schekan Posted September 1, 2012 Share Posted September 1, 2012 I just replaced my knock sensor with a new one, i thought the old one was bad but i still have the same issue. My problem is when the knock sensor is installed in the correct position “positioned at a 45 Deg angle relative to the Engine” the idle surges from normal around 750 rpm to 1300 rpm or so. It’s cyclical and does not go away. When I rotate the knock sensor so the plug is facing the driver’s side fender it idles perfectly, until it gets hot. After driving it for a while the idle gets very high and is very slow to drop. It goes as high as 2500 to 3000 rpm. It does not do this when installed in correct position. But surges when it is in that spot. I am running out of ideas and am thinking I cannot be the only person who has had this happen. Why does the knock sensor have to be oriented in the certain direct as indicated by Subaru? It obviously does something in my case. I have read where people break the bolt and mount it in another location without ill effects, so why is mine so particular about which way it points?? If anyone has any ideas please feel free, I appreciate any help I can get. Link to comment Share on other sites More sharing options...
Gloyale Posted September 1, 2012 Share Posted September 1, 2012 I think you've got the wire looped over the throttle cable. Link to comment Share on other sites More sharing options...
mikaleda Posted September 1, 2012 Share Posted September 1, 2012 I just replaced my knock sensor with a new one, i thought the old one was bad but i still have the same issue. My problem is when the knock sensor is installed in the correct position “positioned at a 45 Deg angle relative to the Engine” the idle surges from normal around 750 rpm to 1300 rpm or so. It’s cyclical and does not go away. When I rotate the knock sensor so the plug is facing the driver’s side fender it idles perfectly, until it gets hot. After driving it for a while the idle gets very high and is very slow to drop. It goes as high as 2500 to 3000 rpm. It does not do this when installed in correct position. But surges when it is in that spot. I am running out of ideas and am thinking I cannot be the only person who has had this happen. Why does the knock sensor have to be oriented in the certain direct as indicated by Subaru? It obviously does something in my case. I have read where people break the bolt and mount it in another location without ill effects, so why is mine so particular about which way it points?? If anyone has any ideas please feel free, I appreciate any help I can get. I'm not sure about yours but I know that on my subie (1990 legacy) if I rotate it in th wrong direction it will not sit flat against the block. Also did you tourqe your sensor to specs? If not that may be your problem. Link to comment Share on other sites More sharing options...
Fairtax4me Posted September 1, 2012 Share Posted September 1, 2012 CEL on? Was the old sensor visibly damaged? (cracks/corrosion) Knock sensor has to sit a certain way so the sensor housing sits flat, otherwise it hits the block when tightened and will deform or crack the housing. Have you checked your theory with the knock sensor unplugged? Strange idle is usually a vacuum leak or a stuck idle control valve. High idle could be something like a sticky throttle cable or the throttle plate being wedged open slightly in the throttle body bore. Link to comment Share on other sites More sharing options...
schekan Posted September 1, 2012 Author Share Posted September 1, 2012 It is sitting flat in either direction, i am positive of that. And no it is not the throttle cable. The old sensor was not cracked and i had no codes. I found it funny that when mounted one way it surges and idles fine the other but goes very high when hot. I thought maybe it was bad for this reason i got a new one from Rockauto, it didn't break the bank at 80 bucks. There are no codes thrown and when i unpluged the old one it threw a code and ran like a piece of crap... I am still thinking vacum as well, just looking for some thoughts on the subject. Also I am going to check my TPS today to see if it is on the mark. Thanks for the ideas so far. I will post my results on that test later today. Also I need new battery terminals so I am going to try the "Idle re-learn procedure " I have seen in this and other forums. It can't hurt..lol:drunk: Link to comment Share on other sites More sharing options...
1 Lucky Texan Posted September 1, 2012 Share Posted September 1, 2012 do I recall that torque is 14 ftlbs ? anyone confirm? Link to comment Share on other sites More sharing options...
schekan Posted September 2, 2012 Author Share Posted September 2, 2012 OK, I install it as per Subaru, And adjusted my TPS and the idle surge is gone. So TPS is messed up. I have a couple I pulled from the bone yard that looked like they were never messed with (Still had the white paint on top screw) My problem is that according to Mitchel On Demand, the voltage at ECM B136 connector terminals 15 and 17 must be between 0.45 and 0.55 when fully closed. This has to be a misprint, i have between 4.5 and 5.0 when fully closed. When the throttle is opened it drops as expected. I am connecting it to the correct terminals according to the diagram. Does anyone know what the correct voltage should be when closed?? Or a link to the adjustment procedure. Link to comment Share on other sites More sharing options...
mikaleda Posted September 2, 2012 Share Posted September 2, 2012 It sounds to me like you have your multi meter on the wrong values. It wouldn't hurt to dOuble check your settings and try reading again. Link to comment Share on other sites More sharing options...
