So I haven't used my AC for a LONG time since it was mostly cold until about a month ago when started to get REALLY hot over here...

First my AC doesn't come on unless I have it 'On' for about ~30 minutes. Then I thought since my car is ~6 years old (00RS), I figure my refrigerant might be low and filled her up (it was a bit low). It was then EXTREMELY cold, which is good, but then it stopped working again the next day.

I later went back to the same shop that filled the refrigerant and asked them to check it out (luckily no charge for diagnostic). It seems to be some electrical issues, but not sure. Since they are not AC specialist, they couldn’t figure it out and the owner said he doesn’t want to replace the pump then find out it’s not the pump (expensive).

I stood next to the tech as he performed the diagnostic and noticed a few things…
- When turning the AC on/off through the climate control from inside, the pump does not engage.
- When driving power to the pump (one of the connectors on the pump), the clutch would engage, but high pressure does not move.
- Seems like hotter the outside temp is, the longer it takes for the AC to come on.

My speculation is that there is a bad ground somewhere… Any ideas?

 

If you can only get the compressor to run by connecting power to it directly, there is probably insufficient charge in the system. There is a low-pressure cutout that protects the compressor in the event of a drop in pressure, which is usually due to a loss of refrigerant.
I suggest you take it to a shop that does AC work and have it properly diagnosed and repaired. This is not a DIY unless you have the appropriate knowledge and tools which you apparently do not, and neither does the first shop.

 

I would love to take to an AC shop that is not going to rip me off. I know I am not going to do the work myself, but I don't want someone to tell me to replace this and that even when they don't have to be. You know what I am saying? I would like to find out what is wrong then tell the shop to fix just that and nothing more. thanks...

 

Well it could be anything from a simple leak to a defective component such as the expansion valve. Unfortunately you do not have the diagnostic tools and AC experience to determine the cause, so you are going to have to leave this to someone else.
The best thing to do would be to post in your regional forum (try NASIOC as well) and ask for a recommendation of a good local shop that people trust.

 

hmm.... was hoping something experienced something similar in the past... I'll try my local forum. Thanks...

 

Even if others have had problems with similar symptoms, you cannot assume that yours has the same cause. You could waste a lot of money and shop time by doing so. It's best to let a skilled professional diagnose the problem.

 

Downloaded everything from http://techinfo.subaru.com/ and follow their diagnostic flow. And guess what? It's electrical...

A lot of probing points in the diagnostic flow are in the main fues box and it's hard to get to. So I just read everything related to AC and for my problem, basically I need at least 10.5v going into the compressor coil, which I have ZERO!

I then trace backward...
Compressor Coil->AC Relay->Power Supply = Good lines but AC Relay not active
So I trace the line to activate the AC Relay...
Power Supply->Thermal Protector->Pressure Switch->AC Relay = I saw strong voltage going into the Pressure switch, but nothing into AC Relay.

I've been doing this during breaks and lunch. I have few more things to try but I am pretty sure is the pressure switch.

Now to be 100% sure would be to check for continuity from Pressure Switch to AC Relay. I can either short the in/out lines at the pressure switch and should observe Compressor engaging, or I can hook up the multimeter again at pressure switch and AC relay and observe resistant.

few more things and I'll be a happy man again!

 

But the thing is, it may not be electrical if you really have low pressure in the system, in that case the pressure switch is just doing its job. The only way to troubleshoot it properly is with a set of manifold gauges to see if the pressure is correct or not.

 

that has been done LONG ago... The shop just couldn't figure out why (I think it more like they didn't want to since they are no AC specialist) the compressor wouldn't come on by itself. If they force it running, everything is fine. I have no leak and pressure is good, as long as the compressor comes on.

After studying the entire system. AC is really not as complicated as some might think. The wiring is simple, basically just thermal, pressure, and ECM control/protection. As for the lines, really just compress, filter, pressure regulator (in our case, a tiny dual pressure switch), and expan.

Wiring diagnostic was easy for me and I am just a CE not EE. If the component requires voltage/current, you should be able to isolate it with a simple multimeter. Assuming you know how to use it and understand switch and relay. But things like expension valve or the evaporator, leak, etc, you'll really need a set of gauges at the high and low pressure valves. I find the biggest obstical is to discharge the system, everything else can be done yourself with limited hand tools...

 

So I completed the two tests I had to do to confirm the Dual Pressure Switch is at fault. I checked for continuety of the line from the switch to relay and of course it's good. And took the line from the thermal switch and connect it straight to the line going into the AC relay, and the compressor came on immediately!

Here is the kicker... I cleaned the wiring plug of the Dual Pressure Switch at the Reciever Drier and guess what? The Compressor comes on normally now. But that doesn't eliminate the faulty Dual Pressure Switch, it could be just on it's edge of going out. So I thought of two more things to try tomorrow.

Probe at the AC Relay and see how much current and voltage is coming in from the Dual Pressure Switch. Maybe I am getting a current/voltage drop across the line (don't know how since the resistant is only 1 Ohm and the wiring diagram doesn't show a potential voltage divider) and barely/not always flipping on the AC Relay.

I should also check the voltage at the Compressor Coil to make sure the voltage is well over the required 10.5v

 

Alright, so I checked the voltage at the AC Relay coming from the Dual Pressure Switch. I got ~14v and ~5amp. I'll go check the Compressor coil during lunch, I was rushing to work this morning when I checked the relay.

If they are both good, I think I'll just wait until it goes dead again... I mean, if the system is running perfect, there is nothing anybody can do (nothing to isolate).

 

Thanks for the update. Even if I never attempt to diagnose an a/c system problem myself, it's interesting to note the steps involved.

--
0==WW==0
"…axles of evil…" - george w. bush

 

You are welcome. I don't know if I would ever replace the components myself, but I definitly would like to know for sure myself, why and what failed. After all, no one would like to get ripped by a shop right.

absorbing knowledge is never a waste of time i say...

 

So here is another update... Of course, my luck has it, it went dead... I was just about to order the dual pressure switch online and didn't feel right. So I went back out to do more testing.

So it's not just a dirty plug. Went through all the excercise again, but this time I got some help. I needed to know if my high pressure line is in the operational range (26-299PSI) and it was. I was then able to check the continuety at the dual pressure switch. Guess what, it's good! So I shorted the line from the Thermal Protector to the AC Relay again, and what do you know, no 12v going into the relay even the thermal protecotor line and the switch to relay lines are good. That means I have no ground! Went back to the wiring diagram and it point to ECU, where it needs to grounded, which is controled by the Thermal Control Amplifier at the evaporator core.

Thermal Control Amplifier: Need 12v coming from the Thermal Protector, into the Amp, where it has a diode to control current into ECU. AND for that to work, it needs a ground, which is coming from the AC button at the console.

Long story short, I hack up some wires and by passed the Thermo Control Amp and the AC system comes on every single time, on demand. I then plug the wires back up and opened the Amp. I probed around and noticed a ~10v drop into the Amp, therefore, I only get ~3v into the ECU.

FYI (00RS):
Receiver Driver + Dual Pressure Switch
73410FA131 Liquid Tank
73039AA010 Pt181000 O-ring

Expansion Valve
73531FC012 Expansion Valve
73561FA030 O-ring
73561FA040 O-ring

Thermo Control Amplifie
73540FC021 Thermo Control Amplifie