Excess Timing in low load areas causes surge/buck issues, and how I fixed it.

[ArKay99]

I've been working through a surge/bucking issue for the longest time. I've logged and changed everything from Throttle cracker/follower to desired air to MAF to VE, etc. I was able to get it to calm down a bit but nothing all that good.

I was running around 28 degrees of timing in the area of the surge/buck and had all my tables in agreement, i.e. Spark High Octane, Base, and Low Octane. The timing didn't move much when surge/buck was occuring. I had done this because I read on the boards that to get a good idle you should run as much spark as you can to get to obtain the lowest MAP at idle. In my case 28 degrees was 'optimum' in that regard.

A friend of mine who runs a h/c LS1 and is tuned with my EFILive had similar timing at idle and the surge/buck points as I did. He decided to take them back to around 20 degrees in the surge/buck area in the High Octane table and the surge almost dissappeared. I did the same thing and got the same results. I also had a problem with stumble when transitioning to idle and that was fixed too. We were running way too much timing in those areas.

I thought about this and it made total sense. I made the idle very efficient. So efficient that the area of peak efficiency was very narrow and almost 'spiked'. This manifested as a constant need to make the Desired Air perfect or false learning occurred. My LTFT's at idle were all over the place and never lined up. I had a very acrid burning smell coming from my exhaust even though I tried to make it richer. Backing off the timing fixed all these issues on both cars. It may have been less efficient but it broadened the area of efficiency enough that low load tuning is now very easy. The car idles great, has excellent low load manners, etc.

Before this I tried to eliminate the surge/buck by making the High Octane and Base Spark close. This meant that I had very high timing in the low load areas and was making the low load efficiency of the motor too large and that was causing the surging/bucking. When I brought the low load area's of the rpm's in question down to the 18-20 degree area the car started to behave so much better. Even the clutch, a Textralia OZ700 which had some chatter, stopped chattering.

I can now take my wife to dinner in the car and hand the keys to the valet.

The throttle response has not been negatively affected because as soon as you move your foot a bit the load shifts and you are out of the cells into more aggressive ones. When I thought about it, it simply came down to the fact that there are some areas of the car's performance that I don't want running at peak efficiency. :doh2:

I guess you can say that I was relying on the recommendations of some very experienced tuners on timing issues of other cars and was not considering that those recommendation might not apply to my setup.

I'm still tweaking the tune, but now I'm doing things like optimizing the AC ramp in and out. :grin:

[Bruce Melton]

Post up your your timing table, I am interested?

[405HP_Z06]

Me too! I've had quite the opposite experience with 224 sized cams.

[ArKay99]

Post up your your timing table, I am interested?

I hope this works...
The High Octane Spark Table.
http://i35.tinypic.com/2h396qf.jpg
The Base Spark Table
http://i34.tinypic.com/2iig1v6.jpg

[Bruce Melton]

Hmmm,
Thanks, Where we you idling DMA/rpm and bucking?

I have found with a cam that "locking up" (my term) the timing below ~1200 or1400 rpms keeps the adjacent cells from "falling in a hole" (another gem) which causes bucking due to dramatic changes in timing as the DMA moves around.

That is, I take ALL the cells below ~<1400 rpm and enter something like 24* and it smooths it all out and gives better throttle response.?

Or were you coming from the stock values in surrounding cells?

[ArKay99]

I was idling in the 800 rpm row between .16 and .20. This doesn't seem to affect throttle response because the affected cells are right on the edge with regard to load, i.e. the engine is just keeping up with the drive train. With higher timing I found that the motor would try to speed up and this is what caused the surge, at least in my case.

I'll take a look for an older set of tables and post them for reference.
I was not coming from stock values, but had bumped the stock values through a series of logs to what I eventually ended up with. ~28 degrees of timing which gave me the lowest MAP. This timing is only about 2 kPa higher, but the idle and low load drivability is much more forgiving like this.

