8.1l (LS1-B) Pulling Timing Down Well Below Commanded/Corrections

[chevyspence]

Hello all. The engine in question is an 8.1l vortec with minor internal modifications, slight over-bore, mildly ported heads, and small cam upgrade, 4l80E trans (also pcm controlled).
I'm trying to gather as much info as possible while trying to figure out which spark correction PID's are actually applicable and correct for this application without logging 30+ channels so please bear with me
First to just address it ahead of time; yes, the log does show that I am seeing some Knock Retard at various conditions, but am currently working that out separately and it does not seem to the culprit of the issue I am looking into here.
Timing when rolling into the throttle hard from idle is dipping well into the negative values (sometimes 30deg less than the main spark table!). When I account for all of the corrections that seem to correlate as they should, they still just don't add up to that timing difference and it seems there is a hidden force at work here...

"Trans Torque Reduction Spark Retard" (GM.EST XMSN DMA) seems to be very active in showing spark retard, but does not seem to correlate to actual timing advance (timing comes back up after the dip even though trans spark retard stays high). The tune file also has all of the torque control (aside from shifts) set to disable and table values zeroed out.

A log file example of the issue and the current tune file are attached. Frame numbers of most interest in the log file are: 826, 1131, 1816, 1922, and 2558 (in that last one, I loaded the engine up some first before throttling all the way up to WOT)

Next logs, I plan to add "EST Base Spark" and "EST Run Spark" to try to gain some more insight.

Anyone have any idea of what may be killing my timing here, or any parameters or tables I may be missing or need a closer look at?

Thanks!

Spark Retard Example 1.efi
040716 modT - DFCO.ctz

[Jetmech442]

I helped my buddy with his 8.1,Completely stock engine. There is a table that has a range of - 64 to 64. Print this table at 0 makes a world of difference. Let me search through my history real quick to find it. (I don't have my laptop handy...)

Sent from my Nexus 6P using Tapatalk

[chevyspence]

Cool, would be interested to hear what that was, thanks!

Another thing I'm looking at now, is since I'm mostly seeing these huge deviations in quick transient conditions, I think having to log more than the recommended number of channels might be messing with me (some parameters slightly lagging vs. others, so the table correlation looks off), so I'll look into that too.

[Jetmech442]

Yeah, 24 is the max channels before it slow down. That's channels, not pids. Some pids use up 2 or more channels.

Sent from my Nexus 6P using Tapatalk

[Jetmech442]

Okay. I haven't been able to look at the Logs yet, but I found the map I was talking about. It's B6614(Max torque by rpm) . From the factory, it was maxed out at 640 lbft. Setting it to 0 actually allows max torque. Seeing to - 640 kills power and you can only idle around a parking lot. This effected certain Operating systems. When I get to my comp I'll let you know the OS my friend had.


Sooo. That probly will help you make more power, but doesn't address your timing mismatch.

It sounds to me (without looking at the Logs) that burst knock is coming on when you get into the pedal. You can desensitize it by adding to b5954(burst knock delta air mass). Generally, burst knock pulls around 6 degrees when at operating temp.

Also, you should paste you high octane spark map (b5913) into your low octane. That way you don't bounce between them due to the octane scaler.

Sent from my Nexus 6P using Tapatalk

[Jetmech442]

You are running OS 12587603. I was running 12606807. This issue may not affect you, but I'd suggest exploring it.

I had a short thread about B6614 here :https://forum.efilive.com/archive/in...p/t-24349.html

Other than that, I looked at the logs and have a few observations....
Your big 8.1L consumes ~.95 g/cyl at WOT. your timing under that first WOT pull is 5 degrees lower than expected from High octane map. HO commands ~20 and you get 14. Low Octane commands 17, so not too far off. That initial dip is probably the trans maps pulling timing for the downshift(you could probly confirm that since you were driving). I think some Burst knock is alos coming into play as I see it grabbing about 5 degrees.

So, no legitimate knock during the WOT run, at least that I can see.

All of my experience is with my 6.L LQ9 engine. my timing is around 26 degrees( and I get all of it), but I only consume .72 g/cyl. Maybe 20 degrees is enough since you are injecting so much more air, but maybe you need more...? I would make the High and low octane spaark maps identical and run good gas(or record the octane scalar PId), that way oyu know which map the commanded timing is coming from.

B3605/ When you are in PE, your only commanding .90 enrichment. I think most people shoot for .82-.85.

Also, you said you did a cam/head work.Have you completed CALC.VET to get your AFR correct?

[chevyspence]

Thanks for the info on that Max torque by RPM table, I'll have to see if that has any effect on mine also!

The WOT tip-ins were all done from a stop (since I was mostly seeing the timing dip at really low RPM) so I should be able to at least rule out the shift spark retard.
As you mentioned, I was definitely getting some Burst Knock retard when I tipped-in very quick but it strangely seemed to kick in just after the timing dip, so looked like it wasn't really contributing to the initial dip.

On the fueling side, I did go through and tune the VE and MAF tables and the dynamic fueling with a serial wideband and now just run in open loop. Yeah, I am running on what I would consider the lean side too; I tow a big heavy race trailer on long trips and am basically in the upper end of the MAP scale for a good deal of the time (especially in the mountains) so I shoot for ~13.2 AFR where I found the best balance of power and reasonable fuel usage. I am fairly comfortable there since, oddly enough, it looks like most 8.1l factory tunes seem to prevent the use of any power enrichment at all!

I dove deeper into my suspicion of some PID's lagging, and its starting to look like even with less than 24 channels logged, the RPM PID is reacting "realtime" but the g/cyl is lagging one frame behind. If I look at what the Ignition Advance PID value is at a certain frame and compare it to the timing table for the current frame g/cyl and the previous frames RPM it usually matches almost exactly considering all active corrections. Basically to fix the timing dip, it looks like I just need to raise/flatten out the timing curve in the upper g/cyl range that I didint realize I was getting to before the RPM started to raise.

Thanks for the tips... now i'm curious and need to check on that max torque table haha.