Timing knock retard inconsistent

[mn_vette]

I'm trying to get my timing figured out for my car and the knock retard that I'm seeing seems very inconsistent. One run it seems fine, another I get 10 degrees. I thought I was some pretty low timing to start with at 8 degrees mid rpm and 10 degrees high rpm, but I'm still getting knock. I'm thinking it isn't real knock at this point, but I'm not sure what else to start looking at to figure out what is going on. The car is a 2012 Corvette LS3 with a vortec supercharger. Other than the supercharger kit everything else is stock, including the exhaust. The plot below was taken after I filled up the tank from almost empty with 93 octane. Any help or suggestions you guys can provide is greatly appreciated.

BadTimingSnippet.hpl

22_09_24_MOREFUEL.hpt

[151LE]

i do see under your fuel/transient/warm-up transient fuel correction/base gain is zeroed out. Try copying a zl1 table in this. This may help with big throttle stabs.

[Realizm]

I'm trying to get my timing figured out for my car and the knock retard that I'm seeing seems very inconsistent. One run it seems fine, another I get 10 degrees. I thought I was some pretty low timing to start with at 8 degrees mid rpm and 10 degrees high rpm, but I'm still getting knock. I'm thinking it isn't real knock at this point, but I'm not sure what else to start looking at to figure out what is going on. The car is a 2012 Corvette LS3 with a vortec supercharger. Other than the supercharger kit everything else is stock, including the exhaust. The plot below was taken after I filled up the tank from almost empty with 93 octane. Any help or suggestions you guys can provide is greatly appreciated.

I remember I had a similar issue, and it was driving me crazy when I switched to a D1sc. The Boost-a-Pump wasn't supplying enough voltage to the fuel pump, so I was running out of fuel pressure even though my Lambda/AFR looked fine (looked super rich, too). Maybe the flow rate was there but the pressure was crap? Not sure. My ID1050X injectors could only do so much with the lack of fuel pressure. Are you on the stock fuel pump? Which injectors and spark plugs are you using? Boost a pump?

I see you are logging fuel pressure. As far as I know, the C6 doesn't have a fuel pressure sensor, so the ECU always reads one thing (desired pressure, I guess). Unless you're on a dyno with a gauge connected to the fuel rail, it is probably difficult to read fuel pressure. Another thing is spark plugs. You'll want to make sure you're a few heat ranges colder than stock. I don't remember the NGK part number on the plugs I have now, but so far they've been working alright. I may want to go 1-degree cooler as well to see if I can push timing up a bit more. If you want to know the type I can try to find the box or part number somewhere.

As a point of reference, on E10 (93 w/ 10% ethanol), my D1sc running up to 13-15 lbs of boost (depending on weather conditions) on a 2011 C6 Grand Sport, I run about 10 degrees of timing at torque peak (~5000-5200 RPM) and increase timing to 16 degrees or so between 6500-6700 RPM. On E85, that is around 16-18 at torque peak and 20-22 at 6500-6700 RPM if I recall correctly. I'll attach my tune so you can reference my spark timings if needed.

If it isn't a fuel pressure-related issue, something you can try is dumping in a can of Boostane Professional to bump up the octane to 100+ to see if that does anything at all. I have two cans I'm not using (switched to E85). If you are in the Dallas area, you can have a can to test.

My issue was fuel pressure related (and belt slip at the time). If yours is false knock, higher octane fuel may will help identify the issue. For me, it was difficult because even with the low fuel pressure issue, higher octane did help reduce knock, but I was never able to get the timing to where 93 should be (where mine is now). Once the fueling issue was corrected, I did have to desensitize the knock sensors a little bit and disable burst knock (yours is still enabled). I wonder if your burst knock is triggering your low-octane table.

Take a look at my knock sensor settings and see what you think. I wouldn't disable them, it is good to have protection. With long tubes, and a supercharger making all kinds of noise, I'd imagine false knock is more prone to happen.

