legacycentral bbs

The patient: 1993 Legacy L 2.2L 5MT. 152,000 miles.

When the A/C is engaged there is a noticeable hesitation/stuttering under throttle all through the rev band. It's rather intermittent. Cabin air goes warm when the hesitation occurs then gets cold again shortly after it goes away.

Idle also varies from 450RPM up to 1100RPM or so at idle, in neutral, with or without brakes when A/C is engaged.

A/C gets quite cold, eventually, when moving even with hesitation/wandering idle.

Idle without A/C engaged is rock-solid at 600RPM or so and power delivery is strong through the revs. No idle issues at all without A/C.

Hesitation only occurs when engine is at operating temp. When engine is cold and A/C is engaged, idle jumps to around 800RPM and sits there until the engine warms up, then the idle wanders and the hesitation is experienced.

I will clean the IACV but I'm not sure it's the culprit as the thing idles fine with the A/C off. It was almost acting like the A/C belt was slipping, but there is no evidence of it and the belt is perfectly tight.

Still has the factory (recalled) knock sensor, which is something else I was going to check.

Any other suggestions? Would a low A/C charge cause these symptoms?

Thanks!

--Andy

I have similar issues. Bump.

The AC system may be overcharged/undercharged. The expansion valve may be freezing up if there's moisture in there, and then causing the high side compressor pressures to go to high, and the compressor to shut down.

Does the air get warm at all during idle, or only when driving? You really need to get a set of AC gauges on the system to determine what's going on. I'd suggest taking it to a shop that will do a free AC inspection. Or if you want, get a set of AC gauges. The less expensive ones are probably around $40-$60.

Did some more digging today.

Cleaned the IACV with Seafoam and went for a test drive. Still had lopey idle and surging under power. Noticed at a stop light the lights were dimming with the changes in revs.

Turned the A/C AND blower off. Idle stabilized.

Turned JUST the blower on HIGH without the A/C. Idle wandered.

I have a bad alternator.

--Andy

Yes + a/c clutch requires power.

ericem wrote:Yes + a/c clutch requires power.

Yep, and the fans that switch on with the A/C compressor too. So that's why it's worse with the A/C but still does it with the blower on high.

Will post back if a new alt fixes the problem.

--Andy

I have not replaced the alternator yet as I do not feel that is the cause. I had it tested at Autozone and it passed with no problem.

The problem is still occurring. It's a definite stumbling through the power band coupled with warmish air - only when the A/C is on.

I have noticed it eventually goes away, about 20-25 minutes into a drive. Then the A/C blows super cold, there is no stumbling, and the idle is fine for a while, until I slow down from highway speed. Then it starts up again.

I think doing a solid redline acceleration run will skip a lot of the stumbling and get it to operate normally.

I'm almost thinking it has a low refrigerant charge and/or a failing expansion valve. I am going to track down an R12 manifold and see what I come up with.

Would a low charge cause the engine to stumble like it is? I have never seen that with an R134a system, but I've never had an R12 system before. Perhaps the stumbling is the pressure switch turning on and off rapidly?

--Andy

might be the battery....

zzGUY46 wrote:might be the battery....

Eeh, I had the alternator and battery load tested and both passed. I made sure I had my stereo, A/C, and headlights on during the test too.

As part of the tune up I replaced spark plugs, plug wires, PCV valve, air filter, oil, oil filter, timing belt, and a VERY broken knock sensor. I cleaned the MAF and IACV and load-tested the battery and alternator. The motor runs significantly better after the tune up when the A/C is off... I'm attributing that to plugs and the knock sensor.

Still planning on checking the refrigerant level. Any other suggestions?

--Andy

O2 sensor?

I *think* I may have fixed the problem... accidentally.

My money at the moment is on an overcharge. Good advice Josh

I was cruising around today, experiencing the same symptoms. I dropped by a friend's shop and he let me borrow his (what he thought were) R12 gauges to see what my A/C system was doing. The low side fitting attached just fine and read around 47 psi... a bit high since it was only 80F out or so.

The high side fitting he had wouldn't fit. I found this out by discharging a 3-4 second long blast of refrigerant and oil, completely accidentally.

I gave up and decided to wait until another time when I had a gauge set that I knew would fit. I noticed throughout the rest of the day the problem was gone.

Took two more trips later in the day and had no problem what so ever. Air is cold, idle is stable, no stumbling. I think an overcharge may have been causing the compressor to resist being rotated, hence what I felt as stumbling. With only 130ish crank HP (new), I'm sure this amount of resistance could have been felt through the drivetrain.

