VE Tuning w/o Wideband?

[ferocity02]

Is it possible to do some rough VE tuning without a wideband? The VE tutorials I've read all require a wideband. I'm trying to dial in my tune at idle just to get this thing to pass smog. Right now it's putting out 36% too much HC at idle, but is 92% under the HC limit at 2500 RPM. It is also misfiring slightly at idle which is probably the issue. I'll get a P0300 after a couple minutes of idling. The misfires are spread out evenly across the cylinders. The tune is stock at the moment.

[ScarabEpic22]

Ok, so you have a cam but havent tuned it at all? You NEED a wideband, like go buy one tomorrow. If you mod the engine at all, you need to adjust the VE/MAF tables and Im frankly surprised your engine runs on the stock tune with a cam. You could be super lean at idle which is causing your excessive HC and your misfire.

You can try doing Calc.VE using LTFTs, its ok for a rough (really rough) start.

I apologize if I sound like a jerk, but you really need to get a WBO2 on it asap. You can rough in a VE/MAF table using Calc.VET in a few hours tops, be done with all of it and pass emissions.

[ferocity02]

Thanks for the tips. I do have a wideband, but it's not installed yet. I'm hoping to get this thing to pass without spending money and time modding my stock exhaust system or doing in depth tuning. In a couple weeks when the turbo goes on everything will change, hence not wanting to put too much effort into it. The wideband will go on with the turbo.

The cam isn't all that big. It runs great other than the misfire issue. But I do baby it around as I know the stock tune is not ideal.

I'll try and do the Calc.VE using the tutorial.

[joecar]

If you're getting misfires at idle, you may be overly lean or overly rich...

unburnt fuel means that unburnt oxygen is present, the NBO2 sensors see this as lean, so the PCM keeps adding fuel (too much fuel).

Or there may be other causes of misfire (coolant leak).


At idle, the PCM may be using either of MAF or VE.

Using the scantool->tunetool cursor linking feature can you determine which cells in the various tables are being referenced (B0101, B3605, B5001).

Try changing the O2 mV switch points.

[ferocity02]

What do you suggest for O2 switch points? The only thing I've read is to set them all to 450 mv.

My O2 activity at idle is not pretty, they appear to be lazy as you have mentioned before. My guess is they are acting lazy because of the misfires.

And what about idle timing? Right now my base table is calling for 18 deg at idle. It varies fairly erratically at idle, from 9 to 24 deg. Tomorrow I may try to add timing using Bidi and see if it helps the misfires. If I fix the timing at idle then it starts to surge +-150 RPM from commanded.

[ScarabEpic22]

I understand not wanting to spend a lot of time before turbo'ing, but the solutions is pretty simple: spend the 2-3 hours, Calc.VET it, pass emmissions.

Id do that, then once you have the A/F done you can tweak spark timing at idle.

Sometimes the only way is the way we've been avoiding, you're going to spend more time trying not to tune it than if just tune it and being done with it.

[joecar]

What do you suggest for O2 switch points? The only thing I've read is to set them all to 450 mv.

My O2 activity at idle is not pretty, they appear to be lazy as you have mentioned before. My guess is they are acting lazy because of the misfires.

And what about idle timing? Right now my base table is calling for 18 deg at idle. It varies fairly erratically at idle, from 9 to 24 deg. Tomorrow I may try to add timing using Bidi and see if it helps the misfires. If I fix the timing at idle then it starts to surge +-150 RPM from commanded.Possibly, or, the O2 activity may be causing the misfires.

Try timing and see what happens to misfires (good idea), and if you now get surging, the you will need to tune in this area.



More info:



Idle Tuning
showthread.php?4661-Idle-Tuning-Helpful-Info-Inside (http://forum.efilive.com/showthread.php?4661-Idle-Tuning-Helpful-Info-Inside)
showthread.php?14439-Idle-Tips-amp-Tricks (http://forum.efilive.com/showthread.php?14439-Idle-Tips-amp-Tricks)
showthread.php?5866-Auto-VE-questions (http://forum.efilive.com/showthread.php?5866-Auto-VE-questions)
showthread.php?14153-scan-tool-wont-log-rafig-or-rafpn (http://forum.efilive.com/showthread.php?14153-scan-tool-wont-log-rafig-or-rafpn)
showthread.php?14435-Need-help-Can-Start-Car-but-dies-when-I-give-it-any-throttle-before-going-to-CL (http://forum.efilive.com/showthread.php?14435-Need-help-Can-Start-Car-but-dies-when-I-give-it-any-throttle-before-going-to-CL&p=129519&viewfull=1#post129519)
showthread.php?7011-ETC-cars-and-IAC-parameters (http://forum.efilive.com/showthread.php?7011-ETC-cars-and-IAC-parameters&p=61455&viewfull=1#post61455)
showthread.php?14544-Working-on-my-idle-tune (http://forum.efilive.com/showthread.php?14544-Working-on-my-idle-tune)
showthread.php?14794-Help-with-open-loop-%28cold-start%29-tuning (http://forum.efilive.com/showthread.php?14794-Help-with-open-loop-%28cold-start%29-tuning&p=133446#post133446)
showthread.php?149-Idle-%28Transition%29-Tuning (http://forum.efilive.com/showthread.php?149-Idle-%28Transition%29-Tuning)
Stumble-3-seconds-after-cold-start&p=178921#post178921 (http://forum.efilive.com/showthread.php?14144-Stumble-3-seconds-after-cold-start&p=178921#post178921)

