Jump to content
SAU Community

Recommended Posts


Oh, so the voltage being displayed is via a cheap volt meter wired in?

my suggestion, and I could be wrong and don't want to offend you, is that may be wired into the wrong pins. Don't take that as a negative or having a go, sometimes going back to the start / basics is what has to happen. The whole 2.4 volts thing has confused everyone except you, because you have seen it and did it. Everyone else's knowledge of a knock sensor doesn't match your experience, and that is why you are getting a bit cranky..

But a knock sensor is a simple thing, understood by many, that doesn't put out a voltage. So something is up, which has impacted this thread in a big way. Can you confirm what pins you connected to?

 

Also, proper terms seriously matter, you can't just use different terminology that may be conflicting or not make sense then tell people to just accept it and move on.

 

  • Like 1

Thank you for being decent this post ben.

I will check the pins again and will check the wires for any previous modifications.

Yes terminology. Im not great with it and does me no favours when explaining to people who know the exact terms.
Why i stated i wont be using it and that you guys are smart enough to decifer or re-correct me.

I do remember reading about it back in the day.
Just cant remember enough.

Ill post back up the pins there on.

You need an oscilloscope not a voltmetre to determine knock, a knock sensor behaves very similar to a microphone.

Like I said in a previous post, if you test the voltage on the knock sensor pin without the sensor connected you'll detect a voltage when using a stock ECU.

  • Like 1

Ok the thing is dosepipe. I know that.
I just couldnt get across that i cant and havent - and i have volts.
I remember reading that it was useless but id already wires it up.

The wires seem correct pins 23 and 24.
Off white almost cream wires as in the pic.

Geebs i havent dove in there for a while hecticly taped and messy. Needs a tidy up.
Ahahaha dont get sus i know theres wires everywhere i did most of em , and i just broke one with the tape lol. Easy fix.

Former owner did safc as standard install with gtr injectors when i bought it back in 2013. Ok for e85 hahaha.
There is also some hacking from b4 2.

What is thar brown plug with all the wires. Is it to do with the wire shielding?20190304_190925.jpeg20190304_193549.jpeg20190304_185301.jpeg20190304_191005.jpeg

This is the most pointless exercise ever, and i love it. 

I couldn't have even dreamed of anyone trying to measure knock with a jaycar multimeter.

Also love your 20/20 hindsight knowledge. Everything anyone posts you 'already knew'.

Good luck figuring it out!

  • Like 5
  • Haha 1

holy f**k Slap you did not take my advice.

the knock sensor is a microphone. it generates an AC wave. that is the sound of the motor. taller wave = louder. spacing of the waves = frequency. its not a DC voltage. a sound wave can have many many many frequencies. it is continuous, not discreet. you only care about noises in the frequency range the engine vibrates at when it knocks.

a decent ecu will filter that noise to a particular "band" of frequencies (say 6-7khz). the peak Amplitude of that signal is your knock level. it can be in volts, it can be in ADC counts. if your engine is not knocking it will be quiet in this range. LOW VOLTAGE. if it IS knocking it will be loud in this range HIGH VOLTAGE.

a shit ecu will not filter the frequency, which is ok if your sensor is resonant to your engine knock frequency. but for a wideband sensor.

the voltage you are reading will be the PEAK voltage seen by the ecu, and probably have a peak/hold circuit that gets reset after a delay. now for this to be useful, you need to know:

is your sensor wideband or resonant?

does your ecu do bandpass filtering on that signal?

the voltage you are looking at is an indicator of the VOLUME of the noise the knock sensor (microphone) is picking up. that is NOT usefull if you have the following:

wideband sensor with no bandpass filter (this will be all the engine noise)

this IS useful under the following:

wideband sensor with bandpass (this is knock frequency only)

resonant sensor with or without bandpass (if sensor is resonant to the frequency your engine knocks at)

if you want to know how LOUD the signal is, you can use a peak/hold multimeter. this will just tell you maximum volume, it could be at 100hz or 15khz. it won't tell you about knock.

if you want to know what FREQUENCY the knock is occuring at you will need a scope. you need both the frequency and amplitude of the signal to work out if its actually likely to be knock or just loud noises at frequencies you don't care about.

