. The guy I hire to do leak detection on closed process systems uses a pretty handy little detector for helium, one that looks amazingly lke my electronic freon detector. New detector head for every project, not cheap.

If you are really just doing a macro test for big leaks, use a dry gas and search for soap bubbles. When you get down to the tiny leaks, use freon if that's the detector you have.

The risk of making an acid in the oil comes when you add moisture. Both PAG and polyolester oils glom onto moisture, and moisture in your test gas will also contaminate the drier. Once the pressure testing is done, vacuum wll degass either oil, but you'll have trouble getting moisture out of either compressor oil at normal system temps, and for sure the dryer will hang onto anything it can. Best to keep itdry if you can.

Don Carter

Definately understand that testing with freon is the norm for most shops. I think the Nitrogen approach becomes appealing when you are talking about DIY mechanics dealing with expensive freons. The DIY'er probably doesn't have electronic leak detectors or R12 recovery equipment. Not much fun to put in $60 worth of R12 (@ Roger's cheap prices), find a leaky seal and realize that you have to release all the gas to fix the seal, then recharge with another $60 worth of freon and hope there's not another leak. Nitrogen is cheap, safe and dry.

When I took my leaky hose to Alamo AC here in Houston to have it rebuilt, I mentioned the Nitrogen testing and they said that was a great way to find leaks and they use the same technique on older R12 cars.

With the Nitrogen, you just spray soapy water to find leaks. Not good for leaks in areas that you can't see, but good for many common leaky areas.