Need some help troubleshooting an A/C issue. The heater and evaporator cores were replaced in January due to a leaking heater core. Now that it warmed up outside, I've noticed an issue with the A/C. Driving around town, I would occasionally notice that the air from the vents would start warming up as if I had switched it over to vent. The volume of air wouldn't change, just the temperature. This would rarely happen and my trips were usually short, so I didn't try to recreate it. A couple of weeks back, I took a 4-hour trip. Drove down during the evening with temps at 75-80 outside and the A/C did OK. Had it set on a few notches above the coldest setting and it maintained a decent temp. On the way home a day later, I was driving in the afternoon with temps outside around 90. I would drive for 15-20 minutes and it was around 55* at the vents, then it would start blowing warm. If I didn't change anything, the temp would slowly climb up to 80-85 and stay there. If I switched over to vent and let it run for 3-5 minutes when I switched back to A/C it would blow cold again. This cycle repeated for the entire drive home. I took it back to the shop that did the evap core. They pulled the dash and found some ducting that wasn't connected up right and did a recover/recharge and said all the pressures were good. Drove it home and it was cooling well. Made that same 4-hour trip a day later, and it started behaving the same way again. Driving at night, with an outside temp between 80-85*. Temps would be cool, then start blowing warm like the A/C was off. If I switched to vent, then back to A/C a few minutes later, it would cool again. Toward the end of the trip, I would notice that it sometimes would coincide with when I would slow down (highway drive passing through small towns). In some cases, I would feel it start to blow warm when I'd be slowing down, then if I started accelerating it would blow cold again. It would still blow warm while at highway speed with no change in engine RPM, but it seems to happen more frequently when slowing down.

Any thoughts on what might cause this behavior. It's in the low 70s outside right now, so it's working great and I can't recreate it. Probably won't go back to the shop until it warms up and I can recreate the problem easier. But I wanted to get an idea of what we could be looking at.

It really feels like the A/C just switches off and runs on vent more, but the controls are on A/C. Same problem happens on regular and max (recirc). The fact that switching to vent, then back resolves it for a bit makes me thing something is freezing up or something is overheating and cutting out. I can't say for certain whether the A/C clutch is engaged or not when it happens. Has anyone found any creative ways (light, etc) to know when the A/C clutch is locked in? I'd pop the hood and check, but it's not so easy at 75mph. Open to any other ideas of things I can check while I'm waiting on it to warm up outside again.

 

If you have a scan tool, you may be able to see if the compressor is being commanded on - I believe that's PCM-controlled on 24-valves, but I'm not positive (I have a 12-valve, and much is different). You should check a wiring diagram or service manual theory-of-operation section to be sure. If it isn't PCM-commanded or you don't have a scan tool, then you could make jumper wires from the compressor connector to a high-impedance (digital) multimeter or something like an LED test light and look for voltage changes. As an alternative to driving, consider something to hold the throttle at 1500 RPM or so with the vehicle parked so you simulate road speed. Google-search "throttle pedal depressor" or "throttle pedal holder" for commercially-available versions, or make your own. Just be aware that the airflow over the condenser core probably won't be the same as if driving - which brings up the question of the condition of your fan clutch.

The A/C system has pressure limit switches. If pressure goes too low (suction/low side) or too high (discharge/high side), the compressor will cut off until pressures get back in range. This will happen regardless of the regular/max setting - the compressor and related components don't really know/care about the air-intake door setting. If the system is leaking refrigerant or there is compressor debris in the system blocking the orifice tube, the cycling effect will become more frequent (especially if refrigerant is leaking). It is possible for the evap core to freeze over. I've never experienced that, but I would expect reduced air flow from the vents in that case (the frozen-over fins would block air flow) - if I'm wrong on this, someone please correct me. An evap core with fins blocked by debris will also cause problems - since the core was replaced, this is presumably not an issue AC gauge readings would be very helpful - to see what the low-side/high-side pressures are in relation to each other and ambient air temperature.

Why did the shop recover/recharge refrigrent when reconnecting ducting? What ducting was not properly connected? I hope they replaced whatever compressor oil was lost during the recovery process.

 

Thanks! I had considered holding the RPM steady, but figured this would introduce a new variable of not having air moving over the condenser coil. I could set up an external fan, but it still changes the test conditions to a degree.

I'll have to do some reading in the FSM. From what I understood, all the HVAC controls on the dash just triggered the doors on the HVAC housing to move. The mode switch is vacuum-driven, and the blend door is electric. The system doesn't care about interior temp or output temp, it just ran the compressor in A/C or defrost to cool the evap coil and the temperature was adjusted by plumbing. That's why I was trying to deduce whether the compressor was cutting out because it was asked to (pressure out of range) or something failed on the electrical side that unlocked it. I would assume that freezing in a line would trigger the pressure cutout, which is why running for a bit on vent would allow it to thaw. However, the odd part was that if I left it on A/C, it would never return to blowing cold. This is what was making me think that maybe it was on the electrical side that overheated and needed cool-down time before it would re-engage. I don't believe the evap core is freezing over since air volume doesn't change. And I wouldn't expect blocked fins since the core is new and I have filters over the air intake to keep out debris. Orifice tube getting blocked/frozen was on my list, but then I wondered why it would never cycle back to cold on its own if the compressor disengaged when the pressures went out of line.

