Stupid Question --- Knock???

[GHuggins]

OK, I've tuned only a handful or two of the newer gen 5's and something just won't stop bugging me... Knock... Scanner will show spark knock on the histogram, but it's not actually knocking or at least I don't see timing getting pulled from spark advance and it shows absolutely no loss on the dyno in these areas. It will only show 2 to 5 degrees knock retard, but again it makes power with this and isn't actually pulling timing. I've even gone to the trouble of pulling plugs and inspecting due to the confusion with no signs of detonation. Doesn't mean that it actually isn't, but all signs point to the contrary? Then the second thing I've noticed is that it will actually knock (audible) in the lower engine load areas where I obviously go in and pull timing out, but on the knock histogram, shows no knock when it's actually knocking? This is always with untouched knock sensor parameters...

This may have all ready been discussed on here, but was just wondering what others were seeing and what your opinions or corrections for these issues were. Possibly just a scanner issue of mine?

Posting config file just in case... It did originally come from another tuner then I've changed it since then, so I'm not sure.

[Redline MS]

I've learned to ignore the knock sensors on the GEN5 motor.....when we got our C7 in late 2013 we filled the tank with MS109 fuel on a stock car with a stock cal....still showed knock....that was enough for me....it is odd to see it record KR yet not take it from the final number. If you have reduced the Base Retard table this limits the amount of aggressiveness the knock function has so I think even though its showing it (which is most likely pure noise) it's not applying it to the base table...

[GHuggins]

I've learned to ignore the knock sensors on the GEN5 motor.....when we got our C7 in late 2013 we filled the tank with MS109 fuel on a stock car with a stock cal....still showed knock....that was enough for me....it is odd to see it record KR yet not take it from the final number. If you have reduced the Base Retard table this limits the amount of aggressiveness the knock function has so I think even though its showing it (which is most likely pure noise) it's not applying it to the base table...

See that's just it. All knock tables are stock. I haven't changed a single one on any of the gen5's I've done... Some even with serious power adders won't show knock at all and then a nearly stock truck with a cai will show constant knock? Like you said, you can watch it take out timing or supposedly take out timing with the knock retard pid only to see the timing stay exactly what's being commanded with the timing advance pid. You also won't have any loss of power in these areas on the dyno. I've seen them show up to 2 or 5 degrees knock retard only to want more timing. Then after adding timing, show less retard? These things do seem to like a little timing, but I'm just wondering what the engines actually have in place to protect themselves if they don't pull timing when they "sense" they're knocking? OR if they have to knock a set amount say 10degrees of knock retard before they'll actually pull anything. I'm also really confused at the fact that they'll have actual audible knock and not show anything at all on the knock sensor read outs, if it was just so sensitive to "noise" that it shows knock when it's not actually knocking? Here in the mountains where you can really load these motors with light throttle you can make these things knock like a sob If there's no built in "safety" that actually works, what would you recommend short of just keeping timing pulled out in the 30 to 90 cylinder mass sections from 1800 or so less rpms?

Then I was also wondering if it was actually "spark knock" at all and not a "fuel knock" which is why it's not pulling timing? I've noticed on some applications it does seem to be much worse if the injection timing is off. Still wouldn't explain why race fuel showed the same knock in your case. Don't know... Just wondering what the engines actually have for "self protection"? Guess it just means we have to get things extra close on these... Would really prefer to be able to "dial" in the knock sensors so power can be added while still having a "safety net" in case anything ever happened such as bad fuel being added to the vehicle instead of just having knock shown and nothing being done for it within the ecm's operating system. Very confusing why gm didn't have them setup to actually pull timing. Maybe they do and they just have to see knock for a given time period before transferring to the low octane table or before it will actually pull timing at all?

Perhaps I'm just thinking too much into it. Not sure???

[jtrosky]

I thought that I read somewhere that the scanner timing advance values don't reflect the reduced advance when KR is pulling timing. Can anyone confirm that? If so, that would explain why you're not seeing the timing values go down when there is KR.

Or, if the scanner really does take into account the KR - you need to remember to add in the values from all of the spark modifier tables as well. So if the base commanded timing is 25 degrees and then another 5 degrees is added by the spark modifier tables, but then you have 5 degrees of KR, the scanner would still show 25 degrees of timing - even with 5 degrees of KR - becuase of the spark modifier tables (damn, that sounds confusing!). I'm just saying that you can't forget about all of the spark modifier tables (and there are a LOT of them on these newer engines!).

