Looking for help or advice with E38 surging/bucking at low throttle

[smokeshow]

I have a mild LS3 (mild cam, Mamo heads, light flywheel, supporting mods) in my Miata. I had bucking that was most noticeable in 2nd gear, around 1600 rpm, off throttle. It was *horrible* when the car was first running, the car was essentially undriveable in that region. I have it 95% gone by *reducing* timing in that area, which on my car is .08, .12, .16 g/cyl area in the spark table below 2000 rpm. I've seen some guys post that even their stock Corvette or whatever does that a bit, so I'm satisfied.

I *did* try increasing spark there, but the car wouldn't slow off throttle. It's interesting to me that it seems to work for many people ... maybe their vehicles weigh more than 2400#, so that's the difference.

Not slowing down is an airflow problem. You can technically solve it by pulling all the timing out to neuter it completely for torque production - or you can get it up near MBT where it should be. In either case, the torque delta per degree spark advance is less. That's the goal...getting out of that steep torque change region.

[blakesmiley]

I think you should try on the order of 20 degrees additional advance in the problem area. At least.

+20 will put it back around the stock table in that area. I'll go back to stock in the lower airmass rows and add from there to see if it starts clearing up. At the beginning of this I went as high 48* in the problem area with little difference, but that was before a lot of the recent changes. I'll let you all know how it goes as soon it stops raining here haha.

Will try this out.

[grubinski]

Not slowing down is an airflow problem. You can technically solve it by pulling all the timing out to neuter it completely for torque production - or you can get it up near MBT where it should be. In either case, the torque delta per degree spark advance is less. That's the goal...getting out of that steep torque change region.

By "airflow problem" you mean base running airflow is too high? The car idles fine.

Still an interesting suggestion, I may play around with it a bit ... I'm always willing to learn. Thanks!

[smokeshow]

+20 will put it back around the stock table in that area. I'll go back to stock in the lower airmass rows and add from there to see if it starts clearing up. At the beginning of this I went as high 48* in the problem area with little difference, but that was before a lot of the recent changes. I'll let you all know how it goes as soon it stops raining here haha.

Will try this out.

That looks better. I'd even extend it further into the higher airmass area. There's essentially a torque hole between 0.36g and 0.64g.

Every cam like that I've worked with has needed more spark at low speed and low load on pump gas. IVC is super late, so the charge motion is lazy and compression low...almost always benefits from extra spark to get the burn going.

In the end, you may/likely won't be able to get rid of all of the buck. Even stock vehicles will suffer from certain scenarios that excite the drivetrain just right. Its a mass and a spring, so at some point it'll eventually oscillate.

By "airflow problem" you mean base running airflow is too high? The car idles fine.

Still an interesting suggestion, I may play around with it a bit ... I'm always willing to learn. Thanks!

Not base airflow, min airflow. Its essentially an airflow floor that the throttle follows if your idle airflow falls low enough for some reason. Lot of people like to raise it because its a simple hack to get them idling...but you then suffer from the sail-on effect.

[grubinski]

Not base airflow, min airflow. Its essentially an airflow floor that the throttle follows if your idle airflow falls low enough for some reason. Lot of people like to raise it because its a simple hack to get them idling...but you then suffer from the sail-on effect.

Yeah, that's what I meant. I thought that was an idle airflow target, but I have noticed when logging that after a certain point, I could move it lower and the idle airflow would stay where it was.

Now you've got me *very* interested to lower it and see if the car will slow off throttle with more advance.

[blakesmiley]

That looks better. I'd even extend it further into the higher airmass area. There's essentially a torque hole between 0.36g and 0.64g.

Every cam like that I've worked with has needed more spark at low speed and low load on pump gas. IVC is super late, so the charge motion is lazy and compression low...almost always benefits from extra spark to get the burn going.

In the end, you may/likely won't be able to get rid of all of the buck. Even stock vehicles will suffer from certain scenarios that excite the drivetrain just right. Its a mass and a spring, so at some point it'll eventually oscillate.

I think you have me headed in the right direction! Got to try a few variations and its the best its been yet, maybe 50% of what it was.

Seems to me like VE/MAF models and fuel trim swings might be some of what's left.

You're a god amongst men, and we don't deserve you!

Will keep the thread updated with how it ends up.

more advance v3 screenshot.png

[Alvin]

I'm with smokeshow on this one.. More timing in those areas is usually what helps.

[blakesmiley]

After many more attempts the car isn't a lot better than it was. In reference to my last reply with the added advance, I think that was mostly false hope. It may have helped some, but not what I was thinking at the time. Tried upwards of 52* in the problem areas, any above that and it felt like peak pressure was happening before TDC causing more violent bucking.

The problem area is still 1400-1600 rpm between 0.17 and 0.21 g/cylair.

