Results 1 to 9 of 9

Thread: How accurate are cam sensors?

  1. #1
    Senior Member
    Join Date
    Aug 2013
    Location
    vodka drinking country
    Posts
    158

    How accurate are cam sensors?

    I am having an odd problem with my newly built engine
    I did not degree it, but used the eccentric timing belt pulleys to rotate the cam pulleys so that all marks line up ideally to compensate for machined deck/heads/non-OEM HG

    now, in my autronic I can see parked cam positions with the engine idling, and one shows 130 degrees which is exactly what the cam center line is, however the other side is showing 120 degrees which is a bigger difference than I am used seeing. I made sure I didn't skip a tooth, all is fine. I did have to modify 255 heads to work with 207 cams by slightly moving the rear cam sensors so they pick up the cam signal

    so at this point, I am not sure what to trust - the cam pulleys marks, which line up, or the software which is telling me my driver side is advanced by 10 degrees in static position

  2. #2
    Senior Member TimH's Avatar
    Join Date
    Nov 2009
    Location
    Wiring looms 'r' us
    Posts
    1,753
    I'm not a mapper, nor do I have any experience with Autronic, but the cam sensor will be producing a waveform and the ECU will be set to detect at a certain voltage level.

    If not set right then an error in the reading seems feasible to me, and maybe moving a sensor could change the voltage the ECU sees? I'm assuming the 2.0l 2-wire/VR sensors rather than 2.5l 3-wire/HE types, since the hall effect are "on/off" rather than producing a sinewave that has to be sampled.

    I wonder if its worth swapping the connections to the left and right sensors to see if the error follows the wiring or stays with the particular head?
    aka JT Innovations www.jti.uk.com No scoob any more Outlander PHEV GX4h for us now !!

  3. #3
    Senior Member
    Join Date
    Dec 2001
    Location
    Norn Iron
    Posts
    4,271
    A sensor is just a sensor, it will report what it sees, and what is then reported to you will depend on how the ecu is configured

    Of course it would be helpful if you mentioned which engine ? presumably it must be an AVCS setup of some sort as you mention sensors on both cams ? Then you mention adjusting cam pulleys in some order ?

    If the ecu is configured correctly, then it should be reporting where the cams are...and the AVCS position should be meeting target. Is it ? If not then you may need to adjust base settings so it does meet target.

    IMO do not start moving sensors...although that in itself would be very difficult.
    speed didnt kill me, but taxation probably will

  4. #4
    Senior Member
    Join Date
    Aug 2013
    Location
    vodka drinking country
    Posts
    158
    I will elaborate

    in my last build I am using 2.5 block, mated with D25 heads (found on WRX and Foresters)
    Because my car wiring was originally made to work with 5 pin 16 bit ECU (JDM STI AVCS), I didn't feel like converting it, and modified the heads - I had to move the sensor closer in so that it reads the cams correctly, and I had to get the cam angle sensor adapter that goes in front of the left head and reads the cam gear








    This worked well on previous builds, I had some cam position imperfection due to decked heads/block which alters the belt length, but I compensated that by using LIC eccentric pulleys to set up the cam gears in such fashion so that they timing marks align perfectly. Even with that, I had slight cam offset from where it needs to be by 2-4 degrees which I can live with. However on the new build even with the cam gears timing marks aligned I am off by 10 degrees on one side, I am running new cams though so each cam varies.

    I've talked to a credible engine builder, and he told me that when he degrees the cams the timing marks on the cam gears might not align perfectly, so I got my answer. Now I need to find a way to degree my cams with the engine in the car, gotta pay for being lazy lol

  5. #5
    Senior Member
    Join Date
    Dec 2001
    Location
    Norn Iron
    Posts
    4,271
    Still unclear...so this is or isnt using AVCS ?

    If so, in many ways ultimately actual position is less important, than both cams being in the same position. If the ecu is configured correctly then what those rear position sensors are seeing...is what you need to be working towards so both are the same. If they are not the same then you need to alter the AVCS control solenoid for whichever side you believe to be incorrect until both rear sensors report the actual camshaft is where it should be.

    Of course the above assumes the AVCS and cam gears are working and not broken too as to why one cam may appear in a different position.

    Whilst I havent done it myself, one way of finding valve events with the engine in car, using a Pico and one of their pressure transducers is a running compression test...it can give very detailed info of when valves are opening and closing...although it'd be cheaper to remove the engine and do it manually lol

    https://www.youtube.com/watch?v=Kwy72aaAFvo&

    No doubt though, there is a way to correlate your cam sensor readings to crank position for a check, have you asked on the Autronic forum ? I only ever used Autronic once years ago and hated it. Software was a nightmare.
    speed didnt kill me, but taxation probably will

  6. #6
    Senior Member
    Join Date
    Aug 2013
    Location
    vodka drinking country
    Posts
    158
    yes, intake AVCS

    going to degree the cams, we'll see what I come out with

  7. #7
    Senior Member
    Join Date
    Aug 2013
    Location
    vodka drinking country
    Posts
    158
    I degreed intake cams today and I can say the cam sensors are lying by 3-5 degrees, at least in my specific custom application

  8. #8
    Senior Member TimH's Avatar
    Join Date
    Nov 2009
    Location
    Wiring looms 'r' us
    Posts
    1,753
    You're using the 2 pin VR sensors, as per 2.0l engines. These produce a waveform, the amplitude of which is related to the proximity of the sensor to the teeth, as well as rotational velocity: the waveform will approximate the shape of the teeth as they pass the sensor, with an amplitude proportional to speed. If the heads are machined the same, and the sensors mounted identically, with the same clearance to the teeth, and both sides are calibrated the same in the ECU then they will read the same.

    If any of those things differ between the two sides, there is a good chance you will get different readings.

    Did you try swapping the wiring between the sides? That would assist in proving/disproving my theory!
    aka JT Innovations www.jti.uk.com No scoob any more Outlander PHEV GX4h for us now !!

  9. #9
    Senior Member
    Join Date
    Aug 2013
    Location
    vodka drinking country
    Posts
    158
    no need, since I degreed the cams, so I just trimmed cam offset in Autronic so it reads the centerline correctly

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •