Resonant knock sensors with internal module

General support questions and announcements for MS3. See also MS3 manuals.

Moderators: jsmcortina, muythaibxr

Post Reply
ravensierra
Helpful MS/Extra'er
Posts: 46
Joined: Sat Apr 19, 2014 10:24 pm
Location: Chch NZ

Resonant knock sensors with internal module

Post by ravensierra »

Hi,

I've recently installed the internal knock module in my MS3X, and am hoping to use the stock resonant knock sensor for tuning.

(I also have a condensor mic bolted to the sensor bracket which I amplify with my laptop for audio knock detection)

However, I can't find any reliable information on the resonant frequency of the sensor (engine is a 6G72 Mitsubishi 3.0L V6). The best answer I could find was 11.6kHz from an unreliable source, which is far away from the theoretical resonance of the block at ~6.3kHz, or even the first overtone at ~12.6kHz.

Empirically testing the sensor for frequency response is probably impractical at this stage, as it's buried in the V of the block under the intake manifold... I'm surprised the settings for the module don't allow for the bandpass filter to be disabled entirely for use with resonant sensors?

What is the best approach here? Perhaps induce detonation at idle and experiment with a variety of bandpass frequencies until I get maximum voltage from the sensor?

Any opinions or advice appreciated. :) Thanks,

Jack
1991 Mitsubishi GTO twin turbo
ravensierra
Helpful MS/Extra'er
Posts: 46
Joined: Sat Apr 19, 2014 10:24 pm
Location: Chch NZ

Re: Resonant knock sensors with internal module

Post by ravensierra »

Have been told I will need to operate at higher load than idle to induce knock?
1991 Mitsubishi GTO twin turbo
dontz125
Super MS/Extra'er
Posts: 4224
Joined: Mon May 11, 2009 7:14 pm
Location: York, ON
Contact:

Re: Resonant knock sensors with internal module

Post by dontz125 »

If your engine knocks at idle, something is seriously wrong! Knock is generally encountered while either flogging the crap out of it in the search for more power, OR by lugging it uphill at high throttle but low (~ 1/3 max) revs.
Temporarily shut down - back soon!
QuadraMAP Sensor Module -- PWM-to-Stepper Controller -- Dual Coil Driver
Coming soon: OctoMAP Sensor Module
TTR Ignition Systems
ravensierra
Helpful MS/Extra'er
Posts: 46
Joined: Sat Apr 19, 2014 10:24 pm
Location: Chch NZ

Re: Resonant knock sensors with internal module

Post by ravensierra »

I know that, I will need to CAUSE knock deliberately in order to test this sensor for its resonant frequency, as I don't have an oscilloscope/tone generator/etc.

I am forced to run a bandpass filter by the MS3.

I can either

a) Record the sensor output using full spectrum audio detection through the microphone on the block and my laptop soundcard, and run a frequency analysis to see the frequencies associated with knock, and set the bandpass accordingly.

or

b) Experiment with different bandpass settings on the MS to see which filter setting allows the knock signal through, in combination with my audio setup for detection of knock.

I would like to know please, what is the best way to safely cause knock for this purpose, whether it be advancing timing so I can cause it at lowish load, or heading up the closest hill in 5th gear.

Thanks in advance

Jack
1991 Mitsubishi GTO twin turbo
Yves
Super MS/Extra'er
Posts: 1505
Joined: Tue Jun 17, 2014 1:23 pm
Location: Belgium

Re: Resonant knock sensors with internal module

Post by Yves »

Isn't the knock frequency based on the cylinder bore ?
ravensierra
Helpful MS/Extra'er
Posts: 46
Joined: Sat Apr 19, 2014 10:24 pm
Location: Chch NZ

Re: Resonant knock sensors with internal module

Post by ravensierra »

Yves wrote:Isn't the knock frequency based on the cylinder bore ?
Roughly, but if knock occurs at say 5* ATDC, how much of the environment for pressure waves is cylinder bore, and how much is combustion chamber?

Anyway, using the formula f=900/(pi*cylinder radius) for my 91.1mm bore gives a resonant frequency of 6289Hz.

The ONLY piece of information I have about the resonant frequency of the knock sensor in this engine says that it is resonant at 11.6kHz, however this comes from a guy on the internet who has proven himself multiple times to be a complete t**ser.

So rather than plug 11.6kHz into the bandpass filter and assume everything is working as it should, at risk of cracking a ring land or two, I'd rather get some hard evidence in my own hands.

Now it is likely that the sensor is tuned to the first overtone of the true resonant frequency of the bore/combustion chamber, which would be 12.6kHz. So I could set the bandpass filter to that, and just ensure that I get an acceptable signal through to the MS when I do encounter knock.

I appreciate if anyone has reservations about telling me how to do something which could be potentially damaging to my engine, but I am fully aware of risks involved.

Thanks,

Jack
1991 Mitsubishi GTO twin turbo
Yves
Super MS/Extra'er
Posts: 1505
Joined: Tue Jun 17, 2014 1:23 pm
Location: Belgium

Re: Resonant knock sensors with internal module

Post by Yves »

If the knock sensor is indicating knock right now with a given type of fuel, I would get me some high octane race fuel and see if the knock disappears. That would be a sure indication that the knock you were seeing is real knock.
ravensierra
Helpful MS/Extra'er
Posts: 46
Joined: Sat Apr 19, 2014 10:24 pm
Location: Chch NZ

Re: Resonant knock sensors with internal module

Post by ravensierra »

I have no idea if I have knock right now, I have yet to decide what to do about this bandpass filter situation and don't make a habit of driving around with my headphones on.

I guess I'll just set it to 11.6kHz and thrash my car silly on some 95RON and make sure I get some knock counts through.
1991 Mitsubishi GTO twin turbo
Post Reply