I'm not sure why they did an evac/recharge. They may have pulled the HVAC housing out, I'll have to ask. When I talked to him yesterday, he just read in the notes that it was done, which should mean the system is properly charged. I'll have to ask about that.

 

It is possible for the orifice tube to freeze. You didn't mention in your first post if the accumulator/drier was replaced at the same time as the evap core. If not, and/or if the shop wasn't very diligent about vacuum-pumping the system, there might have been sufficient moisture left to freeze under operation. As to how long it would need to melt, I don't know.

If by "plumbing" you mean the blend door, then I think you're fundamentally correct about how the system operates and the system not caring about temperatures. Although - I see an ambient air temperature sensor and a refrigerant temperature sensor (described as a coolant temp sensor when the listing is expanded) on RockAuto under the AC/Heating section for your year. What (if anything) does the FSM say about their roles? And is there anything in the FSM about troubleshooting the climate control module? The module looks pretty simplistic - not that much different from the one on my '96

 

The control looks pretty simplistic. The PCM controls the compressor clutch, which is based on two inputs: 1) whether the dash control is in a position that uses A/C, 2) whether the high and low pressure switches are in range. The high and low side switches are in series, so if either one opens, the compressor dis-engages. If the controls are in A/C and the pressures are right, then the compressor should be engaged. I don't see any temperature sensors in the A/C system, so the only thing that would control the clutch cycling would be the pressure sensors. I believe the only ambient sensor is the one located behind the grill and it feeds the overhead console temp display. I didn't see anything for refrigerant temp, only coolant temp. And this doesn't factor into the HVAC system from what I see in the FSM.

I'll have to check if they did the accumulator or not with the evap core. Moisture freezing in the line makes sense. The only part that is puzzling me is why I actually have to go back to vent to restore A/C operation. I would expect that when the line freezes, the compressor would disengage, the line should thaw, the pressures should return to normal values, then the compressor should kick back in. But when I've left it in A/C, it seems like it never comes back to cold. When I go to vent, it would keep the compressor from running for the entire period which means it should fully thaw. This is where I'm just guessing without knowing the specific inputs the PCM is seeing. I had considered ways to trigger an LED from those sensors, but I really don't want to go hacking up wiring to set this up. And I haven't found any scan tool that will tell me the status of those inputs on the PCM.

The FSM has plenty of troubleshooting steps using the manifold gauge set and checking sensor continuity, but that's only useful if it's acting up when stationary. So far, everything is great when I'm not moving so I need to be able to see that data while in motion. Or I throw more parts at it (not my preferred choice).

 

Anything new?

I guess RockAuto had to find a category for that ambient air-temp sensor, so AC/Heating made sense to them...

I wonder - if the o-tube is freezing - if it thaws and the compressor come back on and the o-tube quickly re-freezes, giving the appearance that the A/C never comes back on? The test light idea that you mentioned in your last post would be a way to test that. You could easily make a homemade test light, or extend a pre-made one with wire and alligator clips. Given the clutch connector style, I don't think you can just push the wiring for the light into the connector as you could if it was more open connector like on a GM R4 compressor. I know you don't want to hack up the wiring (I agree) - if you back-probed the compressor connector on whichever wire does not travel through the pressure switches and back-probed the other wire at the output side of the closest pressure switch in series to the compressor, then you could eliminate the possibility of the probes shorting together and you would "only" have to keep them from loosing contact due to road vibration. Straight sewing pins make decent and cheap backprobes (that's basically what the replaceable pins are in my AES backprobe kit) - clip onto each pin/probe with an alligator clip and wrap the wire and "probe assembly" with electrical tape to prevent losing contact due to vibration. If you already have test leads with 4mm banana connectors on each end, then you can get more elegant and use commercially available backprobes.

 

Finally had a chance to get back to the shop now that it's warmed up again. During cooler weather, it had no issue or was really hard to recreate. They checked the levels again and everything was right now. While idling with the gauges hooked up, the pressures were right where they were supposed to be and the temps were good. Their tech took it for a long drive and was able to recreate the issue. They dug into it and found that the A/C compressor clutch had too much gap. I believe he said the gap was 1/8". So the clutch would engage, but it wasn't holding well. They adjusted the shims, took two more long drives in the heat, and said it didn't kick out. I'll run it today and it was good along a route that would usually fail on the way home. I'll run it the rest of this week to see if it's actually fixed.

 

Good! I wonder if they were able to re-shim the clutch without removing the compressor, or if they evac'ed the system and pulled the compressor to re-shim the clutch. Doing it in-place strikes me as a very not-fun job - but then I don't have a lift, and it's almost always worse when you're working on your back.

 

He said it was hard to get a picture of the clutch for the documentation, so I'm guessing it was done while on the truck. I had to run errands yesterday afternoon when it was 100-102 out, it kept steady at around 50* on the center vent the whole time. I'm just glad they stayed after it to find a problem instead of throwing parts at it and running up a bill. Even for all the diagnosis time they had into it, he only charged me an hour of shop time.

 

Sounds like you've got a conscientious shop there! Was that 50* reading with it in recirculate mode?

 

Yeah, 50* was on recirc with the fan in position 3. If I kick over to outside air, it sits right at the 60* mark.