More Idle Tuning
showthread.php?t=149 (http://forum.efilive.com/showthread.php?t=149)
showthread.php?t=5866 (http://forum.efilive.com/showthread.php?t=5866&highlight=RAFIG)
showthread.php?p=86553 (http://forum.efilive.com/showthread.php?p=86553) post #17
showthread.php?t=2630 (http://forum.efilive.com/showthread.php?t=2630)
showthread.php?t=473 (http://forum.efilive.com/showthread.php?t=473)

Throttle Cracker/Follower
showthread.php?t=3568 (http://forum.efilive.com/showthread.php?t=3568)
showthread.php?t=4081 (http://forum.efilive.com/showthread.php?t=4081)
showthread.php?t=5406 (http://forum.efilive.com/showthread.php?t=5406)
showthread.php?t=5940 (http://forum.efilive.com/showthread.php?t=5940)

[ferocity02]

Did one round of CalcVE this morning as a practice. Seems to be working okay. Should I be resetting the LTFT's before starting?

I also messed with the timing at idle and found that reducing the timing for the most part eliminates the idle misfires. My base table calls for 18 deg at idle. Using DVT I subtracted ~10 deg and the misfires went away. It seems to be able to maintain the commanded idle speed better as well, +-20 RPM at the most. Before it was +-50 RPM. The exhaust smells noticeably better too. The idle sounds rougher and deeper with the reduced timing. Any downside to running the idle timing this low? Research shows people running 20-30 deg timing at idle.

[ScarabEpic22]

Yes, reset LTFTs between log/tune sessions.

Not sure on idle timing, really more of what your engine likes.

[ferocity02]

Ok, I've been reading Calc.VET threads up the wazoo... lots of useful information but I wish it was all in one place for easy reference. Again, I'm doing this without a wideband just to get the VE table roughly close enough to pass smog. Do these steps sound correct, assuming the calc pids and maps have been correctly setup?


Tune mods:
1. B0701: Disable Catalytic Converter Protection
Insures accurate stoich and PE mode/WOT Fueling
During tuning only

2. B0120: Change RPM Threshold for Airflow Calculation from 4000 to 400
Eliminates any Airflow Correction from the VE Table
During tuning only

3. B3308: Disable DFCO: Change B3308 (M6) to 140C Change B3313 to 140C
Accurate fueling computations
During tuning only

4. B4105: O2 Switch-points to 450 Millivolts
Smoother MAF & VE Table
Permanent

5. B3618: PE Modifier Based on Rpm (EQ): 1.16
Safe adequate Fueling regardless of Fuel Type
Permanent

6. B3616: PE Enable: make sure PE enables as load becomes significant
(e.g. below 60% TP below 3200 rpm, 35% TP above 3200 rpm).
Permanent

7. B3608 and B3609: PE Delay: set these to zeros.
Permanent


Logging:
Reset LTFT's before logging
Log data
Filter data to exclude for ECT, low cell count, high TP% change


Adjusting tune:
Copy and paste calculated VE table into B0101
Smooth table
Copy and paste (multiply) calculated LTFTBEN table to B5001
Smooth table

Repeat logging and adjusting tune steps until B0101 and B5001 aren't changing much. Revert tune changes 1-3 back to original values.

[joecar]

...

I also messed with the timing at idle and found that reducing the timing for the most part eliminates the idle misfires. My base table calls for 18 deg at idle. Using DVT I subtracted ~10 deg and the misfires went away. It seems to be able to maintain the commanded idle speed better as well, +-20 RPM at the most. Before it was +-50 RPM. The exhaust smells noticeably better too. The idle sounds rougher and deeper with the reduced timing. Any downside to running the idle timing this low? Research shows people running 20-30 deg timing at idle.This looks like what your engine (and its O2's) like, and hopefully this will get you smogged...


do you have any log files of this.

[joecar]

Smoothing: don't overdo this, mostly just flatten out any spikes/holes.

[ferocity02]

This looks like what your engine (and its O2's) like, and hopefully this will get you smogged...

do you have any log files of this.

