Knock threshold voltage?
- Thread starter mkIIIman089
- Start date
mount it to a speaker and run an audio test sweep and tone around 7khz and look for a voltage spike.. Unless IJ has already done something similar, witch of course wouldn't surprise me. :biglaugh:
Last edited:
You're right, it should spike when you apply a 7khz vibration to it, but the amplitude of that vibration will change the voltage. So doing that will simply show me that the sensor works - not what the threshold the TCCS looks for is.
Yea, similar to what I was finding... that or people who think the KS pin on the ECU connector has something to do with knock sensors. 
As I understand it, it's not limited to a voltage, but instead looks for an amplitude in the specific frequency range for longer than a certain duration.
It's absolutely not a simple thing that can be detected with a voltmeter. Which explains why there hasn't been a simple knock display unit put together and sold for this system.
It's absolutely not a simple thing that can be detected with a voltmeter. Which explains why there hasn't been a simple knock display unit put together and sold for this system.
The knock sensor contains a piezoelectric element that generates voltage. It hits max volts at 7kHz. The type used in the 7M generates high voltage over a broader range of vibration frequencies. Unfortunately, the book I have is not specific on what that voltage is. The only way to figure this out it to subject the sensor to increasing kHz frequency from 0 to 15 or so and measure the voltage output.
OK, so the 7M system isn't specifically tuned to 7khz, but it still works off the same principal of voltage over a certain point and duration = take action?
However, grim claims that a monitor cannot be made by monitoring voltage output, yet jdub says you could check the output by running through a frequency range and monitoring the voltage output from the sensor... I don't see why you could not make a monitor that would show something (a light illuminating for instance) once the threshold voltage is reached. Basically telling when the ECU is starting to pull timing.
jdub, am I correct in thinking that the sensor has a range of frequencies that it's "tuned" to and within that range the varying amplitude of vibration will increase/decrease output voltage?
However, grim claims that a monitor cannot be made by monitoring voltage output, yet jdub says you could check the output by running through a frequency range and monitoring the voltage output from the sensor... I don't see why you could not make a monitor that would show something (a light illuminating for instance) once the threshold voltage is reached. Basically telling when the ECU is starting to pull timing.
jdub, am I correct in thinking that the sensor has a range of frequencies that it's "tuned" to and within that range the varying amplitude of vibration will increase/decrease output voltage?
That is essentially correct...peak voltage occurs at 7kHz. The 7M ECU pulls timing based on this voltage, with the max amount pulled at 7kHz. The book I have shows a bell shaped curve with peak at 7kHz.
Gotcha, thanks for your help! Too bad even that book you've got (which has the most detailed documentation I've heard) still doesn't have what the voltage is.
Either it's an engineering super top secret, or it's actually a moving target depending on other factors. Specifically what those factors were escapes me, but I do recall reading about other timing modifying conditions that can exist.
Either it's an engineering super top secret, or it's actually a moving target depending on other factors. Specifically what those factors were escapes me, but I do recall reading about other timing modifying conditions that can exist.
If you have a Scope (or access to one), you could do a simple test using a timing light to monitor timing, a scope to monitor voltage and frequency and a small wrench to create the knock / trigger the sensor. Its not perfect but will get you pretty close to finding out the info you want. With the engine running, if you tap on the block (somewhat close to the sensor with a 10mm wrench, you should be able to trigger the knock system to kick in and you'll see it with a timing light plus you'll possibly notice a ever so slight change in RPM due to the timing retarding. You should be able to see at what frequency the ECU's knock system kicks in and you should be able to see what happen's with the voltage by starting out with light taps and getting harder and harder untill you trigger the system...
There are other veriables that effect timing, but I dont know if they have any effect on knock control. One thing that many ECU's do (but Im not sure if the supra TCCS does this) is ignore / shut down the knock system above a certain RPM due to mechanical noise that can cause false knock sensor output.... This point is dependant on the engine design and the amout of mechanical noise it naturally produces at higher RPM's but most cause, but its usually above 5000 to 6000 RPM. Like for example, the MKII MR2's stop using knock sensor feedback above 6000RPM. Again, this is dependent on the engine.
Hope this helps...
There are other veriables that effect timing, but I dont know if they have any effect on knock control. One thing that many ECU's do (but Im not sure if the supra TCCS does this) is ignore / shut down the knock system above a certain RPM due to mechanical noise that can cause false knock sensor output.... This point is dependant on the engine design and the amout of mechanical noise it naturally produces at higher RPM's but most cause, but its usually above 5000 to 6000 RPM. Like for example, the MKII MR2's stop using knock sensor feedback above 6000RPM. Again, this is dependent on the engine.
Hope this helps...
Well like I said hook it to a speaker and see how much higher the voltage spikes at 7khz relative to other frequencies and apply that to normal driving conditons.
Like I said, frequency alone will not determine peak output. It will just tell me that it's tuned to ~7khz, like I knew, and as jdub elaborated on in his posts.
I ran a simple knock link on my 7M for awhile....
Had to turn it's gain down so far it was next to useless, couldn't understand why until I wore a pair of electric ears during a dyno sessions...
These engines are NOISY under load.
The new MoTeC system I have for this engine uses a sync signal from the ECU to listen only to 10>15 degrees after TDC ib the firing order to help cancel out any false triggers.
You tune the system on the dyno by inducing knock and recording it's signature then you tell the knock sensor "This IS knock" do something about it.
You also have a gain adjustment for each cylinder as the further from the KS the less the amplitude it hears so by trimming the gains you only need 1x KSensor.
Had to turn it's gain down so far it was next to useless, couldn't understand why until I wore a pair of electric ears during a dyno sessions...
These engines are NOISY under load.
The new MoTeC system I have for this engine uses a sync signal from the ECU to listen only to 10>15 degrees after TDC ib the firing order to help cancel out any false triggers.
You tune the system on the dyno by inducing knock and recording it's signature then you tell the knock sensor "This IS knock" do something about it.
You also have a gain adjustment for each cylinder as the further from the KS the less the amplitude it hears so by trimming the gains you only need 1x KSensor.
Interesting! That MoTeC system is quite advanced, didn't think you had that kind of fine tuning outside of very high end racing systems.
mkIIIman089;1420126 said:
It's quite expensive and has been in development for a very long time, they won't sell it to anyone that hasn't done their course in setup and tuning of it.
(that's it in the middle)
Digital power distribution box made by ADD/Pratt & Miller, another one I've seen in use is Life Racing's PDU.
AFAIK they're all very similar, most of the difference is probably in software and the level of tweaking that can be done.
AFAIK they're all very similar, most of the difference is probably in software and the level of tweaking that can be done.