Announcement

Collapse
No announcement yet.

Testing Subaru Pressure Switch

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • #46
    Another interesting note, on the recent 700 mile round trip we just took in the car, the car got the best fuel economy ever. 30mpg, its only rated for 26 or 27 highway. Wonder if it was because the AC wasnt kicking it? Windows were down for most of the highway drive too, 65-75 mph.

    Comment


    • #47
      At 65f outside it would be hard to assess the level of charge, 80+ is a better temp to check, but does sound like it is in the ball park. Glad it worked out. Be sure to thank Ignacio for the wiring diagrams, without him we would all be fumbling in the dark!

      Comment


      • #48
        Duly noted the results are at 65F which is below what I'd like for ambient temps or expected performance. Let it be but more #s line up to charts that indicate "normal" at precisely 80F.

        Do use the listed "charts" of what is normal with a grain of salt. Keep a thermometer in a vent and if possible know the temp of air coming in thru the grille then you really know the performance of a system.

        Factors of charts disregard things like pavement is frying hot, humidity out of control and situations stuck in traffic catching heat from a vehicle in front of you.

        Good luck with it and hope it all lasts now,
        Tom
        MetroWest, Boston

        Comment


        • #49
          Greetings all. I found this thread trying to understand the operation of my 2010 Subaru Foresters pressure switch, so I can diagnose what's going on with my system. Rather then start a new topic I figured I would just add to this one. Hope that's "COOL", haha AC humor.

          So here are the symptoms. Compressor isn't engaging. I have a wiring diagram and have followed each wire from the pressure switch and the AC relay. Switching side of relay has 12V when Ign on then goes to the clutch then to ground (All as it should). Trigger side gets 12V from the same ignition source through the NC side of the pressure switch and gets grounded to engage the clutch by the ECU.
          NO side of the switch is showing "OPEN" Not sure the function of this side. It is simply a NO switch that sends ground to the ECU if it closes.
          Fuses are good, Relay is good. If I send ground to the trigger side of the relay, the relay closes and the clutch engages fine. So I'm not getting a ground signal to trigger the relay from the ECU. Measured it at the relay and it shows open.
          Switch light comes on and stays on. Fan does not come on and idle does not increase when switch it pressed (manual transmission)

          System has been evacuated and recharged 2x and the problem remains. Now here is the irritating part, and why I'm going crazy:
          If I ground the relay and force the compressor to come on, after some time (couple seconds- 1min) the idle will increase, the fan kicks on the ECU grounds out the relay. I can then unhook my temporary ground and it cools great until I turn the AC/car off. On restart I have to repeat this procedure if I want the AC to come on again.

          I got a set of gauges: The high side cycles between 220-280psi (do you quote the psi scale or the r134a scale?) Low side about 40psi. One thing I noticed was that with the compressor running the low side pressure wasn't dropping (It should right?) and I eve saw the high side go down a bit at one point.

          So I guess my questions for y'all after all that are:
          1) What else could be preventing the ECU from sending a ground to the relay, the mystery side of the pressure switch?
          2) Do my pressures sound ok?
          3) Where else could I be looking for the problem?

          Thanks for any help you can offer

          Comment


          • #50
            I think were going to send you to the same place we did the original poster. Ignacio (Nacho) has the wiring diagrams to help trace the problem. Could be the same problem the OP had, bad wiring in the fender well.
            I'm not sure, but I think the ECM ground side of the switch is to raise the idle and possibly turn on fans when the high side pressure is high.
            PSI is PSI, the 134a scale is temp of the refrigerant at that pressure. Quote PSI as it doesn't change with refrigerant so less confusion
            Last edited by Cornbinder89; 07-18-2017, 09:03 PM.

            Comment


            • #51
              I do have a wiring diagram that I think is for the correct model. The pressure switch is getting 12V when ignition on as it should. Seems all I am missing is the ground signal from the ECU to get everything going. Trouble is that sometimes it can be "made" to work as I described so I doubt a bad wire is the cause. Are there any other sensors or switches that tell the ECU not to turn on the AC for some reason?

              Comment


              • #52
                It takes much more current to "pull in" a relay then to hold it in. You may have a bad connection with enough resistance that it can't pull the relay closed but can hold it in if you override it. Trace out the wire in question and check for corroded connections. If it is a problem in the ecm not makeing an internal ground it would mean a new ECM, but I wouldn't suspect it as the 1st thing. I would rule out the rest of the circuit before I suspect the ECM.

                Comment


                • #53
                  to be clear, there are two halves of the switch, one has 12 volt feed and controls the clutch, the other half sends a ground signal to the ECM for the fans, the two halves have almost nothing to do with eachother. I am assuming you are saying the ground for the clutch relay is not working.

                  Comment


                  • #54
                    Originally posted by Cornbinder89 View Post
                    It takes much more current to "pull in" a relay then to hold it in. You may have a bad connection with enough resistance that it can't pull the relay closed but can hold it in if you override it. Trace out the wire in question and check for corroded connections. If it is a problem in the ecm not makeing an internal ground it would mean a new ECM, but I wouldn't suspect it as the 1st thing. I would rule out the rest of the circuit before I suspect the ECM.
                    I will give this try but as its the ground signal from the ECU that seems to be missing to complete the circuit not the 12V supply to either side of the relay, I don't think this could be the problem. But why not rule it out.