Haven't made a log file for this. I may try today.


Smoothing: don't overdo this, mostly just flatten out any spikes/holes.

Will do! Do the steps otherwise look right? Anything I'm missing? I really appreciate your guys' help. Frankly I'd be lost without your tips.

[krisr]

If the car already has O2 sensors in them, why not just pull 1 out and swap it with your Wideband then correct using WB BEN's?

Disable that O2 sensor DTC/MIL codes. Tune car using wideband for couple hours as said and dial it back to 1.0 lambda or whatever your target is. Swap sensors back, enable DTC/MIL codes. Re-enable STFT/LTFT if needed trimming back to 1.0EQ.

I did this when dialling in my mates LS2. Worked great and didn't modify the exhaust at all.

[ferocity02]

Ok, did one round of Calc.VE on the drive home. I had to use 165F for my ECT filter because with the 160F thermostat it will run 167-172F if it's moving. I did not worry too much about high RPM stuff since I'm just trying to get the idle about right.

VE_0000 is the stock '01 5.3L LM7 VE table I started with
CVE_0001 is what I calculated on the drive home, filters on
VE_0001 is the new VE table without smoothing
VE_0001S is the new VE table smoothed
MAF_0001 is the the new MAF calibration table, didn't change much, doesn't need smoothing

For the 400 RPM row I used the 800 RPM row minus 3%, is that okay? I really only smoothed around the outer edges with the stock table, and knocked leveled out a few high/low spots. What do you think? I may do another round tonight.

Is it worrisome that my VE values are much lower than stock?

The idle cell (800RPM / 45kPa) was about 19% lower than stock. Might that cause some of my idle issues like the misfiring?

[ferocity02]

If the car already has O2 sensors in them, why not just pull 1 out and swap it with your Wideband then correct using WB BEN's?

Disable that O2 sensor DTC/MIL codes. Tune car using wideband for couple hours as said and dial it back to 1.0 lambda or whatever your target is. Swap sensors back, enable DTC/MIL codes. Re-enable STFT/LTFT if needed trimming back to 1.0EQ.

I did this when dialling in my mates LS2. Worked great and didn't modify the exhaust at all.

I suppose that's an option. How would having the wideband on just one bank affect things?

[krisr]

No different to tuning it on a dyno with only 1 wideband sensor. Good idea to take the usual steps before tuning and make sure the engine's serviced - make sure your plugs are in good condition, filters are clean, a good batch of fuel is in the tank, etc..

[ferocity02]

Having an issue... I finished the first run of Calc.VE and flashed the tune back in. Then I went out and logged some more and my LTFT's are still poor. The idle LTFT's got even worse, 10.2% and 5.5% now, before they were 5.5% and 2.3% The second run of VE values are similar to the first run but some differ up to 8.5%. The idle cells differed by less than 1%. Seems like the VE table is getting honed in, but my LTFT's are going the wrong direction. What might I be missing here?

I followed the process I posted earlier in post #10. I also disabled EGR and zeroed the EGR spark correction so it wouldn't screw with my cruising cells. And I also turned the base spark table speed enable up to 256 mph from 4 mph so it's always using the base spark table when the throttle is closed.

I've attached the logs, 0027 is the first run, and 0028 is the second.

[joecar]

MAF looks like it's already tuned.

Thanks for carefully naming (using short names) and explaining the pics, that makes it easier to understand :cheers:


I'll look thru your logs later today...

did you apply the transient filter...?

in the meantime do this:
- in tunetool, go Edit->Properites and set VE units to g*K/kPa;
- in scantool, for maps/charts use CALC.VET with units VE (g*K/kPa);

using g*K/kPa units is easier than % units (you don't have to enter displacement).

[joecar]

Some widebands (like the LC-1, LM-1, LM-2) provide a narrowband output signal (to the PCM) concurrent with the wideband signal.

[ferocity02]

For some reason my PID Calc.VE_Table is only offered in %. I think it has to do with how my calc_pids.txt is setup. I copied it as it was in the tutorial, but I feel like there is a line missing for g*K/kPa. So I am using % for now. Attached is my calc_pids.txt file. I used 5.293 for my displacement (5.3L LM7).

I used a filter to exclude cells with more than 5% change in TP in 100ms, if that's what you mean by transient filter. And I'm excluding cells with an ECT under 165F. These are the only two filters I'm using. For the most part I'm trying to stay out of PE, mainly because I'm not interested in it right now, and I'm not sure if PE can be tuned without a wideband.

My MAF calibration B5001 was stock when I started.

Thanks again joe!

