Getting a seal on an AC connection

JHZR2

Staff member
Joined
Dec 14, 2002
Messages
52,855
Location
New Jersey
1991 Mercedes 350SD.

R12 system converted to 134a before I got it.

Has a slow leak. Noted oily residue at service ports and this one connection. Replaced R/D and all o-rings, and the service fittings. Didn’t sniff any leakages. Vacuum held properly.

Right side connection:
IMG_0694.jpeg


Had a 22mm weight grip wrench slip off when trying to tighten this. I can’t tell if it’s tight.

The o-ring was lubricated, nylog was used on all connections.

But I see oil leakage. Seepage

IMG_0700.jpeg


It’s definitely in the large threads. Not just on the pipe heading downward.

I had used a proper fitting 1 1/16 crowfoot flare wrench. I tightened it to the point that it wouldn’t tighten. Didn’t try to gorilla tighten it. But it was tight enough that a misplaced wrench slipped off a bit.

How tight should these fittings be torqued to? Just “wrist tight”, right?
 
Are you sure that it's not a defective area where that fitting for the service port is welded to the tube? Not second guessing you at all; I'd hate to see you damage something trying to seal it up.
 
Tightness is not the answer. The o-ring makes the seal. The ID and OD of the places the o-ring touches must be free of corrosion. The o-ring needs to have the proper section diameter.
I matched the size of the one coming out with a new green one that I put on.

This system seems to be a hybrid of metric and standard fittings. Maybe I got one that is “close” but metric or standard, and is a minuscule bit off?

I certainly haven’t overtightened it. I just went as far as it would go without resistance.

Are you sure that it's not a defective area where that fitting for the service port is welded to the tube? Not second guessing you at all; I'd hate to see you damage something trying to seal it up.
I’m not convinced that the service port adapters are all that great. They are the kind that requires removal of the OE schrader. Maybe that’s part of it. They were installed with rings lubricated and nylon on the threads. Those port adapters can’t be installed very tight as there’s no way to back the threads. But I have no doubt that there could be seepage. I checked with vacuum and sniffing after charging. No signs of leaks. Yet this residue is present.
 
Last edited:
What I've seen before is people using a small sharp pick to get the o ring out but the little gouge it makes is indeed enough to cause a small slow leakage. It's way worse with high pressure systems like hydraulic cylinders. A tiny scratch causes such a pressure loss from the bypass. OIl just goes right under the seals.
 
I’m not convinced that the service port adapters are all that great. They are the kind that requires removal of the OE schrader. Maybe that’s part of it. They were installed with rings lubricated and nylon on the threads. Those port adapters can’t be installed very tight as there’s no way to back the threads. But I have no doubt that there could be seepage. I checked with vacuum and sniffing after charging. No signs of leaks. Yet this residue is present.
I'd inject some UV dye and thoroughly clean the area with brake cleaner. Then you could inspect daily with a UV light to determine the actual location of the leak before disassembling any joints. If it is leaking by the threads of the port adapter, you will need to use teflon pipe sealant (or JB Weld) to get it to stop weeping.
 
IMG_4154.jpeg
IMG_4155.jpeg


1991 MB 350SD. Had a slow leak. Got it pulled down, thought I had fixed it, held vacuum, after a few weeks it wasn’t working. Because I don’t want to vent refrigerants, I went to HC for troubleshooting. Worked fine for a while, but then I noticed the first photo above.

I had even gotten the right flare wrench to tighten it well. Used the Santech green o ring for the fitting. Used nylog.

Same thing with the r134 fitting. Seeped a bit of oil. Tightened as much as I felt acceptable. The whole thing held vacuum and didn’t set off the sniffer.

So, I’ll have to take it apart, but where to start on making this right? I’m not really worried about sham retrofits. I may just go back to the original R-12 Schraders to at least take that out of the equation. But what to do with that one big coupling???
 
The paint distribution suggests that the nut is not on as far as it has gone before. This is often because the o-ring is the wrong size. On this type of joint, the nut should go to the same place whether or not there is an o-ring inside-- i.e test fit it without the o-ring. Do not have the pipe clamp in place while making up the joint. Removing the parts from the car for a close inspection would be a good idea. Look for any corrosion on the sealing surfaces, which are where the sides of the o-ring touch.

Those are the good kind of R-134a adapters, there is no reason not to continue to use them.
 
The paint distribution suggests that the nut is not on as far as it has gone before. This is often because the o-ring is the wrong size. On this type of joint, the nut should go to the same place whether or not there is an o-ring inside-- i.e test fit it without the o-ring. Do not have the pipe clamp in place while making up the joint. Removing the parts from the car for a close inspection would be a good idea. Look for any corrosion on the sealing surfaces, which are where the sides of the o-ring touch.

Those are the good kind of R-134a adapters, there is no reason not to continue to use them.
Yes, that did bug me. I used what I could tell to be the right o-ring from comparison. I bought a car specific set. But the paint thing did always bug me. That’s why I bought the right flare wrench and all. But no luck…
 
last year I replaced a lot of o rings on my 91 Toyota, which also was r12 switched over to r134. It was bone dry after 6 months, in other words successful. I torqued everything exactly to the workshop manual and cleaned all surfaces.

Have you tried the spray bottle and soap method when the AC is on? If that’s the high side, maybe those high level pressures would show some leakage that a vacuum test would not show. You may need to pull the parts apart and inspect them very closely.
 
  • Like
Reactions: wtd
last year I replaced a lot of o rings on my 91 Toyota, which also was r12 switched over to r134. It was bone dry after 6 months, in other words successful. I torqued everything exactly to the workshop manual and cleaned all surfaces.

Have you tried the spray bottle and soap method when the AC is on? If that’s the high side, maybe those high level pressures would show some leakage that a vacuum test would not show. You may need to pull the parts apart and inspect them very closely.
This is what I was going to suggest. Spray bottle with soapy water. Sometimes just because a system holds vacuum doesn't mean it will hold pressure. I've had that happen to me before.
 
last year I replaced a lot of o rings on my 91 Toyota, which also was r12 switched over to r134. It was bone dry after 6 months, in other words successful. I torqued everything exactly to the workshop manual and cleaned all surfaces.

Have you tried the spray bottle and soap method when the AC is on? If that’s the high side, maybe those high level pressures would show some leakage that a vacuum test would not show. You may need to pull the parts apart and inspect them very closely.
No leaks. I did try snoop, and a sniffer, which I wasn’t expecting much for since there was a lot of air moving.

Maybe nylog is insufficient as a lube?
 
No leaks. I did try snoop, and a sniffer, which I wasn’t expecting much for since there was a lot of air moving.

Maybe nylog is insufficient as a lube?
I used the pag oil that the ac system called for retrofitting to r134a. I coated the rings in that stuff and installed. I purchased nylog but didn’t end up using for some reason. Don’t remember why.
 
Back
Top