I will report back tomorrow when I'll hopefully have a correct set of gauges, but I hope this fixes it.

--Andy

I would tend to say people tend to overcharge AC systems because it's hard to judge how much you have in there, and how much may have escaped. When you have too much it will cause a larger load on the engine. Plus the compressors in the older cars had a variable geometry compressor or wobble compressor. So they typically do not cycle.

Let us know if that fixed the problem.

OK, I finally got the proper high side adapter. These are my findings.

Ambient temp: 86 F
Vent Temp: 51 F

Low Side: 32 psi

High Side: 180 psi

I haven't done much digging in my FSM, but it's looking like an undercharge. Tech at the dealership said the low side PSI should be approximately the temperature coming out of the vent. This isn't even close.

--Andy

So I found the performance charts for the Zexel compressor in my FSM. This is what I found>

At 86F:

Compressor Suction Pressure
Acceptable Range: 17-29psi (approx)
My reading: 30psi

Inlet/Outlet Air Temperature Difference
Acceptable Range: 65-72 degrees F (approx)
My reading: 35 degrees F

Compressor Discharge Pressure
Acceptable Range: 150-215 psi (approx)
My reading: 180 psi

Are those readings with the AC on AC MAX and vent fan on 4?

Those readings look alright. The vent temperature should be a little less. The differential temperature thing is a little misleading IMO. You are not likely to get below a vent temperature of lower 40's upper 30's with that ambient temp.

The last set of readings were taken at the end of a day of driving when the system was behaving normally. The controls were set to MAX A/C and the blower was on 4, with the window down.

However, the readings were taken at idle (around 850RPM). The figures in the FSM assume 1500RPM.

So I had a chance today to take some readings when the A/C system hadn't been run in about three days.

I noticed from the get-go that the compressor was cycling, even in MAX A/C.

With the system off, the low side read 110 and the high side read 115.

When first turning the A/C on to MAX, the compressor did not engage. I had to first select regular A/C, then move to MAX.

Ambient is 80F.

Readings at idle were 40psi/190psi 68F.

A few minutes in the readings were 35psi/200psi 61F.

When I took the car up to 1500RPM, the readings were 25psi/225psi and 55F.

Back down to idle, 32psi/175psi 63F. This is when the compressor started cycling again (even in MAX A/C). This exactly replicated my "hunting" idle problem, and I assume is also causing the surging under speed.

I ended at 1500RPM with readings at 30psi/225psi and 54F.

I also got a video of the sight glass while the system was running:

http://www.youtube.com/watch?v=jxWP8ZwNElE

The FSM leads me to believe that this is a low refrigerant issue. From what I can gather the Zexel system doesn't cycle, even in MAX A/C.

I *should* pull all the refrigerant out of the system, replace the drier, and go R134a, but I don't want to be bothered this far towards the end of summer. I'm going to throw a half a can of R12 in it tomorrow and see what that does.

--Andy

Yeah, sounds like there's a charge issue. Just add a little bit of R12. You don't want to overcharge it. Add a little, and keep checking vent temps, pressures, and if the compressor changes time between cycling or stops cycling at all.

Do you hear hissing coming from the expansion valve under the dash?

OK, here's an update.

I was unable to secure any R12. So, I evacuated the system, replaced the drier, pulled a strong vacuum for about 10 minutes, and charged with 134a.

The system is perfectly charged according to the dummy gauge on the low side. The problem still persists. Now that I have the glove box removed I can hear rapid (some times 5-6 times a second) clicking coming from the cut off relay.

SIDE NOTE: When the system IS working, it is A LOT colder than it was before I charged to R12. The lowest temp I could get out of the vents with the R12 charge was about 55F. I was seeing 39F-41F now that it's been converted.

Now I start troubleshooting electronics. From what I can see from the wiring diagram, there are several items that control the compressor clutch:

- The A/C switch
- The pressure switch in the drier
- The cut off relay above the evaporator box
- The thermostatic amp on the evaporator core
- Any one of four relays in the fuse box under the hood.

I'm assuming the A/C switch is OK.

I bypassed the cut off relay above the evap box and the problem continued, so I'll assume that's ok (even though it measured 80 ohms instead of 120 like the FSM suggested).

I was unable to find a replacement relay for the fuse box, so I wired a generic one up with some leads and I have been rotating it through the four positions under the hood. The way it looks like the system is wired, the switched leads for any one relay are also the signal leads for any number of the other relays. So a bad relay anywhere in the system would cause the symptoms I'm experiencing.

