The Top DSM Community on the Web

For 1990-1999 Mitsubishi Eclipse, Eagle Talon, Plymouth Laser, and Galant VR-4 Owners. Log in to remove most ads.

Please Support JNZ Tuning
Please Support Rix Racing

2G AC Pulsing

This site may earn a commission from merchant
affiliate links, including eBay, Amazon, and others.

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

kingje1

Proven Member
84
22
Feb 27, 2022
Nashville, Tennessee
HI all, when I turn my AC on It pulses about once a second. I feel the pull on the engine with every pulse. Coldish air comes out but it's not cold enough to keep me cool in the southern heat.

I'm sorry if this has been answered before, I've been looking through threads and searching now for 30 minutes but haven't found the answer. I am Gen X so forums and tech comes a little harder for me.

Here is what I have done so far:
1. Checked the connections for anything obvious like exposed wires etc and found nothing.
2. Checked the pressure and added until it was good. Air got cooler but still not cold enough.

I've been blessed in the past with cars that have not had issues with AC any more than slight leaks.

I would appreciate any help or a link to a thread that could help me.
 
Last edited by a moderator:
Hey guys, thanks again for your help, I feel like I am really close to closing in on what's causing the issue. I had a death in the family so I was out of town and then my body shop guy called and told me he was ready. Here are some pics of the car at the shop getting prepared for paint.
You must be logged in to view this image or video.
You must be logged in to view this image or video.
 
As soon as I get her back I'll start testing the circuits of the relay to find out which side the pulsing is coming from. While I was in Jalisco, Mexico for the funeral I passed tons of massive junkyards in Guadalajara just wondering how many eclipses were sitting there. I don't know how popular the cars were in Mexico as I only saw one the whole time I was there (2ga). Maybe when I'm there for leisure travel I'll do a junkyard tour. . . Mexico is incredible for fans of old cars as they keep them forever there.

Again, I do appreciate you guys for helping me, please stick with me while I'm waiting to get the car back.
 
Here is where I am murky. Do I test for voltage using chassis ground or should I use socket 1?
I am assuming that if one of these components is bad the voltage will come and then disappear and come again here at this socket. I guess what I don't know is how I would know if the socket is good? We don't know what causes the pulsing and it's intermittent.
Also, How do I test for ground at socket 1?

You're probably talking about socket A-18X which is the socket for the compressor clutch relay. It would be good to test between pin 1 of that socket and any handy ground when the problem is occurring.

If that voltage is always low (around a volt or so) when you have A/C turned on (even if the clutch is pulsing) then the ECM and A/C ECM are seeing all good inputs and your trouble is in the relay itself or somewhere to the right of there on the diagram. If the voltage there pulses between high and low (high probably 12V) then the ECM is seeing something wrong.

And if the voltage there pulses between low and 0 then the trouble is in the A/C ECM or to the left of there.

I would do that test by wrapping a thin wire around that pin of the relay and pushing the relay not quite all the way into the socket: BE VERY CAREFUL doing this because a short there may fry your ECM and/or other expensive parts.

That's gonna be a beautiful car one day soon. I think this is more an electrical problem than an A/C problem and you'll find it pretty quickly once you get your car back.
 
I got the car back and I immediately needed to do some cosmetic work on it because the paint job is beautiful and is making the old ugly parts stand out. I've got to the point thought that I want to start testing this thing to figure out what is going on. I'm sorry that I am not super knowledgeable about electrical testing. I've watched many videos on it but something in my brain is unable to wrap itself around the concept of it.

I want to start with step 1 which is testing socket 4 and 5 which are the sockets that if all is well on socket 1 and 3 will switch on and send power to the compressor clutch. I just want to make sure that I don't fry the ecu or anything else important. In order to test 4 and 5 I have a bread cable with the ends pulled off to create a jumper that I stick one side in socket 4 and one in socket 5 to send the signal to the compressor. Of course while I do this I will have the manifold gauges connected to monitor pressures. If I understand correctly this should send consistent power to the compressor clutch and I shouldn't have any pulsing. If I have pulsing while jumping 4 and 5 my issue is in the compressor clutch or somewhere in the closed system but it would not be electrical. If I don't get pulsing and I get consistent clutch and compressor operation then it's time to move on to step two and test either socket 1 or socket 3.

If anything I said in the above paragraph is wrong please let me know. If you are very confident about testing the AC and I am correct about the above paragraph I would appreciate some validation as I am a little nervous about this. Thanks everyone who is helping out.
 
