Connecting OE knock sensor to stereo / laptop

Discussion in 'Throttle bodies & non-OEM ECUs' started by alexisblades99, Jul 16, 2010.

  1. alexisblades99 Forum Member

    Joined:
    Jun 6, 2009
    Likes Received:
    1
    Location:
    sunny helsinki
    I'm going to be starting mapping on microsquirt fairly soon, and I want to be able to hear detonation if it's happening. I have the engine's original knock sensor, which I gather is a microphone tuned into the frequency of knock for that engine.

    So far, I was planning on connecting it up to a pre-amp stage, then on to the stereo, or if I'm feeling clever maybe the laptop for frequency analysis. Does anyone have any thoughts/helpful tips on this?

    Ta.
     
  2. Trev16v

    Trev16v Paid Member Paid Member

    Joined:
    Oct 22, 2003
    Likes Received:
    451
    Location:
    Oxfordshireland
    Yep, done a lot of this myself.

    First of all you could try connecting the knock sensor directly to the microphone input of the laptop, just to see if you get good enough results. The output level of the knock sensor can be pretty high (hundreds of mV) and the laptop microphone input should be high enough impedance for it. The laptop itself could then act as the buffer / amplifier for your headphones.

    To listen through headphones without the need for your laptop, use something like the Whisper 2000 with the little microphone insert removed, and instead connect the knock sensor via some coax. This is what I use.

    The only problem you'll find is that ignition noise is deafening. You might want to experiment with grounding the coax shield at different points to minimise the ignition noise. The best solution is probably to build a little differential amplifier.

    I used a couple of audio applications on the PC to do FFT on the recordings. On my engine, knock manifested itself at about 6kHz and 16kHz. As well as filtering, commercial knock sensing also tends to use 'windowing' so that knock is only listened for over a given range of angles.

    If you give me a bit of time I could post some small .mp3 / .wav snippets of actual knock sounds from recordings I did with a mate on his Polo, which could be made to pink quite easily!

    PS. Do you still have that G60 Syncro downpipe...?
     
  3. Mike_H Forum Addict

    Joined:
    Oct 16, 2004
    Likes Received:
    17
    Location:
    iQuit
    Shouldn't this be in the tunes section. Sounds a lot like hard house to me :)
     
  4. alexisblades99 Forum Member

    Joined:
    Jun 6, 2009
    Likes Received:
    1
    Location:
    sunny helsinki
    :lol: German industrial techno at its most raw!

    Trev, that would be wicked. I've had a listen to a couple of sound files already, but from a volvo engine. I'm guessing it'll sound different between engines, and yours will be closer to how mine will sound. I have a phono pre-amp which I've been lumbering myself with for the last ten years so maybe it finally has a purpose! Shielding the signal wires though as you say, is a must, so I'll run some extra wires into my loom, which has braided shielding for the sensor wires (I still have the downpipe).
     
  5. HidRo Forum Member

    Joined:
    Jan 30, 2004
    Likes Received:
    12
    Location:
    Portugal
  6. Trev16v

    Trev16v Paid Member Paid Member

    Joined:
    Oct 22, 2003
    Likes Received:
    451
    Location:
    Oxfordshireland
    The recordings I did were actually from a 16V Polo. I need to search through my backup DVDs to find the complete recording. The only thing I can find on my laptop at the moment is this really quick snippet of knocking:

    http://www.introspectiv.eclipse.co.uk/knock/polo_ping_at_11sec_8nov08.wav

    That's about a one second long little piece taken from the recording we did. Open it up in Audacity or Audition, highlight the ringing knock bit, and do an FFT. See peaks at around 6kHz and just below 15kHz. The reason why you can hear the knock occur so loudly compared to the rest of the background noise is - I think - because this is a knock that was produced by making the car move forward in 4th or 5th from standstill with the engine at idle.

    I'll ensure I find the whole lot and stick more of it up.
     
  7. Dave

    Dave *Very Smart* Pedantic Old Fart Paid Member

    Joined:
    Feb 26, 2007
    Likes Received:
    473
    Location:
    Sligo, Eire
    Gosh, you are clever lads. I always reconed that Elec. Eng. were thicker than Mech. Eng.:lol:
     
  8. Trev16v

    Trev16v Paid Member Paid Member

    Joined:
    Oct 22, 2003
    Likes Received:
    451
    Location:
    Oxfordshireland
    Bah. Yeah. Knock sensors, amplifiers, FFT, etc. Let's see you do a pencil diagram of that. :lol:
     
  9. Dave

    Dave *Very Smart* Pedantic Old Fart Paid Member

    Joined:
    Feb 26, 2007
    Likes Received:
    473
    Location:
    Sligo, Eire
    FFT. Furrier Transformations?

    Can't spell!
     
  10. Toyotec

    Toyotec CGTI Committee - Happy helper at large Admin

    Joined:
    Jul 26, 2006
    Likes Received:
    3,324
    Location:
    Creating Pfredstarke
    = Fast Fourier transform (FFT) :thumbup:.
     
  11. HidRo Forum Member

    Joined:
    Jan 30, 2004
    Likes Received:
    12
    Location:
    Portugal
    So you won''t be able to hear this with that DET CAN I posted?
    This is some good info :thumbup:
     

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice