99 GTP air conditioner service port

[DrRadar]

198k. A/C isn't holding charge. The high pressure service port is leaking severely. Learned that the port is supposed to be replaceable. And I bought one at local parts store for $5. Tried to take out the existing one and can't get it to budge. Put open-end wrench on "nut" affixed to line for counter-torque and used a deep socket on air ratchet to try to loosen.

I'm reluctant to use any lubricant to try to loosen it because it will probably be bad if the lubricant got into the a/c system. Reluctant to use heat (torch) to loosen because of the refrigerant.

Any suggestions? Anyone ever done this?

Thanks,

[brcidd]

A 99 GP most likely leak location is the compressor belly- and shaft seal.

Believe it or not, but a good plastic cap with a good rubber element inside will seal any leaky high (or low) pressure port- I worked with the line designers- and the cap is the primary seal- the port seal is just there to hold the charge long enough to put the cap on- My sniffer has proven this too me several times.

[Tech II]

Sorry, Brcidd, I disagree....that valve is not suppose to leak......

That cap is no way the primary seal.....

[brcidd]

Hate to disagree and don't want to argue- I used to think like TechII as well-until I sat down with the hose and pipe designers for GM and was told by them that the caps are designed to hold high pressure- like it or not that is what they told me. I know no one believes it- just telling you the way it is designed.

Also, I test my sniffer with caps off at the ports- and they all leak some--Put the cap on and sniffer is happy- so I was satisfied that the cap is the primary seal.

We used to test systems- welded in pressure ports before and after every component- and never installed valve cores or valve ports- just let the caps seal when a port went uneeded for pressure taps- sealed just fine.

I can't tell you how many times, the cap has popped off in my hand as I unscrew it, due to the pressure it was holding back.

[DrRadar]

I've had valve leak some in the past and trusted that the cap would seal. And the pressure buildup behind the cap was noticeable when removing the cap. Leaky Schraeder valve cores can be replaced. This particular port/seal is not Schraeder (looks like a spring loaded ball) and is leaking severely. It has a "hex fitting" at the base of the valve as well as on the line, so I figured it could be replaced. Found the part, but can't get the blasted thing off.

I'm about to do some extensive maintenance under the hood, but I need to diagnose the a/c before I start pulling stuff off. I've considered replacing the entire a/c system (except the lines if they're too expensive). I have no qualms with replacing the compressor, condensor, and dryer. But I'm dreading replacing the evaporator. It seems like such a hassle. So if I knew the evaporator was good, I would do the stuff under the hood.

So my thought was to replace what I knew was leaking and see if that was good enough. Now I guess the best thing to do is to positively identify the leak location and go from there.

I've used a "torch" leak detector in the past on R-12 systems. Very simple, durable, reliable, and effective. But that apparently won't work with R-134a. I haven't had good luck with the electronic leak detectors in the past...often didn't get a confident feeling that it was working right, and they didn't seem to last long. I could put some dye in there, but that won't tell me if the evaporator is leaking.

So, any recommendations on an R-134a leak detector? or tips on finding the leak?

Or am I kidding myself that a 15 year old car with 198k miles may have an evaporator that doesn't leak?

[Tech II]

Well, we agree to disagree....maybe it's just words....but to me, a primary seal is where the system is initially sealed.....you don't have a secondary/backup seal before the primary.....

As far as detection, electronic sniffers differ vastly in performance, based on price......the two old fashion ways still work.....

Soap solution, and dye with a cheap blacklight.......the black lights are so compact, you can easily get into places the big old ones couldn't...

[brcidd]

I've done warranty reviews on W cars- they had evap problems up until about 1992- I can honestly say, I have not seen a leaky evap on a W car after 1994- the processing was changed for the better-

I reseal those V-5 compressors- replace the body o-rings and shaft seal and get another ten years out of them- the kits are available online for like $15.

[DrRadar]

I put a minimal charge in it yesterday...like 50psi not running. It dropped more than 10 psi overnight. New plan is to add dye and hope to find leak(s) under the hood. I don't doubt that compressor leaks, and maybe I can do the reseal myself. But the condensor takes a beating and would like to know if it's leaking. (Don't see any oil stains on it, although I don't have a clear view.) Then I'll assume evap is ok and charge on.

I'll take the high pressure line with valve out and attempt to replace valve off the car. If still too stubborn, then get a new line (if price isn't outrageous).

Alternative is to have my local shop locate the leak...maybe I'll get quote from them first.

Thanks for the feedback. It helps me make a plan.

[brcidd]

Originally Posted by DrRadar

I put a minimal charge in it yesterday...like 50psi not running. It dropped more than 10 psi overnight. Thanks for the feedback. .

Uhhhhmmm Errrrrhhh--- refrigerant is a condensable, not a compressable gas! Dropping 10 psi overnight is no big deal-pressure of refrigerant will drop with temperature drop- and rise with temp rise- so your test is no good unless of course you lost all refrigerant....but with this scenario, you should hear the leak..

[DrRadar]

Good point about the pressure change with temperature (pV = nRT). Trying the fluorescent dye. Let you know how it goes!

[DrRadar]

So I injected some 0.25oz PAG oil with dye and a few ounces of refrigerant...enough for the compressor to be doing a little work, but only about 30psi low, 150 psi high side. Ran it for a while then looked for leaks. I see a little dye at a seam in the center of the compressor...I guess this is what was referred to as the "belly"? And a few spatters on the back of the compressor beside the fittings, but neither place seems to have much dye. There seems to be more at the evaporator core fittings on the firewall; looks like a ring all the way around the low side fitting with more at the bottom. But that's right below the service port where the dye was injected. The fitting was stubborn going on and I know some dye was splattered; I don't think that's responsible for the ring of dye on the fitting, but it's making me wonder.

I'll continue to monitor. Maybe I'll clean off the existing dye and see if it shows up again.