Page 1 of 2

K24 knock sensor sensitivity differences.

Posted: Thu Apr 09, 2020 6:44 pm
by reliant_turbo
Hello,

Im new to the K-Pro and tuning Honda's and just installed a K-Pro V4 in my 02 CR-V 5-speed (bone stock) and Im working on dialing in fueling. I started with the Stock K24 Accord cal because it should be close to a K24A1. on my first drive down the road at low load just seeing how it was going to respond, the knock counts were cruising up super fast and got to over 100 within a minute. so just as a precaution, i pulled some timing and it got a little better. I knew Id have to work on fueling because that engine was modded so I started to smooth out some of the fueling.

I was poking around in other cals that come with KManager later and I noticed knock sensor sensitivity in the other k24 cals were quite different from the stock 2.4 accord calibration. as in way more sensitive on the accord cal. just as a test, i globally pulled 10 degrees of timing and I was still getting knock counts.

any reason for this? would the tsx knock sensitivity be a better choice for a K24A1? or even starting out with a TSX calibration and dialing back the fuel for the less flowy K24A1?

Also, are the stock CR-V injectors 270 or 310cc? I find conflicting info on the web.

Thanks

Brian

Re: K24 knock sensor sensitivity differences.

Posted: Thu Apr 09, 2020 6:48 pm
by reliant_turbo
I just took a look at all the cals in Kmanager and every single cal in there has similar knock sensitivity..... except the stock k24 accord calibration?

the accord one is super sensitive.

Thanks,

Brian

Re: K24 knock sensor sensitivity differences.

Posted: Fri Apr 10, 2020 11:20 am
by Spunkster
You could reduce the sensitivity , but you need to be sure that you aren't' getting actual knock and just ignoring it.

Re: K24 knock sensor sensitivity differences.

Posted: Fri Apr 10, 2020 2:45 pm
by reliant_turbo
No worries, I have listening device's to determine that.

While I'm new to K-pro, I'm not new to tuning turbo cars or to adjusting knock sensitivity vs rpm tables so that the knock systems still work when the block starts ringing more when the power drastically increases over stock. :-)

Do you have a definitive answer why that sensitivity curve for that cal is so sensitive vs everything else that's built into Kpro? Even the d17 stuff and the appropriate era flashpro cals that I've looked at had similar knock sensitivity settings vs the k24 Accord outlier.

If not, I'm going to set mine to a different k24 curve and take a ride with the headphones / muffs on to make sure it's still not knocking.

Thanks,

Brian

Re: K24 knock sensor sensitivity differences.

Posted: Fri Apr 10, 2020 3:37 pm
by Spunkster
I do not know why it would be different compared to the others as that calibration was made over 10 years ago.

Re: K24 knock sensor sensitivity differences.

Posted: Fri Apr 10, 2020 4:39 pm
by reliant_turbo
ok thanks for the answer.

time to see if a TSX knock sensitivity curve works for a K24A1.

Thanks,

Brian

Re: K24 knock sensor sensitivity differences.

Posted: Sun Apr 26, 2020 8:41 pm
by reliant_turbo
a follow up. I started tuning fuel on my CR-V today on the street using the stock accord kal. with the fueling reasonably tuned and the timing backed off a bit and an entire can of torco accelerator in the tank (this stuff is no joke.... a buddy of mine filled his 500whp dodge spirit R/T up with 87 octane and dumped in a can of this stuff and let it mix up before he hit it.... 25psi.... no knock) and went for a quick leisurely drive around the 'block' (3 miles). 252 knock counts by the time I got home. no WOT, all part throttle mild driving. no audible knock.

copied in the 07 TSX flashpro knock sensitivity curves as they are a bit more sensitive in some areas than the TSX curves in k manager. I didnt move the breakpoints around, i just copied the values for the breakpoints that i had and visually interpolated/lined up the curves so they matched.

same route this time with WOT driving to redline. 2 knock counts at the VTEC crossover point (~2200 on a stock crv).

tomorrow the knock listening amplifier/headphones go on to make sure that I dont have actual knock. with ~100 octane fuel, i doubt it.

Thanks,

Brian

Re: K24 knock sensor sensitivity differences.