Fairtax4me Posted September 2, 2012 Share Posted September 2, 2012 How bout a year, model, and engine? Mitchel and Alldata are semi-good resources but they don't always have the correct info. On the older cars .45 - .55v is correct, but that may have changed between 1990 and say 2010. You may be testing across the wrong two pins. Or that could be right and MOD just has the wrong spec. With a year and model we can look up the ECU pinout in the FSM, and the TPS adjustment procedure. Link to comment Share on other sites More sharing options...
Caboobaroo Posted September 2, 2012 Share Posted September 2, 2012 Sounds like you are testing between the wrong pins. With the correct pins back probed and the key on (engine off), you should have roughly 0.50 volts. If its not near there, then rotate the TPS and watch the values on the multimeter until its within range. Link to comment Share on other sites More sharing options...
schekan Posted September 2, 2012 Author Share Posted September 2, 2012 02 outback, 2.5L. And these are the pins I am checking. So either the diagram is wrong or the voltage is not listed as correct. Does anyone have a different diagram for the pins? Link to comment Share on other sites More sharing options...
schekan Posted September 2, 2012 Author Share Posted September 2, 2012 I found the following listed below in another forum. I have a 4 wire TPS. I think this is the info I am looking for. As per the first link it states voltage should be 4.7 closed. Any other ideas? At least this is in the balpark of the voltage for my unit. I can get to go from 4.4 to 5.1 by moving it. Also my car is Cali emission if that means anything. Unless you have the 4-pin TPS, you don't need to adjust anything. The TPS is simply a potentiometer that is hooked up to the throttle plate. As the throttle is opened, resistance is changed. That information is sent back to the ECU to know what position the throttle is in. The reason I say it's not necessary to adjust the TPS is that the ECU will take a reading of the TPS's resistance at startup. It then uses that as it's "throttle closed" benchmark. If you adjust the TPS while the car is running, you can make the idle go up and down. If you leave the TPS adjusted so the idle is high, turn off the car and turn the car back on, the idle will return to normal. However if you adjust the TPS too far, its readings will be outside the set limits in the ECU, and the ECU will throw a CEL. On the 4-pin TPS's they had the idle switch in the TPS as well. The newer cars have a separate idle switch. So on the 4-pin TPS's you would adjust the TPS based on the idle switch to make sure it's readings are correct. here's the test procedure. Last scan tells you how to adjust the TPS via the idle switch http://www.main.experiencetherave.co...S_testing1.jpg http://www.main.experiencetherave.co...S_testing2.jpg http://www.main.experiencetherave.co...S_testing3.jpg Link to comment Share on other sites More sharing options...
schekan Posted September 2, 2012 Author Share Posted September 2, 2012 (edited) OK this thread this person had the same issue, Seems that the pins you check are 16 and 17. not 15 and 17 as the diagram shows. http://forums.nasioc.com/forums/showthread.php?t=1060234 I did that and got it adjusted. Seems to be working ok for now. Thanks for all the help offered by everyone. I appreciate it. Edited September 2, 2012 by schekan forgot link Link to comment Share on other sites More sharing options...
Fairtax4me Posted September 3, 2012 Share Posted September 3, 2012 OK this thread this person had the same issue, Seems that the pins you check are 16 and 17. not 15 and 17 as the diagram shows. Glad you got this figured out! I did look in the FSM and it says to test across pins 15 and 17, which is actually wrong because it will give the same reading you got, which is the difference in voltage between the reference pin (+5V) and the signal pin. To get the right reading you connect to the signal pin and ECU ground. We can blame Subaru for that screw-up because the diagram you have came straight from their manual. A little late posting the info (pre-labor day party yesterday, shots were involved ) but glad you got it figured out! Now the big Qstion is, did it fix the original problem? Link to comment Share on other sites More sharing options...
Recommended Posts
Please sign in to comment
You will be able to leave a comment after signing in
Sign In Now