Help with AC System Diagnosis

I'm trying to learn AC systems but am failing pretty miserably. The wife's '87 E30 has been retrofitted for R134a but the system was empty upon purchase. I pulled vacuum on it down to about 29 inHg and it held over an hour. Ran the vacuum pump for about 45 minutes and then attempted to charge the system. Here's where I ran into trouble.

The AC compressor engages and remains engaged. The high pressure side seems resistant to being pressurized. As I continue charging, the low pressure side keeps climbing and while the high pressure side climbs it does so very slowly if at all. When turning off the car the high side and low side equalize at about 120 psi pretty quickly. Final stabilized numbers after two small cans of R134a were:

• Ambient: 95F
• Relative Humidity: 35%
• Low side pressure: 60 psi
• High side pressure: 125 psi

Since the low pressure side is high, and the high pressure side is low, would this indicate a compressor issue? Could the dryer have some sort of blockage as it is on the high pressure side before the port? I'm trying to avoid throwing parts at this without a proper diagnosis. Any help is greatly appreciated!

Picture to try and keep this a little less boring:

I'd say either the compressor is weak or if this is an orifice tube system, the orifice tube is damaged and not restricting enough.  

Ditto on weak compressor or not-restricting-enough orifice tube. 

Also if it was empty when you bought it, there is definitely a leak. Add some dye to the charge and poke around at night with a UV light. Then you know which parts to shotgun at it.

rslifkin said:

I'd say either the compressor is weak or if this is an orifice tube system, the orifice tube is damaged and not restricting enough.  

It's an expansion valve system. Unfortunately it's under the dash and hard to get at so I can't quickly try swapping it out for another. From what I've read, wouldn't a bad expansion valve cause the pressures to not equalize quickly when the car is shut down? I'm not seeing that issue. 

maschinenbau said:

Ditto on weak compressor or not-restricting-enough orifice tube. 

Also if it was empty when you bought it, there is definitely a leak. Add some dye to the charge and poke around at night with a UV light. Then you know which parts to shotgun at it.

I did add some die in before charging. I'll have a look at it again but didn't see any leaks nor see a change in vacuum over time. The car had mostly been sitting around for 10 years or so when I bought it so it wouldn't surprise me if something had been disconnected at some point. 

My money's on the compressor.   But, expansion valve stuck open is another possibility, I think.

I thought expansion valves normally fail closed, which would cause high side to read too high and low side to read too low. But you're right, if it's stuck open it would cause what you see.

Vacuum sucks all the seals/orings the opposite direction they normally operate so it isn't a super definitive test, as I understand it the really good guys will pressurize the system with nitrogen for an initial leak check if they do anything like that.

As for the behavior, bad compressor (pull the outlet hose off and look for glitter) or bad TXV sounds right.

Alright I've learned two things:

1) Cosmoline (which happens to be all over the engine compartment in this particular E30) looks just like UV die under a UV light

2) I'm seeing a bit of UV die around the AC compressor. 

I think I did this the wrong way, but how does one add UV die to the system? If you add it to an empty system, it just gets evacuated by the vacuum pump. Add it to the manifold charge line and when you go to purge that line of air, you get a nice UV die spray over everything. Is there a better way?

I thought a lot of the recharge kits had UV dye in the R134a, so you'd get dye when you charged?

I only dropped in here because it seemed like a good thread for someone clueless on A/C to visit...

In reply to Ransom :

Yeah, they do. Problem is I couldn't find one locally with r134a and dye, without stop leak or other junk. 

BrokenYugo said:

Vacuum sucks all the seals/orings the opposite direction they normally operate so it isn't a super definitive test, as I understand it the really good guys will pressurize the system with nitrogen for an initial leak check if they do anything like that.

 

Hmmmm, I’ve never considered this.  

They sell a shot kit thing that will inject a shot of UV die and R134 when you momentarily apply it to the port.  Or buy a can with the stuff in it. 

But... That compressor's shot.  I buy them now from techchoiceparts.com.  Great guys.  Great prices.

In reply to Dr. Hess :

You weren't kidding...those prices look better than anything I've seen in a quick search. 

enginenerd said:

Final stabilized numbers after two small cans of R134a were:

• Ambient: 95F
• Relative Humidity: 35%
• Low side pressure: 60 psi
• High side pressure: 125 psi

Curious if this was at idle? or at an rpm spec'd in the manual? I think all the manuals I've used to service AC on specify an rpm at which to check pressures. And when I raise the rpm from idle, the low side drops and the high side rises as the rpm goes up. I just charged a fresh Miata system and an 87 Benz - both demonstrate the same characteristics. Do the pressures change as expected with rpm increase? Does the vent temp change at all with an increase in rpm?

I would guess it's the compressor. They have one-way check valves in all the cylinders which can leak, and of course piston rings although i think that's more rare to cause problems. Usually when the valves are leaking the compressor is also noisier. 

As far as vacuum leak checks, it is limited in the sense that you are relying on outside pressure (~14.7psi) to push into a ~1-2psi area so you're looking at only finding leaks that leak with a 10-12psi pressure differential. Compare that to a working system where pressure everywhere can reach 100+psi  with the compressor off and pressure on the high side can be 350+.  I think it is still useful for making sure there isn't a big leak before charging the system, it's just not going to find every type of leak that may be in the system. 

In reply to FE3tMX5 :

Good point...everything was at normal idle with the compressor engaged. I don't really have a good factory manual to go off of. The Bentley manual I have only addresses basic procedures with R12 but doesn't cover R134a conversion or charging. When I get a chance I will recheck things with something holding the throttle open a bit and see how things change.

In reply to Vigo :

Thanks, this is helpful. The compressor doesn't seem noisy...I think if anything it is just less efficient. 

The comments about pressure testing the system are really interesting to me. I've always used the vacuum method. So if you do pressure test with nitrogen or something, how do you actually go about finding the leaks? Could you even hook up a compression leak down tester?

In reply to freetors :

What I have been told here and have found out on Teh Interwebz is to hook up your air compressor to the refrigerant hose on your gauges and let it pressurize.  Once the system is under pressure you can check for leaks with soapy water (I use a 1:1 mix of carwash and water and it bubbles like kah-RAY-zee!),  You can even test to see if the pressure switch is working by turning on the system very briefly once it's holding air.  I've done this and it seems to work pretty well, but I haven't actually finished fixing the system yet -- still waiting on parts to see if the diagnosis process worked.

The biggest problem I've heard about this is that compressed air will have moisture in it.  But you're going to put the system under vacuum and replace the drier anyway, so....

I got a 1/4 flare to 1/4 NPT adapter to connect my can tap to the refrigerant hose and that same adapter will fit the typical air compressor quick-disconnect nipple.

In reply to enginenerd :

R134a is going show higher pressures vs. R12. I'm running the original R12 condenser in my 87 'Benz and it's inefficiency shows in the heat, direct sun and lack of movement. And I've got a big volvo/siemens 3000cfm fan pulling air through it. Vent temp will rise 10-15f if sitting at light in those conditions for extended time. Moving to a hydrocarbon/R134 blend improved the cooling. IMO the ideal solution would be to replace the old condenser with a new "universal" parallel flow condenser. I have one- just a project waiting to happen for something that operates sufficiently for my needs now.

In reply to wae :

I have used compressed air successfully to find leaks, but IMO it isn't as good as UV light and dye. I've had leaks too small to diagnose with compressed air. Here's a compressed air success video for you- tiny hole found.   

If you are getting a new compressor, don't forget to get a new receiver/dryer also.  They are kind of like the filter of the system, and should be changed any time you open up the system. It should be pretty cheap.