question on knock sensing

[David Redszus]

Standard motor product KS2, which should be equivalent to a 1986 knock sensor on iron gen I block.

I've found a formula to calculate the frequency based on the bore size. It pointed out 2.730 kHz.

That knock sensor was fitted to a number of vehicles: Buick, Cadillac, Chevy, Olds and Pontiac.
All engines were 4.3 to 5.0 liters in displacement.
The KS2 sensor output range is 5.5-6.4kHz, which is very close to my calculation of fundamental knock frequency based on bore diameter. Your formula might have a problem since it is way off the mark.
All engines have three knock frequencies, depending on which axis is being considered. The highest is often more than triple that of the lowest.

[RednGold86Z]

Standard motor product KS2, which should be equivalent to a 1986 knock sensor on iron gen I block.

I've found a formula to calculate the frequency based on the bore size. It pointed out 2.730 kHz.

That's probably too low - most are above 5kHz. Maybe you missed a factor of 2 in the calculation? Maybe it was a radius input, but you put diameter?

[Belgian1979]

Standard motor product KS2, which should be equivalent to a 1986 knock sensor on iron gen I block.

I've found a formula to calculate the frequency based on the bore size. It pointed out 2.730 kHz.

That's probably too low - most are above 5kHz. Maybe you missed a factor of 2 in the calculation? Maybe it was a radius input, but you put diameter?

I seem not to be able to find the link anymore. Something like 900/pi*bore diameter(in meter). I would think you're right. With radius it will be 5471
Bore is 4.125"

I have a bit of problems seeing a sensor voltage producing noise, but I'm not electronics expert.

[pheyden]

You might want to have a look at this site. They seem to do a good job.

http://www.jandssafeguard.com/

[RednGold86Z]

I have a bit of problems seeing a sensor voltage producing noise, but I'm not electronics expert.

The knock sensor has a floating "seismic" mass supported by piezoelectric material. When the case shakes (it's rigidly mounted to the block, and the block vibrates), a voltage is generated by the piezoelectric material as it compresses as the forces are transferred to the seismic mass. So, it can generate volts, but barely any DC current.

[dieselgeek]

Belgian, if you DO go down the knock sensing road and are still planning on MS-EFI, upgrade to the MS3X (it's a processor and case upgrade) because it contains "data gating" (windowing) capabilities in the software and similar tuning features to the Motec.

Although my opinion on the subject is "not worth it" I very much enjoy what guys like Jeremy (R&G) are willing to share as he works at an OEM level. This thread is a great read.

[BlackoutSteve]

(Hi Scott. )

I bought an MSD knock sensor. It also screws into the water drain plug just above the pan rail.
It's completely stand alone, just give it 12V and it works fine.
It doesn't have a frequency setting, but it does have a sensitivity adjustment. From memory, you bring the rpm up to ~2000 in neutral and adjust the sensitivy dial until the lights just go out.
It has an audible warning as well.
I quite like it. I have 12:1 on pump fuel with water injection and it's been very helpful showing me if I've needed the water to come on a little earlier as well as when a valve lash is required.. Serioulsy, it started flickering in association to the lash I was beginning to hear.

Bummer.. I just looked for the MSD link for you and it's not there.. It might be discontunued.. Oh well, so much for that! Maybe I was the only guy who liked it..

[Belgian1979]

(Hi Scott. )

I bought an MSD knock sensor. It also screws into the water drain plug just above the pan rail.
It's completely stand alone, just give it 12V and it works fine.
It doesn't have a frequency setting, but it does have a sensitivity adjustment. From memory, you bring the rpm up to ~2000 in neutral and adjust the sensitivy dial until the lights just go out.
It has an audible warning as well.
I quite like it. I have 12:1 on pump fuel with water injection and it's been very helpful showing me if I've needed the water to come on a little earlier as well as when a valve lash is required.. Serioulsy, it started flickering in association to the lash I was beginning to hear.

Bummer.. I just looked for the MSD link for you and it's not there.. It might be discontunued.. Oh well, so much for that! Maybe I was the only guy who liked it..

I never saw an MSD. That would have been interesting.

edit, found the partnumber : 8964. Seems to be discontinued.