I got the car back and I immediately needed to do some cosmetic work on it because the paint job is beautiful and is making the old ugly parts stand out. I've got to the point thought that I want to start testing this thing to figure out what is going on. I'm sorry that I am not super knowledgeable about electrical testing. I've watched many videos on it but something in my brain is unable to wrap itself around the concept of it.

I want to start with step 1 which is testing socket 4 and 5 which are the sockets that if all is well on socket 1 and 3 will switch on and send power to the compressor clutch. I just want to make sure that I don't fry the ecu or anything else important. In order to test 4 and 5 I have a bread cable with the ends pulled off to create a jumper that I stick one side in socket 4 and one in socket 5 to send the signal to the compressor. Of course while I do this I will have the manifold gauges connected to monitor pressures. If I understand correctly this should send consistent power to the compressor clutch and I shouldn't have any pulsing. If I have pulsing while jumping 4 and 5 my issue is in the compressor clutch or somewhere in the closed system but it would not be electrical. If I don't get pulsing and I get consistent clutch and compressor operation then it's time to move on to step two and test either socket 1 or socket 3.

If anything I said in the above paragraph is wrong please let me know. If you are very confident about testing the AC and I am correct about the above paragraph I would appreciate some validation as I am a little nervous about this. Thanks everyone who is helping out.

You can jumper pin 4 to pin 5 on A-18X without causing a problem and this should engage the A/C clutch. Don't do it for more than a few seconds though because the ECM ordinarily turns on the fans to pull air through the condenser when the clutch is engaged and shorting those pins won't make that happen. So the A/C high side pressure will go up a lot higher than normal if you keep it going.

However -- I've done that test and it worked fine.
 
Thanks Waltah for the confirmation. I just did the test outside and jumped 4 and 5. The clutch engaged and it did not pulse. I did monitor the gauges and saw that the low pressure got down to between 10 and 18 PSI (hard to tell exactly since the gauge is not marked well) and stayed there while the high pressure rose steadily. I pulled the jumper out as it got slightly over 150 psi. Could this be an indication of me being low on freon or should I just move on to the other tests on the relay? I know that the fans are not engaged when doing this test, will that cause the low side to be lower than normal on the pressure? The car is not consistent when it's idling. The only way for me to get it to start pulsing is to drive around until it starts pulsing and then test it while it's pulsing.

As for the next test, please confirm if I have this correct.
I should have a wire tied around pin 3 of the relay and push the relay into the socket. I should use the multimeter set to test voltage to pin three with one test lead and the other test lead on chassis ground. This measures to see if I am getting voltage at pin 3 which comes when the blower switch is on, the ac switch is on, the dual pressure switches are set to on and the temp switch is on (engine is not overheating). I should have either no voltage or I should see constant voltage if this side is good. If this side gives me a voltage that fluctuates on off then I know that the pulsing is caused by one of the switches on this circuit.
This last one I'm a little unsure of. To test socket 1 on the relay circuit to do I set the multimeter to continuity and put one test lead to the socket 1 pin and one to chassis ground? If I have continuity then I have ground at both places meaning that the ECU is good. If I have fluctuating continuity and then no continuity then I have a problem on the ECU side?

Please let me know if I am understanding correctly with the steps here.
 
Pressures sound good to me. I don’t think it’s low on Freon with the low side getting down that low, it should have been cooling well. 150 on the high side is normal too, although I’m sure without the fan you would have seen a continued climb so you did the right thing.

I haven’t studied the circuit diagram to confirm your question at the moment but I had a thought that I wanted to share, since you confirmed the coil is good by jumping the relay. Sorry if I missed or didn’t remember you already doing either of these steps.

Have you tried moving/wiggling the wire harness around all the connections to the sensors/fuse box while it’s running good to see if you can cause it to start the short cycling? Have you removed the ecu and looked at the board for cap leak damage?
 
Last edited:
Derisch, thanks, that's a great idea. Maybe I can get it to start pulsing by messing some with the wires. I need to pull the ECU anyway to find out what I need for ECM Link but I'd like to do the electrical tests before doing that just so I know what side the issue is on, the ground side or the power side.

Are you a multimeter expert? Did I hit get the testing steps correct? If not, anyone else?
 
Derisch, thanks, that's a great idea. Maybe I can get it to start pulsing by messing some with the wires. I need to pull the ECU anyway to find out what I need for ECM Link but I'd like to do the electrical tests before doing that just so I know what side the issue is on, the ground side or the power side.

Are you a multimeter expert? Did I hit get the testing steps correct? If not, anyone else?
Is there a way to access the underside of this relay to backprobe the terminals that way?This lowers the risk of touching the added wire on something by accident or negatively affecting the contacts inside the terminal from jamming wire into it.

