Knock system

I think the next thing to do is hook the scope up to the knock sensor on the car and try and grab a snap shot of the sensor at WOT.
I think the sensor puts out a high enough voltage to rule out EMI on the car, but that would be the purpose of putting the scope on it.
 
You may also need to record the normal frequencies your engine makes through the rpm range with the knock sensor disconnected. There a bunch of noises the stock engine doesn't make with roller lifters, rockers and timing chains. I imagine at some point they fall in the "danger zone" and fire off the signal to the ECM.
 
I'd be comparing the fueling versus the KR to determine if the KR is real...or not. I've seen a crap load of PL files where the knock sensor was 'hearing' something.
 
You can see everything with the spectrum anaylzer, the problem is I do not know how to translate the db/sound power of the recording level to a microphone voltage. Even though the sensor resonates around 6K hz, and the ESC is tuned around 6K hz, it may not take a large spike off frequency to set it off. Or the normal variation in the 6K spike could also set it off. That is what I think I want to try to use the scope for while hooked to the sensor when the car is running. I don't have any programs to record my laptop, but you play the spectrum analyzer and watch variation in the spectrum during the WOT run. To get a better idea of the off 6Khz frequency I would have to install a newer style wide band knock sensor and record that. So short answer is, I think I have already done what you are recommending, with what I have accept for hooking the scope up to the car.
 
I'd be comparing the fueling versus the KR to determine if the KR is real...or not. I've seen a crap load of PL files where the knock sensor was 'hearing' something.
Fueling is solid, timing is low. Adjustments to both have zero effect. Would get the exact same result from 20lbs of boost to 29lbs of boost. I have to check again, but never any signs on plugs, It has always picked up some minor indications. I am confident that it is just being peaked by mechanical noise, I am gonna run this one to ground one way or the other.
 
You can see everything with the spectrum anaylzer, the problem is I do not know how to translate the db/sound power of the recording level to a microphone voltage. Even though the sensor resonates around 6K hz, and the ESC is tuned around 6K hz, it may not take a large spike off frequency to set it off. Or the normal variation in the 6K spike could also set it off. That is what I think I want to try to use the scope for while hooked to the sensor when the car is running. I don't have any programs to record my laptop, but you play the spectrum analyzer and watch variation in the spectrum during the WOT run. To get a better idea of the off 6Khz frequency I would have to install a newer style wide band knock sensor and record that. So short answer is, I think I have already done what you are recommending, with what I have accept for hooking the scope up to the car.
Actually, if I hunted down the specs for the knock sensor, I may be able to convert db to voltage. Would take some digging and homework though, not sure I feel like doing that exercise if I can easily hook the scope up.
 
Another point is the frequency generator I have tops out at 10v. I can't test if the ESC can be triggered by a steady state signal at some voltage threshold higher than that.
 
Can you subject the knock sensor screwed into a block on a stand and determine what makes that frequency ie; tapping the block?
 
Can you subject the knock sensor screwed into a block on a stand and determine what makes that frequency ie; tapping the block?
My guess is no(see last sentence though). Tapping the block, and I am speculating to a degree, would look like a broadband stimulus. So, if you hit it hard enough, the sound power level or acceleration of the knock sensor will exceed the threshold of the ESC at some frequency. The chart I posted shows that the ESC is most sensitive at the tuned freq/6k hz-ish as expected, but a larger impulse at a different freq will set it off. The combustion noise, piston/cylinder noise, detonation noise all occur at the same frequency where the ESC/knock sensor system is most sensitive. The cylinder is most likely similar to an organ pipe which is why the knock frequency can be picked up and discriminated. It is also why piston slap sets of the knock sensor. Other noisy components and their frequency spectrum is the issue. Tolerance of the factory electronics may play a part. Sudden changes at various freqs will set off the ESC if large enough. There is some expensive knock listening equipment out there, and if I did this for a living, I would think about getting it. I am also thinking about the J&S safeguard. What I want is the ability to record and analyze. I am not running conservative boost, but everything else is conservative, and I really like where the car is. I am not looking to take it further, just looking to say I am good where I am, and maybe to figure out how to do some simple mods to the knock system. i.e. maybe a simple voltage divider to attenuate the knock sensor. Now all of that is not to say that maybe some good data cannot be collected from an engine block on a stand with a knock sensor.
 
How about a Chassis Ears? You can stick it to the block and drive with headphones on and listen for knock. I've watched a few videos on tuning and on a dyno, they purposely induced knock with headphones to hear what it sounds like. Maybe even record it using the microphone input on a cassette player instead of the headphones. To me it sounded like a sizzling type of noise. It didn't sound like the steel ball in a can type sound they you hear with too high timing on a regular engine.
 
I essentially did the same thing by recording the sound with the laptop and the audacity freeware. I can play back and listen while also looking at the frequency spectrum. I listened to a couple on line recordings to get an idea of what the knock sounded like. I did not hear anything engine noise wise that sounded like knock. The audacity freeware lets you speed up, slow down, filter. The thing that I don't want to do is purposely induce knock, I guess if I wanted to turn it way down and then try, but still, that makes me a little nervous. That though would be a sure way to see what it sounds like on my car and see what the signature looked like with the freeware.
 
In the zip file is an MP3 file a run, it is 3rd gear WOT. Had to zip it because board won't allow mp3 extension to post.
 

Attachments

  • 0 bytes · Views: 55
Very good testing there!
This is the video I was refering too.
In your sound clip I don't hear that snapping.
 
Whoops I already posted that video. One advantage of more advanced ECMs is you can zero in on a specific area to work on vs with a regular chip you have wide swaths that you can change.
 
Whoops I already posted that video. One advantage of more advanced ECMs is you can zero in on a specific area to work on vs with a regular chip you have wide swaths that you can change.
Yes, I checked that out the first time and watched again, plus a couple others from this guy. The plex system is nice, if I did dyno tuning for a living that is the way to go.
 
It's not the noise level. The knock sensor measures a certain frequency that has been determined from the factory. Anything that happens to be at that frequency near the knock sensor can cause the signal to be picked up and sent to the ESC. The strength and duration of that frequency is the determining factor.

Electrical issues can also mess with the signal and is what I would be looking for in this situation.
What spark plugs? Non resistor?

It’s obviously not real knock.
 
Electrical issues can also mess with the signal and is what I would be looking for in this situation.
What spark plugs? Non resistor?

It’s obviously not real knock.
I use autolite resistor type, can't remember the number. Kirbans replacement plug wires. I have thought about EMI, and you maybe right. The system is resistant to EMi due to the relitively high voltage the single wire sensor generates. Running a dedicated shielded wire to test would be a bit of a PIA, but I have thought about it.
 
The weather here is deteriorating quickly, so I don't expect to get a whole lot done this year. My plan is to get things set up so I can hook the oscilloscope up to the car and record some WOT runs. I will want to record sensor and ESC output. I think I am going to get a simple screen recorder to grab the scope output. I know better than to say stuff with certainty, but I am confident that I am not knocking.

Like I mentioned earlier, I am also seriously considering a J&S Safeguard. By all accounts this is a quality product. I have no plans to push timing to the knock limit, it would be nice to have a quality product to know that I am not knocking, and it could possibly react quick enough to pull timing if a malfunction occurred.
 
Top