Knock Retard Issue 2.3 Ecoboost

[Smenezes428]

Hopefully someone here can steer me in the right direction. I have a 2.3 Ecoboost from a 2015 mustang swapped into a Mitsubishi Starion. No matter what I do to the tune it is showing -1 to -2.5 degrees knock retard at wot. I thought it could be false knock so i searched everywhere even building a new exhaust, motor mounts, intake to eliminate any possible source of the noise. I tested it in forscan and the knock sensor readings were not showing that it was sensing knock. Today I tried pulling 5 degrees from the global and it is still there. Then i decided to go deeper and zeroed out the kr amount from table 1608. This should have made it show 0 degrees no matter what it heard but it stilled show the same knock retard. I have attached the log.
Where else would the ecu see something that would pull timing in KR? I am stumped.

Knock Test.hpl

[SlowSTExploder]

You might be thinking about this backwards, on the explorers negative knock retard is actually adding timing, to verify log individual cylinder knock retard +/-. That will tell you if its actually adding or taking timing and what cylinder.

[SlowSTExploder]

Maybe it was individual cyl knock adv/ret on vcm scanner i can don't have my laptop with me or I'd tell you what exactly the parameter is called.

[Smenezes428]

Its not, negative is pulling timing. I have verified this. In the log I am monitoring individual cylinder.

[SlowSTExploder]

i see it, are the spark plugs new?

[Smenezes428]

Yes, the spark plugs are new, step colder and gapped t0 .026. The issue is definitely in the programing. Something other than knock is telling the ECU to pull timing. I just cant find what's doing it.

[Rob@HPTuners]

I spoke with Ryan@EMS about a similar issue so I'm assuming its your car. I think something weird is going on under the hood with the control pack strategy and we need to look more closely.

[Smenezes428]

Yep, that's my car. Thanks, hopefully we can get to the bottom of it.

[Seishuku]

That log looks fairly normal to me, the timing is pretty low, but KOM is also 0 (neutral octane)... What kind of fuel are you running on it?
On 93 (US), my Mustang typically normally sees -1 or -2 on cylinders 1 and 2 as well, but my overall timing is much higher than what you're seeing too (around 2 to 6 degrees).
But if you're on relatively junky fuel and the ECU doesn't think it can add any more timing in, the timing you're seeing I would think is pretty normal (for 87 octane).

Edit:
It is a bit weird that you're not able to add in or remove timing though, something definitely isn't right there.

[Smenezes428]

That log looks fairly normal to me, the timing is pretty low, but KOM is also 0 (neutral octane)... What kind of fuel are you running on it?
On 93 (US), my Mustang typically normally sees -1 or -2 on cylinders 1 and 2 as well, but my overall timing is much higher than what you're seeing too (around 2 to 6 degrees).
But if you're on relatively junky fuel and the ECU doesn't think it can add any more timing in, the timing you're seeing I would think is pretty normal (for 87 octane).

Edit:
It is a bit weird that you're not able to add in or remove timing though, something definitely isn't right there.

The overall timing is low because we are testing why it's pulling the timing. I pulled it way back to limit the power and ensure that there was not any really knock. I am able to add timing and am on 93. We are trying to figure out why even with knock sensors set to remove 0 it still shows -1-2? on cylinders 1 and 2. It is having zero knock but still pulling timing.
What year is your mustang?

[Seishuku]

2015, SGMN0Y9 OS, as far as I know, it's basically the same as the control pack system.

I'm generally ok with up to -3 or -4 on one or two cylinders, I would imagine with how sensitive the system is in general that it's always "picking up knock"... That's just how the timing system works, kind of like how nearly all the WOT fueling feedback is done by wideband O2.
It's constantly monitoring timing and testing if it can add back in more, when it does, if it sees too much knock, it pulls it back out... The system is guided by the KOM as a global learned octane value.

If you're really concerned that it's actually knocking, I would set up some old school det-cans and physically listen to it and verify that it is actually knocking and just dial back timing where needed.
I've learned to just leave most of the timing tables alone and let the knock advance system do it's thing and learn from driving, pretty much the only thing I've ever done is increase cylinder pressure limit.

[Smenezes428]

I had the knock system set to 0 and it was still pulling the timing. According to Ryan at EMS and HP tuners it shouldn't be doing this. If you look at the Retard tab and the knock amount that is the timing it should pull per cylinder based on the knock events.
If those were sent to 0 it has to be coming up with that -2° from somewhere else and that is what we are trying to get to the bottom of.

[Seishuku]

Ah, I see.

Not sure if it would be helpful, but if you like I could try zeroing out ECM 1608 (Knock retard amount) on my Mustang and see what happens. I've never had any reason to change that table, but in the sprit of debugging, I'd be willing to try it to help out.

[Smenezes428]

I appreciate the offer but wouldn't want you to risk your car. HP tuners is looking into it so I will wait and see what they say.

[Rob@HPTuners]

Honestly I haven't ever investigated this closely because I've never needed to intentionally disable knock control as a diagnostic measure. But my assumption would be that if you set it to not pull timing, it shouldn't. So I think it's worth looking at the logic (which may have to happen next week at this point). I do think that false knock might still be at the root of the issue, but it's also possible that something else is going on.

I exposed the knock sensitivity tables for Ryan so we'll see if we can get anywhere with that in the meantime.

[Smenezes428]

Honestly I haven't ever investigated this closely because I've never needed to intentionally disable knock control as a diagnostic measure. But my assumption would be that if you set it to not pull timing, it shouldn't. So I think it's worth looking at the logic (which may have to happen next week at this point). I do think that false knock might still be at the root of the issue, but it's also possible that something else is going on.

I exposed the knock sensitivity tables for Ryan so we'll see if we can get anywhere with that in the meantime.

Thanks, I appreciate your time on this.
I was convinced it was false knock as well. The thing that has changed my mind was when I logged the knock sensors in forscan and knock sensor 1 which is monitoring cylinders 1 and 2 was actually reading lower than 2 but cylinders 1 and 2 were showing the knock retard.
Let me know if you need anything from me to help troubleshoot this.

Thanks again