Alternatively, Can you just test terminal 3 with the relay removed? Maybe with a test light would be easier. You could unplug each connector to verify the power comes and goes. And wiggle on the harness and connections to see what happens.

Is this the only ecm controlled circuit that’s acting up? Leaking caps is extremely common and since you wanted to check your ecm out anyway I really think that would be something easy to do to pull it out real quick and let’s have a peak at the board.
 
Last edited:
Check the dual pressure switch. Looks like you have some blockage in the lines which could be triggering the dual pressure switch.
 
I already performed the dual pressure switch check in the Service Manual and when I bypassed it I still got the pulsing so I've eliminated that as a possibility. Sorry I haven't gotten to this yet, my Acura TL (being driven by my son) had the radiator go out and my F150 had the compressor pulley bearings go bad so I got sidetracked. I agree with Dericsh that it's a little shady having the wire in there with the relay connected. I do have some electrical testing adapters thought that might do the trick. I haven't gotten any confirmation on the testing procedures I laid out. Am I on the right path here?
 
I didn't hear from anyone with confirmation for my testing procedures but I watched enough youtube videos on using a multimeter to test relays until I felt confident. I used some small clips (see picture) that grab onto the edge of the relays so there was almost no way I could have shorted anything as there is almost no exposed metal once they are clamped on. I was able to get the pulsing to happen a few times and determine that it was definitely on the power side. My ground Pin 1 had continuity but the multimeter voltage switched between 0.9 and 12v on pin 3 when the pulsing was going on.
Now my next question. When I pulled the connector off the pressure switch and jumped the two terminals on the connector the AC was still pulsing. I took that to mean that it couldn't have been the pressure switch since I was effectively bypassing it and it was still pulsing. Is that correct? Here is what I did according to the service manual.
You must be logged in to view this image or video.

Here is a picture of the clamps I used.
You must be logged in to view this image or video.
You must be logged in to view this image or video.
 
Sorry I didn’t confirm your test, I lost my confidence in what voltages to expect when the power and ground is both ecu switched and hoped someone else would answer. But anyways- Yes, it sounds like you correctly bypassed the dual pressure switch.

I was wanting to move the test to this smaller AC control unit before, but was waiting for your results on the relays.

Let’s move the multimeter to inside the cabin. By the evaporator, pin 1 green/orange wire on the “ACCU”. See if you’re losing voltage there when the compressor starts to short cycle.

References:
This is where the power comes from per the schematic

You must be logged in to view this image or video.


Thread mentioning the unit (with photo):

 
Dericsh, thanks for sending that thread. I was able to find the ACCU which in my 95 was underneath the the evaporator. I found the Green and Orange wire and I measured the voltage there. I was able to get it pulsing and I got a steady voltage during the pulsing (between 12 and 13 volts) I could not find any movement in the voltage as the pulsing was occurring. In all of this testing, however, I have not tested the pin 1 on the A18X relay while it was pulsing. I only tested the PIN 1 ground while the AC until was switched on and I had ground. I guess in my mind since the pulsing was causing voltage to go on and off on PIN 3 there was no need to try to test PIN 1 while pulsing. Is there any relationship between what's happening on PIN 1 and 3 or are they independent? I've tried to go back and test it several times now but I can't get it to pulse long enough during Idle for me to get out and test it. It only wants to pulse while I'm driving.
 
The relation between pins 1 & 3 are just what you see, 3 is the power side and 1 is the ground side of the control coil in the relay. If the ecu is controlling the ground at pin 1 to control the relay then yes you’ll have ground while it’s working, but I’d expect you’re losing it while it’s not. I would backprobe pins 1 and 3 with your meter and watch the voltage. It will be steady battery voltage while the coil is operating.

I’m back to leaning towards an ecu board inspection.
 
OK, I was able to get the car pulsing today enough for me to test the PIN 1 on the relay and sure enough I lost ground when it was pulsing. This means that I am losing both ground and power on the control side when the pulsing is happening. I feel like I need someone with some electrical knowledge to chime in here. Is my testing affected by the fact that the relay is plugged in? Should I be testing the circuit without the relay plugged in? Why did I have the relay plugged in the first place? I'll go ahead and try to pull the ECU and open it up to see if I see anything leaking in there. Is there anything else I should look for?
 
So either something like the pressure switch or evaporator temperature switch are telling the ecu that it NEEDS to turn the AC relay off, and it’s working as it should— OR, the ecu is malfunctioning and losing its control of the relay. We know for sure that your problem is on the control side of the relay. I remember you checked the pressure switch by bypassing it. Remind me, have you messed with the thermo switch in the evaporator?
 
