Rep Power: 0

I hope it's not beads! Crazy. But there's another interesting discovery. When I take the evap temp sensor out but leave its connector plugged in, and dip it in ice water, the compressor shuts off almost immediately! When I take it out of the ice water, it kicks back on in a few seconds. Otherwise, with the sensor installed, the clutch will just go endlessly without turning off most the time. I wonder if the sensor just isn't making proper contact with the evap core, thus not signaling the compressor to turn off.... thus letting the evap core freeze up and stop flow...

Thoughts?

 
Rep Power: 518

I'd say you're damn close to the issue here. .......but if the evaporator core is freezing you will usually notice a significant decrease in airflow from the vents as ice builds up in the fins of the core. Did you?

Temp sensor might be skewed a little.
Ideally you want the compressor to shut off at about 37F (core temp) and turn on a couple degrees higher and it sounds like the sensor may not be quite accurate now, or the HVAC controller isn't reading it quite right.


Check sensor resistance ....at 32F the graph shows resistance should be about 30k ohms, and resistance drops as temp increases.

Back when Radio Shaft was in business and actually selling electronic components you could get a bunch of resistors and alter the temp sensor output.....

 
Rep Power: 0

I didn't notice any decrease in airflow, but I guess I wasn't looking for it either. Is it safe to say that if the pipes coming out of the expansion valve (on low side) are freezing, that the core is also in similar shape?

Most the time the compressor is stuck on it seems, but after driving home (last time I drove) on the highway for almost 2 hours, I suddenly noticed it was cycling again, about every 40 seconds I'd hear it click and the pressures hiss.

I did check the resistance and it seemed to be pretty damn close to the graph in the manual. got the ice water down to about 40 deg. and the resistance was at 23-24 k ohms

 
Rep Power: 518

You mean the suction line from firewall to compressor? Pretty much. I expect the line to sweat, but not freeze.


IDK... this makes me wonder if the issue is a flaky control panel.
If I had a bunch of FREE known good parts to toss at it....

 
Rep Power: 0

Yeah, suction line from the firewall. Super frosty.

Control panel? Is that the ECM? Those go bad often?

 
Rep Power: 518

The HVAC control panel has the 3 dials attached to it, in the dash next to the radio.
NOT the 3 buttons below.

Yes they go bad sometimes but usually they just won't turn the AC on. I haven't personally seen one fail with symptoms like you describe.....it's just a hunch.

The panel actually has a self diagnostic function......not sure it will help but it's worth a shot....check this post for instructions

https://www.civicforums.com/forums/36-mechanical-problems-vehicle-issues-fix-forum/349756-2002-honda-civic-ex-ac-problem.html#post4607711

 
Rep Power: 0

Cool, I'll give that a go. I have messed all around there while doing stereo stuff.

One thing I forgot to mention too is that when I had the evap sensor out of its installed position, I took a 10" rod (3x longer than the sensor) and stuck it in there to see if it would hit anything, and even with the whole thing shoved in there it didn't hit. Is that right? The guy at the shop said that the sensor should actually come into contact with the evap core to get a proper reading. I can angle the rod toward the front of the car and I hit what feels like the metal core.

 
Rep Power: 518

On many designs this is true, in fact some actually insert the temp probe deep into the middle of the core.........

But not on this particular model.

It is designed the way it is designed and it worked just fine until it broke.

 
Rep Power: 0

I also gave the self diagnostic test a go and it didn't provide any codes.

 
Rep Power: 0

Update: I finally got around to replacing the AC thermistor, it took a good while to get here from hondapartsnow.com and it hasn't been hot at all so I wasn't thinking as much about it

Did some testing in the driveway, wedged a piece of wood on the gas pedal to get it at around 3000 rpm with the AC on full. The compressor is now cycling itself after it gets really cold (goes on about every 10 seconds then off for 10 more), which is a promising sign as it was more likely to just stay stuck on all the time before. We'll see though when I get a chance to drive it for an extended amount of time.

One question: Is it normal operation for the engine to drop about 500 rpm when the compressor kicks on, and raise back up 500 rpm when it shuts off? Seems like it would be, but just curious.

 
Rep Power: 518

Absolutely normal. The AC compressor requires several horsepower to run, that's a noticeable load change on the engine.