One thing I havn't tried is trying to induce burst knock by crossing the low load, low timing cells into the the higher load, higher timing cells to see if there is some consequence there.

[Bruce Melton]

Log that high octane table and watch what happens when you engage the clutch and lug around a bit at low rpms. You will hit unlikely cells that may throw the timing around?
Then again, if what you have done works, that is all good.

[ArKay99]

I had come up with the area with the lower values by doing just that. This is by no means finished and some of the values will need to be 'smoothed' into the surrounding cells, but I haven't noticed any laziness with the throttle, even when stabbing it while in those cells.

[SSpdDmon]

Not trying to steal your thunder, but here's some more helpful info...

http://forum.efilive.com/showthread.php?t=7837

I stumbled upon this last year when in the Walmart parking lot. I enabled the bi-di controls and found that timing in the 16~20* range substantially reduced the bucking in light throttle application on my H/C car. So, I had to tweak the timing maps like you did. :)

Personally, I never understood the idea of running a lot more timing in those ranges. Yes, the engine is running at a slower RPM. But in instances where people are running big cams with a nice chunk of overlap, the amount of useable fuel mixture that actually gets into the cylinders and stays there is relatively small. I think that's evident in the VE table when we're lopping off 60~80% at idle to bring the AFRs in line. If there's less fuel, there's less air as well when maintaining the same AFR. I would assume this smaller mixture won't take as long to burn, therefore you don't need to ignite it nearly so soon. In fact, you need to wait a little longer before you ignite it, which is why you want less timing and not more with a load at low RPM/low airflow scenarios. Idle is a little different (a little more forgiving) because there's no load on the engine. I wish I knew the science behind why this works a little better. Unfortunately, I don't. All I know is, it does work.

[ArKay99]

Not trying to steal your thunder, but here's some more helpful info...

http://forum.efilive.com/showthread.php?t=7837

I stumbled upon this last year when in the Walmart parking lot. I enabled the bi-di controls and found that timing in the 16~20* range substantially reduced the bucking in light throttle application on my H/C car. So, I had to tweak the timing maps like you did. :)

Personally, I never understood the idea of running a lot more timing in those ranges. Yes, the engine is running at a slower RPM. But in instances where people are running big cams with a nice chunk of overlap, the amount of useable fuel mixture that actually gets into the cylinders and stays there is relatively small. I think that's evident in the VE table when we're lopping off 60~80% at idle to bring the AFRs in line. If there's less fuel, there's less air as well when maintaining the same AFR. I would assume this smaller mixture won't take as long to burn, therefore you don't need to ignite it nearly so soon. In fact, you need to wait a little longer before you ignite it, which is why you want less timing and not more with a load at low RPM/low airflow scenarios. Idle is a little different (a little more forgiving) because there's no load on the engine. I wish I knew the science behind why this works a little better. Unfortunately, I don't. All I know is, it does work.

First of all, not stealing thunder at all. I'm just throwing the idea of this 'solution' which wasn't mine to begin with, out there for discussion. Having read your thoughts above I agee with your reasoning. I don't know the science behind it either but here is my 'take' on what's happening regarding spark advance.

When running an arbitrary degree of advance, say 28*, at 1500 rpm's the engine will have a long time to burn the mixture and hence it's 'efficiency' will be high. I'm assuming the term efficiency means the power generating potential of the engine. At light loads this efficiency is too great and the engine will produce more power than it should. This will cause it to want to accelerate. Thus at light loads we don't want the engine trying to 'rush' along. It's more desireable to have it just 'keep up'. I looked at the stock values in the High Octane Spark table and they were in the 17* range. One can make the argument that if the engine starts to speed up and you want the engine to maintain the current rpm's you'd just let off the accelerator a bit. However, I think closed loop fueling jumps in at this point and the correction causes overshoot leading to the 'oscillation'.