Hey im jumping in really late but here is my experience. Nice color choice by the way 😁 (see my profile car)

The expansion valve in the dash will not maintain pressures because the evaporator core is typically covered in a nice coating of slime from dead leaves and dirt from coming into the ac vents. The core also has a temp sensor set from the factory to monitor the core temp so it wont freeze over

On my project cars (over 15 years old) they always need a refresh. New ac o rings for the system, new drier, new expansion valve, and an extreme cleaning of the evaporator core. Typically the core wont get enough air flow through it to blow cold through the dash and the expansion valve releases pressure from the low side to high side causing the pressure switch to cut power to the compressor

I know youre doing a lot of electrical work but to me it sounds like it is all functioning correctly, if the expansion valve is working properly and the evap core is cloggued.

Heres what I typically have to scrape off of them in my 1990 hondas to get ice cold ac in texas

You must be logged in to view this image or video.
You must be logged in to view this image or video.
 
Hey im jumping in really late but here is my experience. Nice color choice by the way 😁 (see my profile car)

The expansion valve in the dash will not maintain pressures because the evaporator core is typically covered in a nice coating of slime from dead leaves and dirt from coming into the ac vents. The core also has a temp sensor set from the factory to monitor the core temp so it wont freeze over

On my project cars (over 15 years old) they always need a refresh. New ac o rings for the system, new drier, new expansion valve, and an extreme cleaning of the evaporator core. Typically the core wont get enough air flow through it to blow cold through the dash and the expansion valve releases pressure from the low side to high side causing the pressure switch to cut power to the compressor

I know youre doing a lot of electrical work but to me it sounds like it is all functioning correctly, if the expansion valve is working properly and the evap core is cloggued.

Heres what I typically have to scrape off of them in my 1990 hondas to get ice cold ac in texas

You must be logged in to view this image or video.
You must be logged in to view this image or video.
Woah that layer of gunk is impressive!
 
Deric, I haven't messed with the thermo switch, just checked the voltage from the ACCU. Chrysler Kid, I am getting cold air at the vent of 50-53 degrees but I don't know that it's cold enough. Does that still sound like my evap core might be clogged? Have you gotten the pulsing before?
 
Ok- I just re read this whole thing to refresh myself. Hopefully soon we can figure this problem out. You’ve made great progress, but here’s something we missed or just haven’t made it to yet.

Connector A95 is on the actual AC compressor. Pin 3 sends voltage to the clutch coil. Pins 1 and 2 are the pins for the refrigerant temp sensor. They should have continuity while the temperature is in the correct range. That’s how power makes it to the control side of the coil relay (pin 3 on relay) . If the fans are working and the pressures are correct, I don’t see any reason the temps should be too high, so you shouldn’t see a break in continuity here. If you’re losing continuity at pins 1 and 2 during the short cycling event, it’s likely that the switch is faulty. I’m not sure if it’s built into the compressor assembly or if it’s replaceable separately. I’m strictly going off of info from the electrical service manual. I don’t have a 2g to go look at, I looked at some photos of compressors on eBay to try to get an idea but none of the pictures are very clear. Best guess, It looks build in to the compressor assembly.
 
Woah that layer of gunk is impressive!

Yeah and its been all of them, not just one car.

Typical symptoms include low side pressure readings that spike high and the car refusing to run proper low side pressures when filling them with refrigerant.
Deric, I haven't messed with the thermo switch, just checked the voltage from the ACCU. Chrysler Kid, I am getting cold air at the vent of 50-53 degrees but I don't know that it's cold enough. Does that still sound like my evap core might be clogged? Have you gotten the pulsing before?

Yes thats all they do, you can feel the engine buck back and forth as the compressor cycles on and off as your drive it. The best I could ever do was run lower pressure (like 30psi) on the cold side and they would maybe blow Luke warm at idle, but then when youre driving around and it gets hotter under the hood they start bucking around and clicking the compressor on and off

Your pressure readings were inconclusive. Imagine taking pressure readings while holding the rpms at 2500. The system is extremely dependent on correct pressure and temp readings.

The ac drier, expansion valve and o rings and freon should run you less than $100 and a saturdays worth of work.
 
Support Vendors who Support the DSM Community
Boosted Fabrication ECM Tuning ExtremePSI Fuel Injector Clinic Innovation Products Jacks Transmissions JNZ Tuning Kiggly Racing Morrison Fabrications MyMitsubishiStore.com RixRacing RockAuto RTM Racing STM Tuned

Latest posts

Build Thread Updates

Vendor Updates

Latest Classifieds

Back
Top