Announcement

Collapse
No announcement yet.

RS2 Rough idle

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

  • RS2 Rough idle

    Hi! My freshly rebuilt ADU engine is not idling good and when hot, 90c it stalls if I don't open the throttle.
    Cleaned the idle valve and also tested an other idle valve. No change. So after some reading I suspect the Lambda.
    When I start it cold, 20c. The idle is fine for about 30-50sec, then it's starting to be very nervous.. jumping from 400-800rpm. Could the Lambda cause this? Every hose is checked and exchanged if it was bad.
    If the idle valve was faulty it shouldn't be ok for the first 30-50sec I think?. It's always the same time until he idle goes crazy.
    TPS is checked and functioning.
    Spark plugs and ignition is new.
    ECU gives me no fault codes at all. Any one had a similar problem?

  • #2
    I had a very similar experience recently with poor idle - pretty much as you describe. It would jump about from 400-800 and stall if I didn't give a little throttle when warm.

    What fixed this for me was unplugging the MAF connector giving it a quick clean and then reconnecting. As soon as I fired it up after that the idle was perfect. Definitely something to check if you haven't already.

    Here's some more info that helped me.

    https://www.s2forum.com/forum/techni...nector?t=24544

    https://forums.quattroworld.com/s4s6/msgs/20442.phtml
    93 S2 Avant - SOLD
    94 80 V6 Saloon
    00 S3 8L - SOLD

    Comment


    • #3
      Probably either MAF connector or an air leak.

      S2 Coupe 3B Project


      Ur quattro restoration

      S2 Avant

      Boost is the new rock and roll!
      sigpic

      Comment


      • #4
        Originally posted by newsh
        Probably either MAF connector or an air leak.
        That bloomin MAF connector!
        https://r.tapatalk.com/shareLink?sha...1&share_type=t

        S2 Coupe 3B Project


        Ur quattro restoration

        S2 Avant

        Boost is the new rock and roll!
        sigpic

        Comment


        • #5
          I'm soon going desperate.. MAF connectors/MAF pins are cleaned according to the FAQ. Also tried a different MAF. Still same problem. Let me specify what i have done so far:

          Changed MAF
          Changed Lambda
          Changed Temp sensor (back of the cyl-head)
          Changed ISV to a working one (from my UrS4)
          Changed gasrecirculation valve
          Checked all vacuum/crank case vent hoses
          Checked fuel pressure, though a little high on idle (4.0bar) but with the idle i have (very rough) i cant think thats a problem, if i rev it up and close throttle i have about 3,2bar.'

          Could it be the regulator or injectors? Injectors was a problem when i tried to fire up the engine at first. After ultra-sound cleaning the engine started right up.
          The car works fine when i give it full throttle, all the way up to 7000rpm. No hesitation. On revs or cruising in the 1000-2500rpm it can give signs of hesitation/missifre.
          Plugs are new and so the ignition (converted to 1,8t coils)
          Also the morning i failed the MOT due to high emissions.. But that was expected with my rough idle. Lambda was good though.

          One thing that i find is that VAG-COM is showing a engine load of 40-45% at idle, i find this a little suspicious, why could this be?

          Any one have some ideas?

          Comment


          • #6
            Throttle position sensor. Make sure it's rotated so you hear it click when the throttle closes.
            91 Modded 3B
            14 A6 Avant Black Edition

            Comment


            • #7
              Originally posted by SteveH View Post
              Throttle position sensor. Make sure it's rotated so you hear it click when the throttle closes.
              Forgot that one, the tps is clicking and the ECU confirms (Idle) i have seen that through VCDS. Also did the (TPS) FAQ.

              Comment


              • #8
                Well you would have seen that others have had a similiar issue. Unfortunately they have not indicated whether they solved the issue despite replacing a whole bunch of things. Other contributors implied they knew the answer but did not give us a clue as to what it was. So back to square one!
                So here are some ramblings that may or may not be helpful although the intention is always the former!

                Something obvious that perhaps we overlook is, according to a Factory Publication, is the necessity of, after disconnecting the battery from the ecu, is to allow the engine to start and idle at operating temperature for at least 10 mins. Presumably this to facilitate some baseline/adaption. I can see the logic! So you could try just pulling the motor fuses and then reconnect them. Start the engine and see what happens without touching the throttle. Obviously if it starts to misbehave after the first few minutes then its not the solution! It is curious as to why the problem starts after a certain period of time. Presumably the ecu either ignores signals and uses some pre programmed values or does not activate something (or both!). As the lambda value is static within the diagnostics until a certain period of time, I had assumed it was the lambda. Only one person would know for certain I guess. I have also read that the lambda control is not active until after 62C. Not sure how long the engine takes to get to that temp but it seems inconsistant with a repeatable time period!

                Fuel pressure. With the pipe connected then the value should, with a healthy engine be @ 0.5 bar less than the rated value of the FPR. The specs for the RS2 are 3.2-3.7bar with pipe connected and 3.8-4.0bar disconnected. So your values are almost the same as having no vac! I would check the pipework from the manifold and in particular the t piece in case that is allowing a vac leak. Again I am assuming you are using the RS2 FPR.

