For any discussion not specifically related to your project
By JHolland
#48719
Black Knight wrote: Tue Mar 02, 2021 3:40 pm
JHolland wrote: Tue Mar 02, 2021 9:50 am What knock sensor are you using?
The stock one that came from factory.
I think I should trust the research of the factory.
They spent the time to to make it work as well as possible and find the location and angle that works best.
Subaru through hole one wire.
So far it looks like they did a very good job.

Black Knight
I couldn't agree more. I don't have any data for that sensor though, if it had been a more common Bosch type then it would have been possible to determine what magnitude of G you were seeing, its not of any real importance but it would have been interesting (to me anyway!).
By Black Knight
#48724
JHolland wrote: Tue Mar 02, 2021 4:05 pm
Black Knight wrote: Tue Mar 02, 2021 3:40 pm
JHolland wrote: Tue Mar 02, 2021 9:50 am What knock sensor are you using?
The stock one that came from factory.
I think I should trust the research of the factory.
They spent the time to to make it work as well as possible and find the location and angle that works best.
Subaru through hole one wire.
So far it looks like they did a very good job.

Black Knight
I couldn't agree more. I don't have any data for that sensor though, if it had been a more common Bosch type then it would have been possible to determine what magnitude of G you were seeing, its not of any real importance but it would have been interesting (to me anyway!).
Thanks for that. It is nice to get support for these new ideas and testing.

You have shown me something I should do.

Test a very common knock sensor so it will help more people and could become a standard.

I will try to find a very common one to test soon.

Do you know of a knock sensor that would be good for this.

It would be great if it was found on Ebay and was a through hole as they are easier to mount on non stock aps.

Thanks for this info.

Just going out now to record both sound signal and hopefully data log to TS with real knock.

If I get good data I will try to post a pics of the change from raw signal to filtered and amped.

Black Knight
By Black Knight
#48725
Great day getting knock data on the sound recorder software.

Far far better than I hoped for. I can see knock signal change with timing change.

I set up a test area in the timing table at 3000 to 4500 rpm and 60 to 80 kpa.
I set up with 5% and 15% and 30% timing increase from my good timing table.
I could see and hear on the recording program very distinctive spikes that were not there with the good tune timing table.
I did see some random very small spikes on the good timing, so I am a little advanced in a few spots.

Wow, this sensor is great.

This is on a subaru flat 4 2.2. The sensor is closer to #4 cyl and far away from #1 cyl.
I can see by the db/volt of this sensor a stronger sound on #4 cyl and lees on #1 cyl.
I can see much more than that and it will take time to learn how to use it.

I found the knock frequency is centered at 2000hz and blends up to 3000hz and down to 1000hz.
I did not get to test it on TS because I need more parts for the noise filter and I don't have them.
They are on order now I know what I will need.

With the limited filtering it was great and I could tune with it.
With the right filter set up and amplifying it will be better.

It is amazing what I could hear and distinguish.

I mean every thing including the fuel pump until I set up the limited filter.

Very fun and cool.

Just sharing.

Black Knight
By JHolland
#48727
When I dealt with Bosch about 10 years ago I was told that their wideband sensors are all the same electrically, newer ones are probably the same too. I tested some Volvo sensors and they matched with the datasheet that I had, I did a brief write up:

http://rhinopower.org/knock/knock.html

Edit: The spec for the current Bosch Motorsport knock sensor is much the same and has the same 26mV/g response.
By Black Knight
#48733
JHolland wrote: Wed Mar 03, 2021 8:59 am When I dealt with Bosch about 10 years ago I was told that their wideband sensors are all the same electrically, newer ones are probably the same too. I tested some Volvo sensors and they matched with the datasheet that I had, I did a brief write up:

http://rhinopower.org/knock/knock.html

Edit: The spec for the current Bosch Motorsport knock sensor is much the same and has the same 26mV/g response.
Thanks for the link.

Great write up.

You are way ahead of me in many ways.

I will read and learn.

Thanks

BK
By Black Knight
#48871
JHolland wrote: Wed Mar 03, 2021 8:59 am When I dealt with Bosch about 10 years ago I was told that their wideband sensors are all the same electrically, newer ones are probably the same too. I tested some Volvo sensors and they matched with the datasheet that I had, I did a brief write up:

http://rhinopower.org/knock/knock.html

Edit: The spec for the current Bosch Motorsport knock sensor is much the same and has the same 26mV/g response.
Thanks again for the links.

The signal is isolated now and I can set this controller to any knock threshold I want. The engine noise can be eliminated or still be seen as a very low db/volts/hz. At this setting the knock is the only signal there with a slight engine noise I choose to have there as a reference.


I have the signal converted to volts also.

The question is.

Will this work or will I need a logic controller to take my volt signal to the ecu?

It looks like the volt signal should work. It is clear and at the threshold I choose.

I hope my amateur wording is clear enough.

Thanks for any input.

Black Knight
By Black Knight
#49489
Thanks to Pazzi and one other member on speeduino I have been informed that there is a knock sensor firmware and a window for the knock sensor on the current firmware.

Very good news.

I just got my controller to log with auxiliary input with a good clean square wave signal.
I do have about 25 ms signal delay that was 250ms yesterday because of the limitations of the aux input. More adjustment of the controller and the inc file should help some here.

I may try UN-commenting the section of the firmware that is there for knock sensor and see what happens. I am set up as a bench test with only a mega board with no shield. I have the amps and volt correct for the mega board and no noise signal.

The knock signal is responsive and misses nothing. I can even hear/see a pencil eraser tapping on the test bench set up. Then I can adjust it so that is gone and then knock onset is there.

Noise signal is now less that .03v and knock is showing between 1.1v and 4.8 volts on TS logs.
Any signal above 2.5 vols is far too much knock and would most likely cause engine damage.

Got it done. TS is logging knock with minimal delay. Still needs work but it is good enough.

I hope my controller will work with the new firmware knock set up.

Thanks to all of you that helped.

Black Knight
Last edited by Black Knight on Tue Apr 06, 2021 11:35 pm, edited 1 time in total.
By Black Knight
#49532
LPG2CV wrote: Sat Apr 10, 2021 9:47 am I took a quick sqwiz at the code and ini, and cant see that it's putting out any monitoring info to gauges/indicators or log.
Good point to bring up LPG.
Don't want to mislead anyone to thing this done yet.
Josh will let us know when is is working

I am looking forward to it working because my knock controller is working and would benefit from this window.

I will be posting the signals I am getting on TS logs very soon.

Thanks LPG

BK

Will this have an updated version about this featu[…]

See my earlier post . The Base Tune sets t[…]

Vr Conditioner Noise when cranking

I did the fix, but it's not work. I put my setti[…]

Perhaps some different points and pictures. Instr[…]

Still can't find what you're looking for?