I have swapped the "new" relay in for #1 and #4 and the problem persists. I'm hoping it's in #2 or #3, but I'm not expecting miracles.

Next on my list is the thermostatic amp, but this is an $81 part that I am not interested in replacing needlessly.

Last would be the pressure sensor in the drier. I would also hate to replace this needlessly since I would have to discharge the system. It also seems from the parts book that the sensor is not available separately, only with another drier. This is $95 from the dealer instead of the $17 for just the drier that I already paid from NAPA.

This is a simple system but something is most definitely not right with it. I hate throwing parts at a problem. Any suggestions are welcomed.

--Andy

Andy,

Do you have any pressure readings now that you've done the 134a conversion?

Does the cycling only happen while driving, or while idling as well?

I have a reading on the low side from the gauge that came with the charging kit. It shows about 40-45psi (ambient temp was about 90F today). The gauges I have currently are for R12 and I added the 134a fittings, so they won't work. I may pick up a set of 134 gauges for myself tomorrow, since I'm sure you're going to want an idea of the high side pressure.

The cutting occurs at idle and at speed, still.

A thought just occurred to me. A faulty clutch would probably break the circuit too, which would explain why every relay in the circuit would be energizing and de-energizing. It would probably appear as an electrical problem too, which it is.

I will try running 12V directly to the clutch tonight and see what happens.

--Andy

I got the 143a gauges on. When the system is running it's at 45psi on the low side and 250psi on the high side - almost identical to what my 2008 runs.

The problem is not any of the relays under the hood.

I want to figure out a way to test the pressure switch, but I'm having a hard time reading the wiring diagram I have.

The pressure switch shows two switches inside of it, one normally closed and the other normally open.

The wires on the normally closed switch are brown/yellow and brown/red.

The wires on the normally open switch are red/lime and black. Black goes to ground.

I can't make heads or tails of this I don't want to screw anything up any worse than it already is. Since BrY and BrR are normally closed, I think may just jump RL and B and see what happens.

Any input appreciated, as always.

--Andy

Since Josh's FSM scans didn't seem to include the A/C section, I scanned the two pages with the wiring diagram.

http://picasaweb.google.com/lh/photo/gs ... directlink

http://picasaweb.google.com/lh/photo/Xg ... directlink

I've tested the "Air Conditioner Cut Relay" and all four of the relays in the relay holder.

I'm trying to figure out how to test the "Pressure Switch" and "Evaporator thermo-switch". Any suggestions welcomed.

--Andy

The pressure switch, assuming it is the one on the reciever drier, it's actually a trinary switch. There's actually three settings. The diagrams you posted I went over them a couple years back....because like you said, they are rather confusing. There's some other diagrams that talk about the trinary switch.

Unfortunately I won't have access to my manuals until Thursday to review them.

The evaporator thermo switch only cuts out the AC if the temperature reading on the evaporator is around 32 deg F. I'd probably focus on the pressure switch, but without testing/back probing the circuit it'd be hard to completely rule it out.

Finally think I found the source of the problem. The thermostatic amplifier.

The harness side of the amp amp has three wires - a 12v source, a signal to the A/C relay under the hood, and a ground. The actual "brain" of the amplifier is housed in the connector from what I can see. There is a diode that sits on/in the evaporator that measures the temp and controls the amplifier. If I jump the signal and ground, the system works perfectly. Naturally it doesn't deice and gets supernaturally cold - like 33F with fog coming from the vents. But it works.

I will try to find another thermostat in the junkyard this week or next. It's an $80 part new from the dealer and that's hard to swallow at the moment.

Josh - I found the pages in the FSM talking about the pressure switch. It has four wires. Two control the high/low cutoff, two control the "sub" fan (I believe the driver's side fan). The high/low cutoff is just that - off or on. The fan side of the circuit seems like it provides 5v or 12v, depending on the pressure in the system. This provides the high/low feature on the fans.

So, not that confusing once you're able to wrap your head around everything.

Now I just have to figure out A) how to get the evaporator box off without removing the evaporator and B) if it's even possible to replace the amplifier without removing the evaporator.

EDIT: For the record, this is covered in the following Endwrench article:

http://www.endwrench.com/pdf/heat/FtACTechTipsSp00.pdf

--Andy

Andy,

Based on this diagram
http://picasaweb.google.com/lh/photo/Xg ... directlink

What are you calling 12v source, signal, & ground. Which pins on the B71 connector?

You will have to discharge the AC system in order to get the evap box out.