                The vehicle speed sensor also has an input in idle regulation. It apears that it is not activated above 6Km/h. Obviously if it reads the rather confusing stationary 4Km/h then thats not it either!

                Then there is the load. Above 35% is indicative of a electrical/mechanical load, a spurious air con signal or unmetered air between the turbo and throttle. The first can be eliminated by having no consumers on and making sure no steering lock is applied. The second is a possibility so is the third! Guess a leak test would not go amiss if you have not already tried this. I think I have read recently that the aircon boost does not work on the RS2 due to some code issue. Perhaps I am mistaken .
                A higher load would also be indicated by increased inj time (above 1.3ms) although in addition to the faults above that in itself could also be caused by a faulty MAF!

                Due to the cyclic nature of this issue I have wondered if it was related to the evap solenoid. This appears to be cycled for 6 mins at 0-7% duty cycle and then at 0% for 1 min. A malfunction of this component might be indicated by a lower inj time (below 1.0ms).

                Other than that I can only assume that as it has not solved by changing a lot of components then it points to something on the car! A pesky air leak of some sort perhaps. Its difficult to use the values provided by the diagnostics if the engine is cycling and there is no other clues from the ecu in terms of fault codes. You seem to be a methodical and thorough sort of guy so testing and elimination might be the solution.
                Last edited by twoqu; 3 September 2019, 16:34.

                Comment


                • #9
                  Thanks for your reply Twoqu! Very usefull information there! So you have read that the ECU need about 10 min of no touching the gas pedal for the ECU to adapt for the idle? This in that case i haven't tried. The car works fine the first 40sec or so, then starts to idle rough. When look at VCDS the engine load is about 40% and injector duty is about 3 ms.. Maybe i have a leak althoug the car is running fine without hesitations on load from 1500rpm up to 7000rpm.
                  This is from my log at VCDS (yes the coolant temp is low) but still load at 40% and inj opening time 3ms..
                  Have to look at the speed sensor also, so if im sitting still and the ECU register over 6km/h the car wont idle or what?
                  Could it be my injectors thats the problem? only thing that worries me about that teory is the injection time, and why it is so high, same as the engine load %
                  What do you say?
                  45764-ad24c7f8b8967359d25304c04dff20a2.jpg

                  Comment


                  • #10
                    Once I had idle problem, then I accidentally mixed IAT wires with ISV wires.

                    Comment


                    • #11
                      Originally posted by maiks View Post
                      Once I had idle problem, then I accidentally mixed IAT wires with ISV wires.
                      Yeah, problem is that i have working IAT and ISV is working when i do an output test with VCDS. So that easy is not my problem.

                      Comment


                      • #12
                        If I knew the answer I would just tell you!
                        So to be a bit clearer, It was just something I noticed tucked away in some info. When reconnecting the battery, i.e. after performing work on the vehicle, it was recommended to let the car warm up without touching the throttle to operating temp. Basically you can do this without disconnecting the battery just by removing the ecu fuses unter the Motor cover in the fuse box. Then put them back in and start the car but do not touch the throttle during the warm up. It will be clear whether this works or not within the first minute. It removes any learned values/adaption I guess so the ecu starts again. Worth a try.
                        The load and inj opening time are related so an increase in one increases the other. Not sure how much the coolant temp effects things such as enrichment and all the factory data applies to a normal temp engine 90C.
                        As for the VSS, the ecu needs to know the car is stationary for idle activation as I understand it. Stationary is 4 km/h!
                        If we knew the significance of what happens after the time period we might be nearer the solution.

                        Comment


                        • #13
                          Before someone shoots me down, the load/inj times were lifted from documentation fror the AAN not the ADU. It was just to illustrate the point rather than be definitive. Having said that, I doubt whether there would be much diference at idle, probably less!

                          So what is load?
                          Load inside the Motronic ECU is the injector time open to reach lambda 1.
                          It is calculated from the following:
                          1) Engine RPM
                          2) Mass airflow (kg/h)
                          3) The injector constant programmed into the ECU.

                          Comment


                          • #14
                            This is what Dmitri had to say about MAF's a while back :
                            https://www.s2forum.com/forum/techni...858#post927858

                            That is different to how load is expressed in other groups though.

                            Comment


                            • #15
                              I guess you could compare the RS2 maf with your AAN maf on the AAN if thats running ok. That should give you a better comparsion as to its condition. There is no electrical difference between them. The RS2 ecu has a change internally which gives a different reading compared to for example the AAN. If it runs ok on the AAN and is within spec more or less considering its age then you have eliminated one more thing. If you use the AAN maf on the RS2 and the problem is on the car we haven't eliminated it. So get readings from AAN first then swap to RS2 .
                              I can't remember exactly the calculation for coolant temp but if in group 000 field 1, 184 - 215 represents 85C - 105C, surely 161 should be more than 25C! Would have thought they both read from the same sender. Does it make sense when engine is warmed up?
                              Last edited by twoqu; 3 September 2019, 22:29.

                              Comment

                              Working...
                              X