Attachment 124661

[Tunercharged]

Once the fueling issue was corrected, I did have to desensitize the knock sensors a little bit and disable burst knock

I think that this is a good way to help ensure correct timing. When the tune is happy and working consistently, I would disable or at least limit the amount that can be pulled out. No matter how tight everything is, to lose some timing while getting out of the hole because of bumps in the pavement can be very annoying. In the old days, 80s, not too many had onboard black boxes, we just went with it and once satisfied the timing that was set just did its thing. I thank technology for helping us keep our FI things together easier these days but if not running at the extreme limit is can sometimes be beneficial to let some of the built in over engineering handle the occasional slight hiccup in fueling or timing instead of counting on the knock retard sensors. Of course I’m not saying to put anything in jeopardy. I don’t turn off knock retard, I want to know if something is picked up, but my tables don’t let much timing be pulled if it is picked up. If it is real, I will change fueling or timing to fix the problem. I only use the knock retard as a tool, not as a safety system.

[mn_vette]

I took some data with the octane booster. first file is with about 1/8 of a tank, the second file is with a thing of octane booster and and extra 3 gallons of gas to mix it up a bit. It did not seem to help. So I think this, and the fact that the values are pretty low, show that I'm getting false knock. I can de-sensitize my knock sensors to try an help. Is there anything else I can do to try to find the source of the possible knock? I'm not even sure where to start. I could modify my low octane table to keep the knock from going too low, but I also don't know how to determine when I've gone too high either. I see the videos of people saying that you keep going up until you start seeing knock, but that doesn't really work when you have knock a low timing.

Realizm, I couldn't download the file you attached. It was complaining about a file extension or something. Could you try reposting it?

151LE, I did put in the transient fuel for the ZL1. I'm not sure I can tell anything better from the logs, but it feels like it has more throttle response with that. Thanks for the suggestion.

[Realizm]

Hi mn_vette,

Not sure what is up with the attachments. It says attachments not used within an hour will be removed. I'll try this method, let me know how it goes. Attached is my current tune as of yesterday night.

[mn_vette]

Hi mn_vette,

Not sure what is up with the attachments. It says attachments not used within an hour will be removed. I'll try this method, let me know how it goes. Attached is my current tune as of yesterday night.

I managed to do a bit more testing. Even taking another 6 degrees out of the timing table above 1g. This didn't do anything to dispel the knock being seen.
Next I tried to increase the timing in the Low Octane table to help limit the knock. Changing this did not do anything. The timing was still going very low.
Next step was to increase the knock recovery rate and to limit the maximum amount of knock retard. This seemed to help quite a bit. Perhaps it was just a good day for my car or this really did help, I need to run the car a bit more to find out. Next I thought about doing the knock sensors to make them a bit less sensitive. So now my questions.

1.) Knock sensor multiplier table - Does a higher number in this table make the sensors more sensitive or less?

2.) Knock sensor multiplier table values seem pretty random in the table changing quite a bit by RPM and airmass. There is no general trend like in most tables. Is there a generally "preferred" way of setting these tables or do people just do a percentage multiplier across the entire table?

3.) If I want to start increasing timing to get closer to optimum power how do I know if I've gone too far? With the false knock there I'm not sure how to know if I'm getting real knock. Is it all by feel? Data logging and using times? Do I need to do it on a dyno and look for loss of power? Or do I just stick with a conservative 12 degrees and leave it there?

Oh, and I think I forgot to answer the question about the boost-a-pump. Yes, I am running one of those with the supercharger kit and 60# injectors. I have a video of a run with a fuel pressure gauge and a voltmeter on the output of the BAP. the pressure seems to go from 60psi to 52psi at the end of the run with the BAP running at 17.7V at the device.