[joecar]

You have the wrong calc pid... the one you want is CALC.VET...

see the calc_pids.txt (http://forum.efilive.com/attachment.php?attachmentid=14575&d=1360953394) file in post #1 here: Calc-VET-correcting-MAF-and-calculating-VE-(in-single-log) (http://forum.efilive.com/showthread.php?15236-Calc-VET-correcting-MAF-and-calculating-VE-%28in-single-log%29)

[ferocity02]

I am using this tutorial: http://download.efilive.com/Tutorials/PDF/Calc.VE%20Tuning%20Tutorial.pdf

What is the difference between Calc.VE and Calc.VET?

Is my calc_pids.txt file okay other than it's missing g*K/kPa? The equations look a bit different too.

[joecar]

Calc.VET is an update of Calc.VE.

The Calc.VET thread contains the most upto date info, it's a lot of effort updating a pdf, it is easier to keep the thread updated instead.

Just copy the calc_pids.txt from the thread into your User Configuration folder, and go again...

if you don't have a wideband yet, edit calc_pids.txt as follows:

replace this:


*CLC-00-032
factor 0 4 .0 "{GM.EQIVRATIO}=1"


with this:


*CLC-00-032
factor 0 4 .0 "1"


this will use LTFT's only.

[ferocity02]

Will do! I'll try this in an hour or so. Man I really wish this info could be compiled into a new tutorial or something.

I'll post results later tonight. Thanks again!

[ferocity02]

Ok, just did some logging and I think it worked. There was one issue which was that "Calculated VE, not corrected" (CALC.VEN) would not validate because "Dynamic Air Temperature" (CALC.DAT) was not selected. In the Calc.VET link it does not show CALC.DAT being a selected PID in the screenshot. So I ended up selecting it anyways, is this correct?

The results were similar to what I got with Calc.VE. One thing that worried me is why my idle cell (800RPM/45kPa) is much lower than those around it.

CALC.VET PIDS are the pids I selected
CALC_VET 0001 is what I got with my log
SELBEN 0001 is also what I got with my log (pasted it into Excel to show more sig figs)
VET 0001 is my VET table with the new data pasted in
VET 0001S is the smoothed version, mostly cleaned up around the edges.

The MAF table stayed pretty smooth.

[ferocity02]

And the log.

So unless I'm mistaken, I should be able to flash the tune back in with the new VE table and MAF calibration, not changing anything else (including keeping the RPM Threshold for Airflow Calculation B0120 at 400 RPM), and the LTFT's should be near(er) to zero while I do another log?

[joecar]

Ok, just did some logging and I think it worked. There was one issue which was that "Calculated VE, not corrected" (CALC.VEN) would not validate because "Dynamic Air Temperature" (CALC.DAT) was not selected. In the Calc.VET link it does not show CALC.DAT being a selected PID in the screenshot. So I ended up selecting it anyways, is this correct?
...Yes, correct.

[joecar]

...
The results were similar to what I got with Calc.VE. One thing that worried me is why my idle cell (800RPM/45kPa) is much lower than those around it.

...Idle with a cam plays havoc on the O2 sensors...

unburnt oxygen causes the O2 to register unburnt fuel as lean so the PCM tries to add more fuel...

so try experimenting with the VE table at idle, give it more air (which gives it more fuel) and see how the engine runs, then give it less air and see which the engine likes better...

see if you can smell unburnt fuel at the exhaust tips, this will give you guidance.

[ferocity02]

Makes sense. So if I flash in the tune with the new VE and MAF tables to do another logging session, should the LFTF's be closer to zero without changing any of the tune mods I listed in post #10?

[joecar]

Since your MAF seems good, I would leave it alone and only edit VE.

[joecar]

...
should the LFTF's be closer to zero without changing any of the tune mods I listed in post #10?Set B0120 back to 4000.

[ferocity02]

Set B0120 back to 4000.

Thanks joe. I put B0120 back to 4000 and re-enabled cat protection and DFCO, flashed in the tune with the new VE and MAF tables, and drove to work. The LTFT's are more reasonable now, in the 0-3.6% range. Bank 2 was much closer to 0%, and bank 1 is always a few percent higher than bank 2. Overall the truck seems to behave a little better now. Still misfiring at idle, but perhaps it's false since it's cammed, albeit a small cam. Exhaust still smells of fuel at idle too :(

Coincidentally, I got a P1415 code this morning (AIR injection system bank 1). I got this code about a year ago and it was a stuck check valve, but I freed up the valve and it's been working fine since. I doubt this has anything to do with my issues but I thought it was strange that it came up now.

Would doing RAFIG help at all? I did it once, but it might be time to do it again with the improved VE table.

[joecar]

Try reducing VE in idle region and observe for misfire and ehaust fuel smell.

AIR system is only active immediately after startup (to help initiate cat lightoff) and then it stops (see B0201).

RAFIG will help, yes do it again.