                    Comment


                    • #55
                      Originally posted by Cornbinder89 View Post
                      to be clear, there are two halves of the switch, one has 12 volt feed and controls the clutch, the other half sends a ground signal to the ECM for the fans, the two halves have almost nothing to do with eachother. I am assuming you are saying the ground for the clutch relay is not working.
                      Not quite.
                      Ignition on 12V goes through the switching side of the relay to the clutch
                      The same 12V source goes through the NC side of the pressure switch - into the ac control module - through the trigger side of the AC relay - to the ECM where it need ground to activate the AC and turn on the fans and increase the idle. There is no direct electrical connection to the fans from the AC system so I assume its all controlled by the ECM
                      The NO side of the switch only goes from the ECM to ground and doesn't seem to effect the fan operation. Maybe a high pressure cut off?

                      Comment


                      • #56

                        1st and foremost, I don't have access to the wiring diagrams, so I am going off how it is described in the post. My understanding is the relay is not in the ECM so there must be wiring and connection on the ground side of the relay. A bad connecting on the ground side will give the same results as a bad connection on the 12 volt feed. Since you can get the compressor to turn on without the fans, I would say the other half of the switch is sending a signal to the ECM to turn the fans on, otherwise they would come on with the compressor. This is how a normal trinay switch would work also.
                        Again, I would be looking for a bad wire or connection between the relay coil and ECM.

                        Comment


                        • #57
                          Sorry if a repeat and parts look ups can be a way to see what relays are there to do what can be helpful. They can be tested off car and show good and not be enough. Many would respond to a tap once then go looking. Points inside may close and not have the power for a clutch. Some relays (if you just follow wires from clutch if possible find it if willing to ruin it will come apart and can clean points like ignition points of long ago just need tiny files.

                          Yup ~ 8 years old can be these suckers or where they plug in and any plugs along the way screw you all up.

                          Shows this is picture show from Autozone can also be totally wrong picture or part not corrected at their site at the bottom. Can lube spades, twist gently if off/loose, replace (shows $20) or keep testing LED test light backprobed with a "T" pin if anything happens.

                          LED is because if you use plain test lights or back powering looking for ground you do risk blowing ECMs.

                          Probably said sorry - long thread, limited time today. Gauges and pressures mentioned first post do NOT tell you system has correct charge at all just within operational ranges.

                          Many once empty for any reason or if ECM thinks it was the reset is to disconnect Battery negative, lose all stored info if not with the proper code reader should allow a "grace" period for it to work/engage for testing rather than jumping stuff,






                          Fits Your Vehicle Duralast Relay - A/C Compressor Control















                          ‚Äč Fits Your Vehicle Duralast Relay - A/C Compressor Control
                          Tom
                          MetroWest, Boston

                          Comment


                          • #58
                            Originally posted by Cornbinder89 View Post
                            1st and foremost, I don't have access to the wiring diagrams, so I am going off how it is described in the post. My understanding is the relay is not in the ECM so there must be wiring and connection on the ground side of the relay. A bad connecting on the ground side will give the same results as a bad connection on the 12 volt feed. Since you can get the compressor to turn on without the fans, I would say the other half of the switch is sending a signal to the ECM to turn the fans on, otherwise they would come on with the compressor. This is how a normal trinay switch would work also.
                            Again, I would be looking for a bad wire or connection between the relay coil and ECM.
                            I see what you are saying and will try this today. I also PM Nacho and he is sending me the wiring diagram today, just in case its different from the one I found online.Thanks

                            Comment


                            • #59
                              Originally posted by Tom Greenleaf View Post
                              Sorry if a repeat and parts look ups can be a way to see what relays are there to do what can be helpful. They can be tested off car and show good and not be enough. Many would respond to a tap once then go looking. Points inside may close and not have the power for a clutch. Some relays (if you just follow wires from clutch if possible find it if willing to ruin it will come apart and can clean points like ignition points of long ago just need tiny files.

                              Yup ~ 8 years old can be these suckers or where they plug in and any plugs along the way screw you all up.

                              Shows this is picture show from Autozone can also be totally wrong picture or part not corrected at their site at the bottom. Can lube spades, twist gently if off/loose, replace (shows $20) or keep testing LED test light backprobed with a "T" pin if anything happens.

                              LED is because if you use plain test lights or back powering looking for ground you do risk blowing ECMs.

                              Probably said sorry - long thread, limited time today. Gauges and pressures mentioned first post do NOT tell you system has correct charge at all just within operational ranges.

                              Many once empty for any reason or if ECM thinks it was the reset is to disconnect Battery negative, lose all stored info if not with the proper code reader should allow a "grace" period for it to work/engage for testing rather than jumping stuff,
                              I have a pile of known good (tested) relays that I have swapped in. No change
                              The system had been evacuated and recharged 2x the proper amount 500g (18oz) of r134a. Car spec is 450-550g (16-19oz)

                              Comment


                              • #60
                                OK - sorry - the long thread is just harder. Now question is if that relay puts OUT to the clutch presumably next in line does it get out of fuse box or where it is? Is so right there this becomes a wire chase even the best of diagrams don't really tell you they ran it thru the glovebox on to sun visors or who knows?

                                Don't forget there may be another way to cut ground or power to clutch COIL as a safety when refrigerant is too low requires a reset some posted on a sticker underhood that you have to?

                                I just doubt that but can't be certain with this car,
                                Tom
                                MetroWest, Boston

                                Comment

                                Working...
                                X