[Dodge Freak]

Maybe they pulled it off the market since it was saving too many engines, lol

Gotta keep the part suppliers busy

[Belgian1979]

Belgian, if you DO go down the knock sensing road and are still planning on MS-EFI, upgrade to the MS3X (it's a processor and case upgrade) because it contains "data gating" (windowing) capabilities in the software and similar tuning features to the Motec.

Although my opinion on the subject is "not worth it" I very much enjoy what guys like Jeremy (R&G) are willing to share as he works at an OEM level. This thread is a great read.

Scott,

I know. Since I have the sensor I want to use it as much as possible. If not by having MS detecting knock, then by listening to the sound...It might be usefull.

Something like this maybe : http://www.youtube.com/watch?v=GGJZ6kKzLus

[BlackoutSteve]

I never saw an MSD. That would have been interesting.

edit, found the partnumber : 8964. Seems to be discontinued.

If you're interested...
In Australia.. http://www.gumtree.com.au/s-ad/sandgate ... 1007623253
or Rocket down here still might have stock. .. https://shop.rocketindustries.com.au/products/MSD8964
Even amazon list them. There's bound to some company on this planet that still has old stock.

Listed as in stock.. http://www.prostreetonline.com/pnsku/msd-8964.asp

If not, save a search on eBay and wait for one to show up locally.

[Belgian1979]

Standard motor product KS2, which should be equivalent to a 1986 knock sensor on iron gen I block.

I've found a formula to calculate the frequency based on the bore size. It pointed out 2.730 kHz.

That knock sensor was fitted to a number of vehicles: Buick, Cadillac, Chevy, Olds and Pontiac.
All engines were 4.3 to 5.0 liters in displacement.
The KS2 sensor output range is 5.5-6.4kHz, which is very close to my calculation of fundamental knock frequency based on bore diameter. Your formula might have a problem since it is way off the mark.
All engines have three knock frequencies, depending on which axis is being considered. The highest is often more than triple that of the lowest.

Hello Red,

Can you tell me if in between the 5.5-6.4 kHz the reading always means knock or are other things causing this frequency as well ?

[RednGold86Z]

It's difficult to say for sure, but it's typically caused by knock. Other components are smaller, and would make higher frequency sounds - but, the sensors aren't completely immune to those sounds, and several different frequencies combined can cause some vibrations at that frequency. Piston slap might also make a similar frequency sound, because it's slapping the bore, but it might not be as loud. Also, as many people have found - those "tuned" sensors are easily fooled by other sounds.

This is one reason why flat response sensors, combined with multiple tooth crank wheels are so nice - you can measure a quiet area in the cycle as a reference, and you can measure the "amount" of knock frequency noise in a region just near and after TDC.

[Belgian1979]

It's difficult to say for sure, but it's typically caused by knock. Other components are smaller, and would make higher frequency sounds - but, the sensors aren't completely immune to those sounds, and several different frequencies combined can cause some vibrations at that frequency. Piston slap might also make a similar frequency sound, because it's slapping the bore, but it might not be as loud. Also, as many people have found - those "tuned" sensors are easily fooled by other sounds.

This is one reason why flat response sensors, combined with multiple tooth crank wheels are so nice - you can measure a quiet area in the cycle as a reference, and you can measure the "amount" of knock frequency noise in a region just near and after TDC.

Red,

What sensor would you advise for my 4.125 bore SBC (347 ci)

[Kevin Johnson]

It's difficult to say for sure, but it's typically caused by knock. Other components are smaller, and would make higher frequency sounds - but, the sensors aren't completely immune to those sounds, and several different frequencies combined can cause some vibrations at that frequency. Piston slap might also make a similar frequency sound, because it's slapping the bore, but it might not be as loud. Also, as many people have found - those "tuned" sensors are easily fooled by other sounds.

This is one reason why flat response sensors, combined with multiple tooth crank wheels are so nice - you can measure a quiet area in the cycle as a reference, and you can measure the "amount" of knock frequency noise in a region just near and after TDC.

Sorry for the aside:

Do you know of any published data on the variable bouncing behavior of NaK valves with respect to operating temps, rpm sweeps, etc. being incorrectly interpreted by signal processing as knock?