Bucking & overshooting idle

[JimMueller]

Hi All,

I have some remaining issues from a pro tune which the tuner can't seem to resolve. I'm attaching two logs.

coldstartACoff: Cold start until 235*F, then fans on until back to 185*. Ignore AFRerror as I used the wrong AFRCommanded PID in this log. Fixed in the drive log.

drive: Driving around trying to log the symptoms. AC off for the first 10 minutes, then AC on the remainder of the drive.

Frame 259, 340, 371 & 421: overshoots idle but recovers on clutch-in quick stops
Frame 2530: clutch-in quick stop idle overshoot in traffic
Frame 4530: Bucking
Frame 5460: Bucking on decel off gas, overshoots idle RPM
Frame 8445: Engine sensor fuse blows
Frame 11085: Bucking on decel off gas
Frame 11229: Severe bucking on light throttle tip-in
Frame 11520: Bucking
Frame 11670: Bucking

The car is a 98 Z28 with a LS3 longblock w/ 24 tooth reluctor, LS3 injectors, 13lb flywheel. Engine sensor fuse regularly blows due to a short we haven't been able to locate. IAT sensor is in the stock location.

Concerns:
- Bucking at off throttle / light throttle transients
- Idle timing tables don't mesh with the main timing tables
- Seems like my fuel economy isn't very good, but I guess that's subjective
- While I didn't try to reproduce it in these logs, turning on the A/C can sometimes make the motor stumble/die not warmed up. It can happen at idle, or if clutching-in preparing to stop; it's more likely to occur during a stop the harder I stop.

I haven't tried the EOIT adjustment because I am not clear how I apply the spreadsheet numbers to my tables. But if it helps, my EVC & IVO are both 1, IPW @ idle is 3.5ms.

I'd appreciate any detailed information I can use to fix the remaining issues without incurring new issues. I tried a highly recommended tuner over a year ago but that didn't turn out as well as I had hoped

Thank you very much for your help.

Jim

[WS6FirebirdTA00]

Just glancing over your comments real quick, you need to modify the AC torque tables to bump up AC airflow if the idle trims are + with the AC on.

[JimMueller]

The tuner said he had already bumped up the AC torque table more than he had for any other car. Do I just keep bumping it up by a certain percentage across the whole table? Is there such a thing as having it too high and it causing other side effects?

I took my first stab at trying to mesh the main & idle tables but it doesn't look as if it's any smoother at idle. Is it adequately smooth in the new log/tune else which method do I use to make it smoother?

I didn't notice as much bucking in the teen's, but have a few large dips past idle coming to a stop.

[Nitroused383]

Is this a bone stock LS3? Was a case learn performed when the engine was installed?

[JimMueller]

The only thing stock in the rotating assembly is the crank. The top end has aftermarket valvesprings but is otherwise stock internally. It has a NW 92mm TB, TSP 105mm MAF and the matching aftermarket screen. The motor was installed June 2010. I don't recall specifically doing a case learn after the engine was put in the car, but I'm pretty sure I did.

[Nitroused383]

What size cam?

[JimMueller]

222/230 112+2

[MikeGyver]

I can't look at your logs on my mobile, and I don't know if it applies, but this concerns your TB. I have a FAST 102 DBW, and I see that it looks similar to a NW. I had to hog out the idle air passages as much as possible to get good idle air control.

[Nitroused383]

Your open loop fueling table isn't commanding stoich, it looks to be commanding 13-14 AFR (1.03-1.06 EQ). Set that entire table to 1.0 above 158 degrees. You should log IAC counts and see what you come up with at a hot idle, should be around 40-70 counts. Throw tps % in your scanner too. Sounds like you have excessive airflow so I would search how to do the russk idle airflow and redo your base running airflow.

LS3 cars do not like much timing at idle and light load. I would setup your idle table around 15 degrees of timing. I would lower your main spark table cruising cells too, below I attached a picture of a timing table you should try. After you get your base running airflow dialed in you will need to do some fine tuning of your fueling. Looks like you're running open loop MAf, so your adjustments will need to be made to the MAF transfer.

Also your idle over / under speed tables are very agressive and not helping your situation at all, they are amplifying your idle issues. Below is also an over / under speed values that you should try, I realize they are not the same increments as your 98 tables but try to match them up in the appropriate cells.

[Russ K]

You need to diagnose why the engine sensor fuse is blowing. Also your Maf fails at frame 8445.