 

to understand what is happening you really need to plug the knock sensor into a usb soundcard and use some spectrogram/ real time fourier (frequency) graph software. something like this: https://digh.com/product/deka/

 

some free software I use is this which works great.

http://www.sdradio.eu/weaksignals/spectran.html

 

 

 

 

Edited by burn4005
  • Like 1

* Waits for "I know this* Response then reply asking about why there is voltage there *

I am not sure how this is taking so long to understand, especially with some smart people giving some smart answers.
I feel like you have backed yourself into a corner and now just want to attack everyone saying your wrong rather then admitting that you made a mistake and you now have been given some knowledge.

If you really want to get some information out of the knock sensors, something like the below may help:

Ebay Oscilloscope

  • Like 1
Quote

So you are saying the voltage is manipulated to bend in amplitude according to vibration causing it to become a frequency. Not changing volts.

Would the voltage displayed be the threshhold in wich the frequency works at.

why does the voltage drop on e85? Less noise? Less frequency needed? Magically? or im full of shit and it doesnt wich makes no sense to my eyes that have witnessed it ?

Anolog is as simple as a boost switch or bleed valve.
And i know it needs a digital tune to make the most. Geebz nearly any mod manualy controled or manipulated is anolog tuning.
The anolog gauges are just a unique trend . 

this is one of the dumbest f**king things I've ever read.

the only thing that might make sense here is you are actually Japanese and are running all your answers through google translate into english. 

the reason you are being attacked is you have been unable to convey any coherent concept whatsoever but are still somehow attempting to claim the intellectual high ground here.

Edited by burn4005
  • Like 1
  • Haha 1
24 minutes ago, burn4005 said:

the only thing that might make sense here is you are actually Japanese and are running all your answers through google translate into english.

Maybe you should run it back and forth through google translate between Japanese and English until it starts to make sense.

  • Haha 1
holy f**k Slap you did not take my advice.
the knock sensor is a microphone. it generates an AC wave. that is the sound of the motor. taller wave = louder. spacing of the waves = frequency. its not a DC voltage. a sound wave can have many many many frequencies. it is continuous, not discreet. you only care about noises in the frequency range the engine vibrates at when it knocks.
a decent ecu will filter that noise to a particular "band" of frequencies (say 6-7khz). the peak Amplitude of that signal is your knock level. it can be in volts, it can be in ADC counts. if your engine is not knocking it will be quiet in this range. LOW VOLTAGE. if it IS knocking it will be loud in this range HIGH VOLTAGE.
a shit ecu will not filter the frequency, which is ok if your sensor is resonant to your engine knock frequency. but for a wideband sensor.
the voltage you are reading will be the PEAK voltage seen by the ecu, and probably have a peak/hold circuit that gets reset after a delay. now for this to be useful, you need to know:
is your sensor wideband or resonant?
does your ecu do bandpass filtering on that signal?
the voltage you are looking at is an indicator of the VOLUME of the noise the knock sensor (microphone) is picking up. that is NOT usefull if you have the following:
wideband sensor with no bandpass filter (this will be all the engine noise)
this IS useful under the following:
wideband sensor with bandpass (this is knock frequency only)
resonant sensor with or without bandpass (if sensor is resonant to the frequency your engine knocks at)
if you want to know how LOUD the signal is, you can use a peak/hold multimeter. this will just tell you maximum volume, it could be at 100hz or 15khz. it won't tell you about knock.
if you want to know what FREQUENCY the knock is occuring at you will need a scope. you need both the frequency and amplitude of the signal to work out if its actually likely to be knock or just loud noises at frequencies you don't care about.
 
to understand what is happening you really need to plug the knock sensor into a usb soundcard and use some spectrogram/ real time fourier (frequency) graph software. something like this: https://digh.com/product/deka/
 
some free software I use is this which works great.
http://www.sdradio.eu/weaksignals/spectran.html
 
 
 
 
Voltage explained!

And the rest of you trolls spittin shid like your skyline preists can go lik an arse.

You couldnt answer it and thats that . You went on about everything else making me confused as shid.
And maybe when ur intelectual abilities extend beyond the bible of tuning and mechanics you might actualy understand what i said , where i was confused , why it was said, and why a lot of what you have posted is erelevent to not only the point but to me. Because i dont dyno tune or have sufficient consult equiptment.
As i stated i know it isnt usefull and asked why i see it. Wich atleast was answered by someone with intelect and wisdom.
Not a load of shid about its impossibleness.