I did end up with it in SD with the custom OS. The VE table has some weird humps in it that don't make sense to me, but flattening them out doesn't make it any smoother and throws trims out in left field.

Went to AC Delco plugs for the LS9 with no change, then went back to the BR7EFs.

I suppose I'm accepting defeat with the surging and chalking it up to the largeish duration cam.

Currently working with the spark smoothing settings to find a happy medium. The low torque gain table (32950) almost seems to be an on/off switch below 0.400, so still trying different things to get it to be smoother.

Thanks again to everyone for your input.

Current log and tune with spark smoothing on will be attached.

If anyone thinks they could fix it remotely I would absolutely pay for that!

SD v3.9 SS on.hpt
SD v3.9 SS on.hpl

[ns158sl]

Judging by the logs, still spark related. It is literally pulling upwards of 20 degrees. Need to address that. Do a back to back log, one with spark smoothing on and another with off for comparison.

[edcmat-l1]

Disable spark smoothing. You're welcome.

In the log that's posted, and the screenshot, spark smooth is disabled. It's enabled in the current file posted because its the only thing that makes the car somewhat drivable through a parking lot.

I love this forum. Especially all the "Nah that ain't it" replies.

spark smoothing.jpg

Spark smoothing is still on.

spark smoothing and misfire.jpg

Red arrows are spark smoothing. The downward spikes are all throughout the log in the exact RPM range you're feeling the surge or buck or whatever.

Yellow circle sure does look like misfire events.

[blakesmiley]

Spark smoothing is still on.

spark smoothing and misfire.jpg

Red arrows are spark smoothing. The downward spikes are all throughout the log in the exact RPM range you're feeling the surge or buck or whatever.

Yellow circle sure does look like misfire events.

I agree with you, I think it is misfiring under only those conditions. Its perfectly smooth any where outside of 1400-1600rpm .18-.22 g/cyl. In that rpm range and any other load its ok. In that load range outside of that rpm range, its ok. It doesn't seem to be on a certain cylinder or bank. It's had several different sets of spark plugs and plug wires, currently has NGK BR7EF and AC Delco 748UU wires. Relevant compression on a scope is even on all cylinders. Compression with a gauge is within 5%, all around 95psi. No vacuum leaks that show up on trims with propane.

The working theory is random misfires from the long duration cam like smokeshow said in post #24.

I'm still hoping there is a way to fix it outside of using spark smoothing to "tame down" the oscillations, or mostly to have it not misfire. So, I'm all ears! If its something you think is fixable with tuning I'd be more than happy to pay for your time!

My next post below this will have logs and tunes comparing spark smooth on and off.


[Edit]
In my experience a real misfire (plug wire coming off, cracked plug wire etc.) shows up very clear on the wideband too, like 1.15 lambda. My wideband is in the right bank B2 1" downstream of the narrowband. I'm not seeing it do that in the logs, but would like your input on it!

[blakesmiley]

Judging by the logs, still spark related. It is literally pulling upwards of 20 degrees. Need to address that. Do a back to back log, one with spark smoothing on and another with off for comparison.

I took two logs one with spark smooth enabled, and spark smooth disabled. I tried to make them as close to each other as possible, same drive, same warm up/cool down drive before them. In both of them I used cruise control in parts and used the pedal in parts. Also, turned closed loop off with the scanner special functions. I will do my best to post time stamps of that here.
The only reason I'm trying to use spark smoothing is that it makes it somewhat driveable through parking lots and the low speed cruising areas through town.

Here is Spark Smooth ENABLED.

SD v3.12 Spark Smooth Enabled.hpl
SD v3.12 Spark Smooth Enabled.hpt
Spark smooth enabled.png

With cruise control @ 2:38
Open Loop with cruise control @ 3:15
back to Closed loop using the pedal @ 4:31

It does it through the whole log so wont be hard to see.

Here is Spark Smooth DISABLED.

SD v3.12 Spark Smooth Disabled.hpl
SD v3.12 Spark Smooth Disabled.hpt
Spark Smooth Disabled.png

With cruise control @ 2:40
Open Loop with cruise control @ 3:12
back to Closed loop using the pedal @ 5:02

[edcmat-l1]

The best results you will have with that combo is speed density with spark smoothing disabled. If you're looking for stock driveability you probably aren't going to get it.

You can now give excuses for why you want to use spark smoothing, but you've been told now twice what one change will do for your bucking. There's no denying it will remove some if not a lot of it.

[ns158sl]

I guess im not familiar with the APP never reading 0. I think I would leave spark smoothing disabled for now, and log the bucking in a parking lot, or where you have other issues. My guess is your coastdown spark is too aggressive while your Minimum airflow is too high, so they are going to fight each other.

[edcmat-l1]

I guess im not familiar with the TPS never reading 0.

That means you've never tuned an electronic throttle vehicle.

[ns158sl]

My bad, lol, APP.