Russ Kemp

[JimMueller]

I can't look at your logs on my mobile, and I don't know if it applies, but this concerns your TB. I have a FAST 102 DBW, and I see that it looks similar to a NW. I had to hog out the idle air passages as much as possible to get good idle air control.

I purchased the recommended TB & MAF which the pro tuner recommended, based upon his exposure and/or limit of problems. I'm not crazy about hogging out passages.

Your open loop fueling table isn't commanding stoich, it looks to be commanding 13-14 AFR (1.03-1.06 EQ). Set that entire table to 1.0 above 158 degrees.

I set OLFA colums 158*-284*, all rows, to 1.00.

You should log IAC counts and see what you come up with at a hot idle, should be around 40-70 counts.

This PID is included in the coldstartACoff log in the very first post. I let it idle up to 235*, where the IAC was then 48, but is usually mid-50s to mid-60s.

Throw tps % in your scanner too. Sounds like you have excessive airflow so I would search how to do the russk idle airflow and redo your base running airflow.

These PIDs are included in the coldstartACoff log in the very first post. Filtering the histogram to only include TPS<1 and ECT<235, the airflow is between -0.14 & -0.26 between 140* and 248*. So unless there's a reason we want it farther from zero, I think the RAF is pretty close. I don't see any instances where TPS>0 in the idle log.

LS3 cars do not like much timing at idle and light load. I would setup your idle table around 15 degrees of timing. I would lower your main spark table cruising cells too, below I attached a picture of a timing table you should try. After you get your base running airflow dialed in you will need to do some fine tuning of your fueling. Looks like you're running open loop MAf, so your adjustments will need to be made to the MAF transfer.

Looks like you adjusted my file in that snapshot. I modified my high octane table to match, then copied with labels to low octane and both idle spark tables.

Also your idle over / under speed tables are very agressive and not helping your situation at all, they are amplifying your idle issues. Below is also an over / under speed values that you should try, I realize they are not the same increments as your 98 tables but try to match them up in the appropriate cells.

I zero'ed my current tables and pasted in your numbers, integrating the missing values.

You need to diagnose why the engine sensor fuse is blowing. Also your Maf fails at frame 8445.

I noted that the engine sensor fuse (and thus anything on that circuit such as the MAF) failed in my original post at the same frame number. I have a local guy who worked for GM as the last resort to repair vehicles before they were lemon lawed. I'm having a horrific time getting him to honoring his word on hooking me up with a lift and fixing it. Dealer wants an open budget to isolate the problem.

[MikeGyver]

I purchased the recommended TB & MAF which the pro tuner recommended, based upon his exposure and/or limit of problems. I'm not crazy about hogging out passages. ... I let it idle up to 235*, where the IAC was then 48, but is usually mid-50s to mid-60s. ...

Yeah, disregard my comment. My IAC stayed at 310 until I made the holes bigger.

[Nitroused383]

My bad I totally over looked the IAC counts, they appear to be in the ball park. I was refering to add the TPS% to your driving log. Do another driving log but include idle cracker and follower airflow. You will definitely need to figure out whats up with the fuse issue. Do another driving log and post it up.

[JimMueller]

I guess I missed the thread notification, as I just found your latest reply today. I'm confused on the TPS%, I already have TPS in the log, am I supposed to use TPS(SAE) instead? I'll add that, idle cracker & follower airflow to the scanner config I posted on 8/24. Anything else I can add/remove to make the updates quicker? What's your recommended method to smoothing the VE table? Highlight it all and smooth just once?

I'm attaching the tune I'm currently using, and I know the VE needs to be smoothed, but I figured I'd do that just once after the bulk of the cells are <2% off.

[jimmyblue]

I disagree with changing the A/C torque tables. The A/C is what it was
and it takes the same torque under the same conditions as ever.

What's jacked is the motor torque model vs air, spark, fuel @ RPM. This
back-drives IAC airflow. We can't get at this model as far as I can see.
Other than by VE table. Maybe it's telling you that the VE table is holding
the bag for some other low-end errors (like injector offset).

One thing nobody seems to look at is the IAC PID (proportional, integral,
derivative) settings. I've played with these cluelessly, I'm not a control
theory guy, maybe we've got one and they'll step up. But anyway, for
surgey behavior I have bumped up the derivative values and seen the
thing behave better (fights fast changes in idle air). Might be better to
leave it sporty on the too-little side and just fix the too-much side
(not let PCM drive IAC down so quick just in case).