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



  • Similar Content

  • Latest Posts

    • Thanks for the reply mate. Well I really hope its a hose then not engine out job
    • But.... the reason I want to run a 60 weight is so at 125C it has the same viscosity as a 40 weight at 100C. That's the whole reason. If the viscosity changes that much to drop oil pressure from 73psi to 36psi then that's another reason I should be running an oil that mimics the 40 weight at 100C. I have datalogs from the dyno with the oil pressure hitting 73psi at full throttle/high RPM. At the dyno the oil temp was around 100-105C. The pump has a 70psi internal relief spring. It will never go/can't go above 70psi. The GM recommendation of 6psi per 1000rpm is well under that... The oil sensor for logging in LS's is at the valley plate at the back of  the block/rear of where the heads are near the firewall. It's also where the knock sensors are which are notable for 'false knock'. I'm hoping I just didn't have enough oil up top causing some chatter instead of rods being sad (big hopium/copium I know) LS's definitely heat up the oil more than RB's do, the stock vettes for example will hit 300F(150C) in a lap or two and happily track for years and years. This is the same oil cooler that I had when I was in RB land, being the Setrab 25 row oil cooler HEL thing. I did think about putting a fan in there to pull air out more, though I don't know if that will actually help in huge load situations with lots of speed. I think when I had the auto cooler. The leak is where the block runs to the oil cooler lines, the OEM/Dash oil pressure sender is connected at that junction and is what broke. I'm actually quite curious to see how much oil in total capacity is actually left in the engine. As it currently stands I'm waiting on that bush to adapt the sender to it. The sump is still full (?) of oil and the lines and accusump have been drained, but the filter and block are off. I suspect there's maybe less than 1/2 the total capacity there should be in there. I have noticed in the past that topping up oil has improved oil pressure, as reported by the dash sensor. This is all extremely sketchy hence wanting to get it sorted out lol.
    • I neglected to respond to this previously. Get it up to 100 psi, and then you'll be OK.
    • I agree with everything else, except (and I'm rethinking this as it wasn't setup how my brain first though) if the sensor is at the end of a hose which is how it has been recommended to isolate it from vibrations, then if that line had a small hole in, I could foresee potentially (not a fluid dynamic specialist) the ability for it to see a lower pressure at the sensor. But thinking through, said sensor was in the actual block, HOWEVER it was also the sensor itself that broke, so oil pressure may not have been fully reaching the sensor still. So I'm still in my same theory.   However, I 100% would be saying COOL THE OIL DOWN if it's at 125c. That would be an epic concern of mine.   Im now thinking as you did Brad that the knock detection is likely due to the bearings giving a bit more noise as pressure dropped away. Kinkstah, drop your oil, and get a sample of it (as you're draining it) and send it off for analysis.
    • I myself AM TOTALLY UNPREPARED TO BELIEVE that the load is higher on the track than on the dyno. If it is not happening on the dyno, I cannot see it happening on the track. The difference you are seeing is because it is hot on the track, and I am pretty sure your tuner is not belting the crap out of it on teh dyno when it starts to get hot. The only way that being hot on the track can lead to real ping, that I can think of, is if you are getting more oil (from mist in the inlet tract, or going up past the oil control rings) reducing the effective octane rating of the fuel and causing ping that way. Yeah, nah. Look at this graph which I will helpfully show you zoomed back in. As an engineer, I look at the difference in viscocity at (in your case, 125°C) and say "they're all the same number". Even though those lines are not completely collapsed down onto each other, the oil grades you are talking about (40, 50 and 60) are teh top three lines (150, 220 and 320) and as far as I am concerned, there is not enough difference between them at that temperature to be meaningful. The viscosity of 60 at 125°C is teh same as 40 at 100°C. You should not operate it under high load at high temperature. That is purely because the only way they can achieve their emissions numbers is with thin-arse oil in it, so they have to tell you to put thin oil in it for the street. They know that no-one can drive the car & engine hard enough on the street to reach the operating regime that demands the actual correct oil that the engine needs on the track. And so they tell you to put that oil in for the track. Find a way to get more air into it, or, more likely, out of it. Or add a water spray for when it's hot. Or something.   As to the leak --- a small leak that cannot cause near catastrophic volume loss in a few seconds cannot cause a low pressure condition in the engine. If the leak is large enough to drop oil pressure, then you will only get one or two shots at it before the sump is drained.
×
×
  • Create New...