News

360 knock sensors

Discussion in 'Technical Q&A' started by 4i2fly, Aug 17, 2004.

  1. 4i2fly

    4i2fly Formula 3

    Apr 16, 2004
    1,330
    SF, Bay Area
    I got gas this morning at the usual place a Chevron station locally where I live and headed down to FCA Concourse at the Quail Lodge. On the way down there after 30 - 40 miles I noticed the car is hesitating at the about 7k rpm under acceleration otherwise ran o.k. Concerned about the problem, on the way back I stopped at a local shop and explained the problem. He took the car for a ride and noticed the same thing. He hooked up the sdii and his laptop and we went for a ride. He immediately noticed both knock sensors were showing faults. He also checked everything else (variators, throttle position, air induction, etc...) all normal. I suspect a bad case of gas at a reputable station. Under his suggestion, I filled up the tank, it was 3/4 empty and added the redline additive and suggested to put some mile before it goes away. I drove home about a 25-mile drive the car still feels the same way. I am going to the track this weekend and like to get this fixed as soon as possible if it is not the gas, any suggestions. Thanks in advance.
     
  2. To remove this ad click here.

  3. SrfCity

    SrfCity F1 Veteran

    Nov 1, 2003
    9,842
    Orange County, CA
    FWIW, I heard that Chevron with Techroline sucks and can cause problems. I use the new Shell V-Power and it seems good. My suggestion is to try and fill with another brand of gas and see what happens. Good luck.
     
  4. Diablo

    Diablo Formula Junior

    Bring it to the shop and have them take a look at the knock sensor torque.

    Some of the 360's loosened up and started acting up that way.
     
  5. 4i2fly

    4i2fly Formula 3

    Apr 16, 2004
    1,330
    SF, Bay Area
    I usually mix 100 and 91 octane but sometimes I just pump 91. I never had any issues with Chevron in any of my cars.

    If it doesn't go away after a few mor miles I will bring it in to check the sensors. But isn't it odd that both would show the fault?
     
  6. thomas_b

    thomas_b Formula Junior

    Sep 15, 2003
    765
    it is also odd that this would result in a fault - I believe the system generates only out of range and wiring errors - must be really bad gas :)

    the additive is a useless exercise since they don’t boost octane enough if they would mix evenly in the first place

    looks like the best option is to dump the gas for good one or to go for a mixture with 100 octane
     
  7. To remove this ad click here.

  8. 4i2fly

    4i2fly Formula 3

    Apr 16, 2004
    1,330
    SF, Bay Area
    Just spoke with FoSF and apparently this is a known problem. There are four knock sensors on the car and only two of them showing the fault in my case. After talking with FoSF's shop foreman, he confirmed that gas couldn’t be the culprit. There should not be a knock sensor fault if there is detonation. The ecu examines the output of the knock sensor and retards the timing to alleviate the detonation. He wasn't sure about wiring diagram but said either the ground or power is shared. He noted that it is possible either the sensor is shorted or there is an issue with the common ground or power. Unfortunately for me they are short handed due to Monterey event and won't be able to look at it until next week. I hope I can get it fixed else where on Friday or Saturday before the track event.
     
  9. 4i2fly

    4i2fly Formula 3

    Apr 16, 2004
    1,330
    SF, Bay Area
    FoSF has been working on the car since last week and yet to identify the root cause. Apparently the knock sensors were not the issue so they advised me of possible bad ground in the control system. Having read so many threads with 360s having bad ground wirings, I wasn't a bit surprised of the suggestion. After almost a week of trouble-shooting they are going to take the rear bumper out and examine the CAT(s). They have a theory that the CAT(s) may have gotten clogged and hence lack of power at higher rpms. The theory is, since I mix 91/100 octane in the car the CAT(s) may have burnt and clogged due to the effect of high-octane gas on exhaust valve timing. Any input to make sense out of this issue would be greatly appreciated.
     
  10. ferrarifixer

    ferrarifixer F1 Veteran
    BANNED

    Jul 22, 2003
    8,520
    Melbourne
    Full Name:
    Phil Hughes
    Not so sure. I haven't had any dealings with road car knock sensors yet, but the race cars, particularly the N-GT, always flag knock sensors when fuel quality is an issue.

    I think the problem lies in the terminology of the computer. Although it says knock sensor "fault", I think it's just telling you that they're working too much.

    I think the 91 is the problem. When the race cars have done it, I clear the problem, use fresh 98 RON and try again. Has worked every time so far.

    In the N-GT if you use ELF 102 RON WRF fuel (Rally cars) it goes the same, feels no different at all, but flags the knock sensors. If you use ELF 102 RON LMS (Le-Mans) fuel, it works perfectly at all times.
     
  11. thomas_b

    thomas_b Formula Junior

    Sep 15, 2003
    765
    the only other scenario I could come up with is that something is loose and hits the engine block or creates vibrations - the knock sensor basically picks up vibrations from the engine block - successive processing steps filter the frequency spectrum and are looking for a characteristic pattern, which identifies the presence of knocking

    now if the knock sensor fault indicates that the occurrence of knocking is above the max. threshold as Phil says than I would look for some rattle, loose stuff, etc.
     
  12. To remove this ad click here.

  13. 4i2fly

    4i2fly Formula 3

    Apr 16, 2004
    1,330
    SF, Bay Area
    I did try the higher octane and didn't help. That's when I realized there is some other issue tripping the knock sensor.

    This is a good point and I don't think it can be ruled out, since they have checked everything and all electronics seem to function properly. I guess I should have asked them to bypass the knock sensors and see if the power gets restored (I don't know if this is safe?) It is interesting if "any" vibration could trigger the knock sensors to show a fault... What I mean by "any", obviously the input frequency (bandpass filter) of the detection circuitry to pick up the knock vibration is at a certain range and for a random vibration to fall in its range is pretty coincidental. And at the end it is the amplitude of the signal that trips the switch for knock detection also another coincident. And why only knock 1 & 2 showing the fault? Something must be happening in one cylinder bank that is not happening on the other. I may be missing something but years ago I used to develop knock circuit test as part of an engine controller for Ford Motor cars in automatic testers and the detection skim is not that complicated. It just makes me wonder, nothing other than a vibration or electrical fault (ground/power) could not trip the sensors. How about the ecu itself, and detecting knock when there isn’t one present? I am not sure if they have swapped the ecu around to see if the fault follows. Thanks for the inputs, please reply if you can think of anything else...
     

Share This Page