Posted: Thu Apr 30, 2020 3:10 pm
by reliant_turbo
I have another question..... I see that knock level is a loggable value and its in volts.. and knock threshold is a loggable value but mine sits at 0v the entire time. am I missing something in parameters that causes the knock threshold to not track with knock sensitivity? i have "knock sensor enabled" checked and thats it.

since knock threshold is in volts and knock sensitivity is in %..... is that % relatable to the 0-5v that the knock sensor is translated to?

thanks

Brian

Re: K24 knock sensor sensitivity differences.

Posted: Thu Apr 30, 2020 3:50 pm
by Spunkster
Have you read the info in the help file: http://www.hondata.com/help/kmanager/kn ... tsub=knock

Re: K24 knock sensor sensitivity differences.

Posted: Thu Apr 30, 2020 4:51 pm
by reliant_turbo
yes. i was reading it while i was creating the question and searching the forum and looking at every thread where the word knock appeared and didnt find an answer why knock threshold was 0v in my logs.

the definitions of knock level and knock threshold in the help file indicate that Knock level should be the current knock sensor value translated to 0-5v and knock threshold should be somehow correlated to what value of the knock sensitivity table the engine is operating in.

the only thing i found that might explain this is "Knock Level is translated to show the ratio of knock level to knock threshold." in sensor overlay though im not sure if that is in the correct context of what im wondering about.....

im trying to log knock level and knock threshold to make sure the sensitivity curves are set correctly or see how close i am getting to the threshold.

there is nothing in the help file that explains why when logging, the knock threshold would be 0v all the time (even though i had a few knock counts) or why knock threshold and knock sensitivity tables are in 2 completely different units. unless knock sensitivity commands the knock amplifier circuitry to change its gain based on whats in the table but that seems silly when you can just set a threshold voltage and compare to that....

any insight?

Brian

Re: K24 knock sensor sensitivity differences.

Posted: Thu Apr 30, 2020 6:02 pm
by reliant_turbo
i might have figured out why the knock threshold stayed at 0v.

it seemed to affect the PND ecu back in the day but was fixed. viewtopic.php?f=12&t=2660
Hondata wrote:Only the base RSX was affected.

Conrad, you must remember that we start with what Honda uses and then modify the behaviour to suit us. With the RSX Honda does not use any form of knock retard. We have not changed any of the behaviour in that area. There never have been editable knock tables.
im wondering if the base accord calibration PNF may share this behavior. or the K20A3 Civic Si considering the non performance vtec.

perhaps thats why the knock sensitivity tables are so sensitive in those calibrations? if there isnt knock retard than what does it matter if the table is sensitive. unless that thread was old information and im incorrect....

im going to try to copy all of my tables into a RSX-S calibration and change all the parameters to match and see if the behavior changes.

i assume i can run a PRB based calibration on a PPA ecu without issue? (no 6 speed to worry about reverse lockout anyway)

Thanks

Brian

Re: K24 knock sensor sensitivity differences.

Posted: Fri May 01, 2020 2:24 pm
by Spunkster
Yes you can try that, but fuel values may need some adjustment.

Re: K24 knock sensor sensitivity differences.

Posted: Fri May 01, 2020 2:42 pm
by reliant_turbo
I have my street tuned fuel and ignition values that i have started copying in and I had to change the breakpoints for high cam fuel/ign/etc to get them low enough for the 2200 vtec point of the non performance cam setup that i still have for now. all of the parameters are changed over to what i was running which was stock k24 accord cal based.

now i didnt have the ecu connected to my laptop when i was doing all the setup but "secondary runner" was greyed out so i couldnt change it but it was checked.

will the PRB based cal operate the secondary runner solenoid that my PPA ecu definitely has. does it allow me to change this value once the ecu is connected to the laptop?

Thanks,

Brian

Re: K24 knock sensor sensitivity differences.

Posted: Fri May 01, 2020 3:14 pm
by Spunkster
No, a PRB calibration will not operate the secondary runners.

Re: K24 knock sensor sensitivity differences.

Posted: Fri May 01, 2020 5:02 pm
by reliant_turbo
damn.... so i have to choose between seeing knock threshold in the logs or my dual runner manifold working correctly.

i guess its time to swap on my stock PRB manifold i have sitting there and retune the fuel.

Brian