PID controls are used for all sorts of machinery and process control and
can be tuned for just about any system. If you know the characteristics,
which we unfortunately don't. That makes it cut-and-try.

Reducing the integral gain can also help, too much integral makes any
loop unstable. Cutting I then just takes longer to learn to null-error.
That's probably more acceptable than overshoot (windup). From the
little reading I've done you get P right (P alone will leave a finite error),
then I (for learning without overshoot) and then you cover up your
residual mistakes in time-domain behavior with D.

[Nitroused383]

Did you get a chance to do a driving log scanning the added parameters I mentioned? How is the car driving now with the lower timing table? Sorry I had my scanner setup to show the SAE TPS so it was not showing up in the graphs.

[JimMueller]

Here's the current tune and the full driving log from key on to key off this morning. I haven't made any adjustments based upon this log.

I haven't loaded the former tune for a true A-B comparison. But from what I recall, the current tune...

1) wants to idle at a lower RPM but the idle MAP is 5-10kPa higher
2) seems to not quite have the same throttle response off idle
3) still has similar bucking when TPS around 5% or less and very minor throttle adjustments up through the upper teen RPM's
4) I'm rarely using the A/C because of being irked about not knowing how to isolate and resolve it

Within the first 2 minutes on this log you'll see it overshoot idle and surges to try to recover as I'm slowing down to go over speed bumps. There is warm bucking around frame 4580. The AFR seems to have gotten out of line after performing a single smooth on the entire VE table.

[Nitroused383]

Your Base running airflow looks to be a bit low, try bumping the entire table up .5 g/sec across the entire table. Or just redo your idle airflow after you get your fueling in more in line. I noticed in your 1st cold start log your afr was around 13:1, get your afr around 14.5-15 at an idle. Doing this will change your idle airflow needs. Your Open loop fuel adder table is quite rich at your startup temps (90 degrees), I will include a picture of a revised table. If you think the lower timing didn't help with the bucking at all than revert back to your previous timing tables. Your current tune appears you are running off of the maf sensor, so smoothing the VE table will have no affect on your afr. You need to get your afr dialed back in line by adjusting your maf table. Your MAF also fails around frame 4008, you really need to figure out whats up with that. After it fails you are than running off the VE tables.

Once you get these other things dialed in if you are still having issues with the rpm dropping too much play around with the throttle cracker table. Add airflow in the 1000 rpm column, start around 1.2 g/sec. If it still drops too much keep adding airflow until the rpm hangs, than back it down.

I think one of your a/c problems may have to do with idle > airflow > cooling fan airflow tables fan 1 and fan 1 & fan2. Log desired airflow and pay attention to the number while idling. Using the vcm controls turn on fan 1 and record the change in desired airflow, enter this into fan 1. Repeat the same thing for fan high (fan 1 + 2). When you turn the ac on your fans automatically kick on and your fan airflow values look pretty low. This may or may not help but give it a shot.

[Rinkrat456]

Nitroused383, the required additional cooling fan airflow is subject to alternator load, correct? I ask because I have seen folks on this site and on others who have the same Lincoln Mark VIII 18" fan as I do and their Fan 1 and Fan 1&2 airflow values are much greater than mine. I have a 280A alternator with a smaller pulley than what was intended on my main project truck (off road winching, off road lights and dual batteries) so my desired airflow barely sees any difference (.008lb/min) when the fan's low speed kicks on. ECT's have never gotten hot enough where the fan's high speed has been needed but when I manually trigger high (which I haven't logged desired airflow for this yet) the factory .026lb/min values still seem to hold the RAF just fine. Others who have my fan (28amps low speed 3300cfm and 35amps high speed 4500cfm) seem to have airflow values for their fan settings 4x that of my fan values.

Long question short...cooling fan 1 and fan1&2 settings just have to be what the desired airflow is asking for, correct?

[Nitroused383]

If desired airflow is 7 g/sec at an idle with no fans on and you turn fan 1 on it bumps up to 7.2 g/sec you need to put .2 g/sec in the fan 1 airflow table. Same deal for the high fan (both fans 1 + 2) if it jumps to 7.6 g/sec than you need to put .6 g/sec in the fan 1 + 2 table. You can switch the unit values lb/min, lb/sec, lb/hour, or g/sec by clicking the green lettering to the right of the table. I think you're on the right track about the alternator load.