[GHuggins]

I thought that I read somewhere that the scanner timing advance values don't reflect the reduced advance when KR is pulling timing. Can anyone confirm that? If so, that would explain why you're not seeing the timing values go down when there is KR.

Or, if the scanner really does take into account the KR - you need to remember to add in the values from all of the spark modifier tables as well. So if the base commanded timing is 25 degrees and then another 5 degrees is added by the spark modifier tables, but then you have 5 degrees of KR, the scanner would still show 25 degrees of timing - even with 5 degrees of KR - becuase of the spark modifier tables (damn, that sounds confusing!). I'm just saying that you can't forget about all of the spark modifier tables (and there are a LOT of them on these newer engines!).

Unless I missed some which is more than possible, I went through all of the modifier tables just for this reason. You always see that a lot on the older gens as well... If it's indeed still pulling timing even though it reports otherwise in the scanner, it still wouldn't explain why you can add more timing and have less knock retard reported plus make more power on the dyno. Then there's also the fact that it won't show knock retard when there's actually knock. That more than anything is what I've had trouble wrapping my head around... All of this added to the fact that higher octane fuels make no difference at all really has me wondering if were even able to see the correct knock retard values in the scanner? Would explain things better... Tuning the gen 5's seems to be pretty straight forward and for the most part easier other than the obvious timing problem. Torque models have to be pretty spot on if not slightly under actual and smooth. Makes for best driveability if the dd's are slightly under and airmass tables have half what you would normally add to them actually added to them. MAF's and VE tables have to be very spot on to really make power. Then fuel supply seems to be very important with the DI. If fuel pressure drops much below target at all they seem to always have "power making" issues. You can still play with soi to fix the fueling, but power will never come up like it will with adequate supply with the soi decreased in the higher ranges. Other than all that, they don't seem too bad...

[Redline MS]

I wouldn't let these issues with the knock sensor logic get to you. If you find optimal advance by doing a spark hook test you should never be near knock anyway. As far as I'm concerned the difference between optimal spark and knock is pretty obvious if you use this method. Just for clarification regarding this test; you need a load controlled dyno that can read torque in real time. You keep adding spark until the torque doesn't increase. That's peak cylinder pressure (for at least that fuel). I think we put to much weight into the knock controls and most certain the GEN5 stuff has us all scratching our rears. At some point you need to look at all the cars you have done prior and be realistic on where spark should be. GM uses the spark hook test to cal there cars then uses other adders to pull back which is why we can pickup power on a factory car at times. Anyone who leaves 3-4 degrees of spark on the table is just shorting there work. Optimal is optimal if you know how to find optimal!

I like these customers who say "I want a conservative tune not on the edge.." Why be conservative if you know what makes peak power and NEVER be on the edge.

That fuel test we did put it all in check and who the hell knows what GM was thinking....LOL!

[GHuggins]

I wouldn't let these issues with the knock sensor logic get to you. If you find optimal advance by doing a spark hook test you should never be near knock anyway. As far as I'm concerned the difference between optimal spark and knock is pretty obvious if you use this method. Just for clarification regarding this test; you need a load controlled dyno that can read torque in real time. You keep adding spark until the torque doesn't increase. That's peak cylinder pressure (for at least that fuel). I think we put to much weight into the knock controls and most certain the GEN5 stuff has us all scratching our rears. At some point you need to look at all the cars you have done prior and be realistic on where spark should be. GM uses the spark hook test to cal there cars then uses other adders to pull back which is why we can pickup power on a factory car at times. Anyone who leaves 3-4 degrees of spark on the table is just shorting there work. Optimal is optimal if you know how to find optimal!

I like these customers who say "I want a conservative tune not on the edge.." Why be conservative if you know what makes peak power and NEVER be on the edge.

That fuel test we did put it all in check and who the hell knows what GM was thinking....LOL!

I like that All though I personally like to go by hp when dialing in timing up high in the rpms and use peak torque only down low or 3500 or less rpms... Guess the whole direct injected thing combined with how cheaply the pistons seem to be manufactured now in GM vehicles just had me a little over worried with knock?

