ParetosFew

To follow up on this, it was in fact just over charged. Dealership evacuated and recharged by weight and it is now running great. Anyone looking at this later, make sure your high side gauge is reading correctly as this would have confirmed what ammcinnis was telling me early on.

I setup an appointment at the dealership, but thought I'd try to retry my measurements. Well, you were right ammcinnis and it only took another few dumb moves on my part to prove it out. I setup the gauges again since I was suspecting my original measurements of being inaccurate as well. Saw pretty much the same thing I measured previously (100PSI on low side while compressor not running, dropped down to about 55 when compressor tried to run but no change visible on high side which was at around 100 PSI the entire time). Being stuck in the mindset of that sensor not reading correctly, I forced a voltage of a little over 3V using AA batteries to simulate what should be an OK pressure. Admittedly, this was not a very safe thing to do. This did cause the compressor to run for about 30 seconds until what I'm assuming was the safety blow-off valve let out a blast of refrigerant. Of course after this I was now highly suspect of my 100PSI high side reading since clearly there was a lot of pressure there. I'm embarrassed to admit this, but I did not have the high side valve on the gauge quick-disconnect opened enough. Not quite sure how I got that wrong, but I swore it was moving and seemed to be reading prior but this was probably the same issue on the readings from my previous post. Once that was open, I saw the high side pressure max out at 500 PSI every time the compressor tried to run for a few seconds. This matches up with the voltage reading on the high pressure transducer, and now makes me very confident it is in fact just grossly overcharged. Another evidence of it being overcharged is this started out intermittent, and it is now not working at all since I added more refrigerant. I still have the appointment for July 6th, I will let them evacuate and charge properly and I'm pretty sure that will get it working again. I will keep updated what I found out then. Thanks for the suggestions and help, and as far as your question about the DTCs I'm not sure how to read them. I did connect the $10 bluetooth ODBII reader with Torque app, confirmed I was connected and checked for fault codes. Nothing was reported, although I'm not sure that would get enough detail.

I think at this point I will have it evacuated and recharged since I don't have the equipment to properly evacuate and recover. Once I'm sure the charge is at least correct, will continue to monitor for intermittent issues at higher temperatures. I may have not given it enough time when I took that first photo. Will possibly try measuring again from cold system just to see if I can recreate that measurement.

Got the gauge set, here are some updates. Started based on vehicle ambient temperature sensor reading 80F. With system off, the the high and low side pressures were very close. Both were within a few PSI of each other, around 75 PSI. With the system on, max cool set on AC and engine at idle, and given about 5 minutes or so of trying to run AC saw the following readings from first image - LOW 30 PSI, HIGH 20 PSI. Following the table in the instructions added refrigerant until the low side was a little at 60PSI (accidentally went a bit high, was aiming for 55). After the engine warmed and after finished adding refrigerant, was reading LOW 100 PSI, HIGH 100 PSI, which seemed reasonable to me given the 80F ambient plus engine heat. Wish I'd have measured the temp near the engine. At this point, the compressor was repeatedly trying to run for several seconds then stopping. I've attached a video of this, you'll see the pressures drop down when this happens. Video: https://streamable.com/7kitvx Still suspecting the high side pressure sensor, I measured the signal voltage on the center conductor of the sensor. At 100PSI was reading 3.97V, when the compressor tried to run this voltage quickly jumped up to 4.5 and climbed to 4.9 before the compressor stopped trying to run. I'm confused as to why the reading would increase despite the gauge pressure dropping. Regardless, the voltage readings seem way high for a 0-5V sensor and the pressure ranges involved. I've been having trouble finding concrete information on what the range of that sensor is from Subaru manuals, but I found a general guide for this type of compressor that says the general range should be 0-5V, 0-500PSI ( GUIDE HERE, PDF ). Makes perfect sense to me that the controller would shut off of the compressor with a reading of near 5V on that sensor since that should mean the pressure is very high. I am going to keep researching that part to see what else I can find, please let me know if I've made any errors in my troubleshooting so far.

Thanks for the responses here, will definitely update with what I find. I saw your comments mentioning this on some other AC threads found while searching, and have since ordered a gauge set to see both high and low side pressure. My next step is to use this and take temperature into account to check the charge, hopefully this arrives today. Forgetting how to multiquote here, been a while since I've used this type of forum, but to answer some other things: - While I have the gauges attached, if the pressures seem in reasonable range I will try to measure the voltage from the high end pressure sensor. - Will try swapping relays. - Ordered one of those $10 bluetooth ODBII readers, although I think this may not get see the codes I need for this problem I like having these for check-engine lights etc. Maybe it could help here. - Painless - thanks for that, it will be helpful. - Car owned by my since new, no recent accidents or anything that would effect the front end. I somewhat remember working on another vehicle and possibly "topping off" the refrigerant on this one having some left in a can (realize now not a smart move). I will try these few things and check back in, may end up taking it to dealership for a SSM scan if I don't make any headway with any of these steps. Edit: Doh! Should have hit m quote button, didnt' see it once I'd already hit the quote button.

2015 Legacy Limited, 2.5 Ironically, only when it is extremely hot it out and most necessary the AC goes intermittent. It will run for a few minutes then blow warm air, then cold again. It does this many times, but I've noticed on longer trips it eventually just runs consistently after 20 minutes or so. I checked the refrigerant with one of the can-top gauges and it was in the lower end of the green range - added a bit to bring it to center of green range but made no difference. My searching showed several forum posts about the compressor clutch sticking, but my clutch looks different than a traditional compressor clutch I'm used to seeing. I think this may be because this car has a variable displacement compressor? That's a bit beyond my knowledge, but read something to that effect on one of the posts. I suspect the pressure sensor but am having real trouble finding information on that. I'll attach a diagram from subaru parts with the part I'd like info on - it's the only thing on that diagram without a part number. Diagram from this page: https://parts.subaru.com/Subaru_2015_Legacy-25L-CVT-4WD-Limited/Heating-and-Air-Conditioning.html Any information on testing this sensor would be helpful. I found a service manual online for a car that should be close (17 outback) and it said that sensor should read 0.5-4.9V when the ignition is on. Leads me to think that is an analog transducer but not sure if I can test that with the connector disconnected. Thanks!

I actually came here looking for info on this issue for my 2015 Legacy. I'd like to do the DD C-lights but don't want to put them into these in their current condition, so might be looking for new headlights for that. I'm sure very frequent automatic car washes didn't help this any. Some pics...