[Realizm]

I managed to do a bit more testing. Even taking another 6 degrees out of the timing table above 1g. This didn't do anything to dispel the knock being seen.
Next I tried to increase the timing in the Low Octane table to help limit the knock. Changing this did not do anything. The timing was still going very low.
Next step was to increase the knock recovery rate and to limit the maximum amount of knock retard. This seemed to help quite a bit. Perhaps it was just a good day for my car or this really did help, I need to run the car a bit more to find out. Next I thought about doing the knock sensors to make them a bit less sensitive. So now my questions.

1.) Knock sensor multiplier table - Does a higher number in this table make the sensors more sensitive or less?

2.) Knock sensor multiplier table values seem pretty random in the table changing quite a bit by RPM and airmass. There is no general trend like in most tables. Is there a generally "preferred" way of setting these tables or do people just do a percentage multiplier across the entire table?

3.) If I want to start increasing timing to get closer to optimum power how do I know if I've gone too far? With the false knock there I'm not sure how to know if I'm getting real knock. Is it all by feel? Data logging and using times? Do I need to do it on a dyno and look for loss of power? Or do I just stick with a conservative 12 degrees and leave it there?

Oh, and I think I forgot to answer the question about the boost-a-pump. Yes, I am running one of those with the supercharger kit and 60# injectors. I have a video of a run with a fuel pressure gauge and a voltmeter on the output of the BAP. the pressure seems to go from 60psi to 52psi at the end of the run with the BAP running at 17.7V at the device.

Going back and looking at my tune, looks like I put all the knock sensor settings back to stock but I did change stuff on the Spark retard menu. Typically though for question 1 and 2, reducing those values reduces knock sensor sensitivity.

Before you change knock sensor settings, try disabling burst knock as that is the computer's estimation of when knock may occur, then pulls timing and all that annoying stuff. For tuning timing, once your fueling is all set, create a graph the same as your spark table, generally cylinder airmass (g) vs RPM and monitor knock retard in the cells. Where you see 3 here, 4 there, 1 other places, you can reduce timing in those areas as needed. Once you fix your problem, you can start increasing timing until you hit knock and back it off a few degrees to be on the safe side.

Here is my settings currently when compared to a stock 2011 C6 Grand Sport file. Green are changes from stock.
Engine -> Spark -> Retard
Engine Spark Retard - C6 GS modified - compared to stock.PNG

Everything under the Knock Retard Amount table was cut by half. IE. 4.0 to 2.0.
My recovery rate is upped significantly. So if there is any knock and it goes away, it ramps timing back up and doesn't wait forever. It will still switch to your low octane table if you do end up getting into bad knock, so there are still safety factors in place.
Recovery Rate.PNG

Under Burst Knock, the base air vs cylair is all zero'd out.

Using 60# injectors? Are you using the Siemens-Deka injectors? Those were the exact injectors I had all kinds of trouble with for knock, fighting fueling, but I may not have had the injector data correct either. Sounds like your fuel pressure is spot on though, so that is great.

Regarding the Siemens-Deka 60LB injectors, they look like this:
Deka-60-resized-1.png

Hopefully you can get it figured out. I know it drove me crazy. Timing would pull so bad and it would lose power, my rpm would start dropping on a dyno pull around 5,000 rpm.

I'm building a PWM setup using a microcontroller and some beefy mosfets for PWM, so can try to test injectors at different frequencies. I need to get a fuel rail and a decent pump and I should be good to start testing. I'm curious if some of my Deka 60's are underperforming, not spraying correctly, etc. Anyone ever tried to ultrasonic clean injectors? That work decently?

[mn_vette]

Realizm, thanks for the feedback. I have a file all modded up to reduce the knock retard and de-sensitize the knock sensors. I haven't had a chance to go run it though. With the way work has been going this week it gets pushed down on the priority list.

The 60# injectors I have are Bosch bodies, but I believe they are modified by an injector company. My vendor was able to give me an entire spreadsheet of HP tuner tables specifically for those injectors. That fixed up quite a bit of my problems when I first started tuning.