I think bucking is momentary surge but at a much larger degree. I also think it may be some sort of 'soft' detonation, i.e. the mixture is burnt too soon and peak pressures are reached sooner than they should. Not before TDC but sooner ATDC than it should be. On the other end of this is no or little fuel getting into the cylinder and resulting in no or a very inefficient ignition. I've had some bucking occur so bad that I thought the timing chain was going to break. i.e. a very short burst of the engine resulting in an audible 'clank' in the drivetrain.

By bringing the advance down the 'efficiency' of the engine is reduced and now it is only making enough power to keep pace with the drive line.

I once saw a spark map of a Top Fuel dragster. At the beginning of the run the advance was actually negative, like -10 degrees, then by half track it was around 20* then rose steadily above that. I'm sure they have to keep peak cylinder pressures in line much more than we do and they only need enough power to keep the tires from breaking loose at launch, Not the full 7000hp. We've all seen the results of too much power too quickly...smoke. I think this same principle applys here.

[Bruce Melton]

This is all good stuff.
The Bidi is a good way to see what the engine likes and running a bit leaner @ idle is cleaner and smoother I find. The engine will stumble however if it is too lean coming off idle under load.

On that note I attach the MSD file for our "extremely aggressive" LS7 powered road race track car which uses the LS reluctor and MAP to trigger the ignition and a GMPP carb intake to run a big Holley. The only reason we even use a curve is for hot starting which would be tough @27*. The thing idles @ 800 rpm and has nothing but go @ any throttle position or rpm < than the 7900 rpm rev limiter
.
I only mention this to show how simple the timing curve CAN be for an LS with a big ass cam. In this case a GA cam (239-251/590/106). I have tuned this cam in a streetable C5 LS7 and it runs fine but with a little less low end than more streetable cams.

3704

[SSpdDmon]

This is all good stuff.
The Bidi is a good way to see what the engine likes and running a bit leaner @ idle is cleaner and smoother I find. The engine will stumble however if it is too lean coming off idle under load.

On that note I attach the MSD file for our "extremely aggressive" LS7 powered road race track car which uses the LS reluctor and MAP to trigger the ignition and a GMPP carb intake to run a big Holley. The only reason we even use a curve is for hot starting which would be tough @27*. The thing idles @ 800 rpm and has nothing but go @ any throttle position or rpm < than the 7900 rpm rev limiter
.
I only mention this to show how simple the timing curve CAN be for an LS with a big ass cam. In this case a GA cam (239-251/590/106). I have tuned this cam in a streetable C5 LS7 and it runs fine but with a little less low end than more streetable cams.

3704
You might be able to bump up the 3-5K rpm range a little more to get you out of the corners a little faster... ;)

[Bruce Melton]

It is making ~700 hp in a 2390 lb car with no knock sensors, so the trick is application of power and not detonating. We ran like motors on the engine dyno and with no vacuum advance, centrifugal advance etc and this is about all the timing these NASCAR spec motors (LSs) can take and survive.
I guess a couple degrees might work in that range but if it is too much, really bad things will happen quickly.. For sure, more power in that range is would be unwelcome. How many more times can I spin the car and stay lucky?

http://home.new.rr.com/meltn/Monty%20LS7%20%28Small%29.jpg

[SSpdDmon]

It is making ~700 hp in a 2390 lb car with no knock sensors, so the trick is application of power and not detonating. We ran like motors on the engine dyno and with no vacuum advance, centrifugal advance etc and this is about all the timing these NASCAR spec motors (LSs) can take and survive.
I guess a couple degrees might work in that range but if it is too much, really bad things will happen quickly.. For sure, more power in that range is would be unwelcome. How many more times can I spin the car and stay lucky?


I was just being a smart-ass Bruce. ;)

BTW - You wanna help put one of those in my Camaro? :D

[Bruce Melton]

I was just being a smart-ass Bruce. ;)

BTW - You wanna help put one of those in my Camaro? :D

Sure, but think of it in your CBalt !!-
However-
I have to warn you that one of the advantages of our sophisticated GM PCM driven control systems is fuel economy. This thing can leave a month's carbon footprint in an < hour.