Thanks

[Redline MS]

I like that All though I personally like to go by hp when dialing in timing up high in the rpms and use peak torque only down low or 3500 or less rpms... Guess the whole direct injected thing combined with how cheaply the pistons seem to be manufactured now in GM vehicles just had me a little over worried with knock?

Thanks

Its a good reason but we have plenty of LT1's getting pounded with power with similar setups as cars that have had failures. Whenever we've seen a failure there was no clinical reason to explain why other they crap parts. The spark hook test was really developed for the part of the spark table that is not WOT. Which in reality is the reality of a good driving car. The WOT part is pretty easy and yes HP is the way to look at it. I think people forget that an engine is producing power from idle to WOT and getting peak power out of every load vs rpm is what makes a car feel great...its just where all the work is.

Ultimately it would great to have pressure transducers in the cylinder as that will solve if you really have knock and to find peak cylinder pressure.....expensive tools..

[GHuggins]

Yes. Most out there don't realize that drivability is where nearly all of the time needs to be spent on the best of tunes... Seems like most think you can slap on a blower or turbo, tune wot and be done... Of course even good wot tunes need a little time invested

What did y'all ever find out with your exhaust temp testing?

[Redline MS]

Yes. Most out there don't realize that drivability is where nearly all of the time needs to be spent on the best of tunes... Seems like most think you can slap on a blower or turbo, tune wot and be done... Of course even good wot tunes need a little time invested

What did y'all ever find out with your exhaust temp testing?

Still playing with it...we are crazy busy and a few weekends in a row I've been traveling. I just need to come in on a Saturday and get in the zone. There is no doubt the correlation between EGT's and lambda will make a whole lot more sense of when injection timing is in the range. What we need a SOI summit meeting to real take everyone's experiences and run it through the grinder!

[GHuggins]

Still playing with it...we are crazy busy and a few weekends in a row I've been traveling. I just need to come in on a Saturday and get in the zone. There is no doubt the correlation between EGT's and lambda will make a whole lot more sense of when injection timing is in the range. What we need a SOI summit meeting to real take everyone's experiences and run it through the grinder!

Sign me up for that one

[MikeOD]

I noted knock retard in a C7 ZO6 I'm working on the other day, and the final delivered timing matched up to the amount it said knock retard was pulling, after all the other adders and multipliers (AFR ECT IAT...) I pulled a little timing back and it stopped showing knock, and the delivered timing lined up with the adders and main table again. I am a little confused with these engines still...mostly need to learn more about the torque modeling I think, but overall, I'm getting there.

[Blue-Z]

Working on a c7 now, added cold air and cat back exhaust. No tune, car was knocking 4 degrees out
Of it. Fuel is in line now and randomly see 1-2* out of it. Is there a way to desensitize? Or is that not a good idea. Car likes a bit more timing, don't hear any sort of pinging whatsoever. Thanks

[Blue-Z]

Also, I am getting same results, knock shows 2* retard but advance doesn't change. Is it really pulling the timing??

[10_SS]

ive got a 2016 Camaro here, LT1. Heard lots of audible knock... havent logged it yet, but now Im expecting to see 20+ deg or retard due to all the noise I hear. Must be 87 octane, but still shouldnt hear it to that extent.

[LSChev]

Noticed the same thing on our 2015 Denali. Checked the table and it should be commanding 14 degrees. Checked all the modifiers and am logging all the modifiers I can. When we do a run, it only gives me 12 degrees. Added another degree. Only showed 12 degrees but made a bit more power on the dyno. Added another 2 degrees for 17 total and it showed 13.5 with 1.5 degrees of knock.

Then I went back to the 14 total tune and it ran 12 degrees again, but showed 2.5 degrees of knock. Made the same power as it did when it ran 12 degrees and showed no knock. Weirdest part about it is it didn't show KR until halfway through the run. Timing advance logged 12 the whole way, halfway through 2.5 degrees KR pops up, it doesn't change the timing advance report though.

Not sure if it is something to do with the scanner? I am using the 2.24 scanner and 2.25 editor.

As for audible knock, after the long tube install I could hear lots of knock at part throttle. Not sure if it is just the acoustics of the longtubes or what? Curiously, when I leaned the stoich afr out to 14.7 (since this truck will never see E10) the part throttle knock got better.