LS1A Tuning

[RADustin]

I was working on my Dad's ls1A this weekend. It's a ls1A PCM on the 385 horse Z06 engine with a cam and drive by cable. T56 car. All swapped in a 1961 Impala. Geoff at EPS provided the initial base line years ago and I was finally going in to fine tune things.

I set the MAF to fail so that I could calc.ve it. I ensured it stayed open loop and my WB02 was in the left bank with the NB02 normally goes. NB02 for left bank was unplugged, and right bank was left in tacted, plugged in.

This lead to a couple challenges-


The car seemed to always want to idle off the backup VE table. Maybe it runs all the time off the backup, but for sure the idle. Is this normal? Is this because I failed the MAF? Maybe because we unplugged the 02? It makes for a very spiky VE for idle cells as the resolution of the backup is so terrible. Can anyone provide some insight? Is there a PID for backup mode enabled? I always ensured to copy the main VE with labels and paste with labels into the backup table during changes. If I tried to smooth the VE at all after a modification from the log file, the AFR would be way off from desired. This seems to be from poor map resolution and interpolation and such.
The VE lookup PID does not work for LS1A. What is the proper way to tune the MAF? I tried to improvise and got sane values so I used them. I logged the cylair_dma pid. I made a custom calc to do {calc.w02ben}*{cylair_dma}*{sae.rpm}/15. I charted this against MAF hz. It gave me numbers about 1 gram/s lower on the mid to lower end of the scale and about 30g/s lower on the high end from stock. This is to be expected as the engine has a tiny airfilter mounted right on the MAF, which is mounted right on the throttle body. Anyways, are there any other good ways to perform this task of re-scaling the MAF?


I appreciate any help. Swapping PCM is out of the question, I need to make what he has work- and it does right now for the most part. I just want to understand more about what's going on and ensure everything is 100%.


Thanks!

[joecar]

I set the MAF to fail so that I could calc.ve it. I ensured it stayed open loop and my WB02 was in the left bank with the NB02 normally goes. NB02 for left bank was unplugged, and right bank was left in tacted, plugged in.

This lead to a couple challenges-


The car seemed to always want to idle off the backup VE table. Maybe it runs all the time off the backup, but for sure the idle. Is this normal? Is this because I failed the MAF? Maybe because we unplugged the 02? It makes for a very spiky VE for idle cells as the resolution of the backup is so terrible. Can anyone provide some insight? Is there a PID for backup mode enabled? I always ensured to copy the main VE with labels and paste with labels into the backup table during changes. If I tried to smooth the VE at all after a modification from the log file, the AFR would be way off from desired. This seems to be from poor map resolution and interpolation and such.


With MAF failed, if the PCM has a Backup VE table (which LS1A does), it will run the Backup VE table (and not the Main VE table)... and yes, the Backup VE table has poor resolution.

[RADustin]

Crap OK.

So can I leave MAF enabled and working and still tune VE?

EDIT: If I disable STFT and LTFT? would that work and allow me to tune VE without going to the backup table?

[joecar]

2. The VE lookup PID does not work for LS1A. What is the proper way to tune the MAF? I tried to improvise and got sane values so I used them. I logged the cylair_dma pid. I made a custom calc to do {calc.w02ben}*{cylair_dma}*{sae.rpm}/15. I charted this against MAF hz. It gave me numbers about 1 gram/s lower on the mid to lower end of the scale and about 30g/s lower on the high end from stock. This is to be expected as the engine has a tiny airfilter mounted right on the MAF, which is mounted right on the throttle body. Anyways, are there any other good ways to perform this task of re-scaling the MAF?

Yes, unfortunately the pid VETABLE_DMA does not work for LS1A.

you're using {cylair_dma} to calculate MAF... but take note of the following:

{GM.CYLAIR_DMA}
PCM calculates this [g/cyl] internally from MAF

{GM.DYNCYLAIR}
{GM.DYNCYLAIR_DMA }
PCM calculates both of these [g/cyl] internally from VE

( see post #2 here: Summary-Notes (https://forum.efilive.com/showthread.php?14188-Summary-Notes) )

so it seems to me that you want either GM.DYNCYLAIR or GM.DYNCYLAIR_DMA (whichever one works on LS1A) since these are calculated from VE.

[joecar]

Your airfilter being so close to MAF may be causing additional problems.

[RADustin]

Yes, unfortunately the pid VETABLE_DMA does not work for LS1A.

you're using {cylair_dma} to calculate MAF... but take note of the following:

{GM.CYLAIR_DMA}
PCM calculates this [g/cyl] internally from MAF

{GM.DYNCYLAIR}
{GM.DYNCYLAIR_DMA }
PCM calculates both of these [g/cyl] internally from VE

( see post #2 here: Summary-Notes (https://forum.efilive.com/showthread.php?14188-Summary-Notes) )

so it seems to me that you want either GM.DYNCYLAIR or GM.DYNCYLAIR_DMA (whichever one works on LS1A) since these are calculated from VE.

I was originally looking for GM.DYNCYLAIR or GM.DYNCYLAIR_DMA but didn't notice either one as available. I'll need to retune though it appears so I can look again.

Thanks!

[RADustin]

Your airfilter being so close to MAF may be causing additional problems.

Its a street and performance air cleaner. I'm sure it's junk for performance but it's what he has so it stays. I'll have to make it work one way or the other.

Like this one-

http://www.fquick.com/images/vehicles/full/632949105.jpg

[joecar]

Several problems:
1. air is still turning a sharp corner when it hits the sensor, so it is colliding with itself and gets tubulent, i.e. air is not going straight and uniform, see pic below.
2. when fans turn on, the airflow will be disrupted by fan-wash.
3. pulls in hot air (not just from engine, but from fan-wash).
4. filter element is small, engine will out-pull it.

1. and 2. will directly affect how the MAF sensor reads the airflow going past it.

3. and 4. are performance related, and may cause knock to occur.


If there's room, find some flexible clothes-dryer style duct to pull air from elsewhere, and use conical filter (has much larger surface area).

I know, easier said than done, I understand
you -> :chair:<- me


21538

[RADustin]

I completely understand and agree.

Unfortunately his priority is shiney showcar >>> anything else. So it is what it is.

I'm just going to tune it with the numbers I can gather and call it good enough.

Do you have any ideas to tune the full ve chart?

[RADustin]

correct me if I'm wrong but I need to do more than just open loop but also speed density to correctly tune the ve table. Right?

so by turning off LTFT and STFT, I'm still not there. I need to make it listen to the ve only, without failing the MAF.

[joecar]

Using LTFT gives you only the low load parts of the VE table.

In your situation, it might be best to use wideband-only with CL/LTFT/STFT disabled, and MAF disabled also (i.e SD).


You have to fail the MAF (force a MAF DTC to show up) otherwise the PCM will still use the MAF...

the PCM uses the MAF as follows:
- above B0120: MAF only,
- below B0120: MAF for steady state (airmass changes less than some threshold), VE for transient air (i.e. where airmass is changing faster than the threshold),
where B0120 is typically 4000 rpm.

As you can see, with MAF enabled, the MAF will be used as much as possible.

[joecar]

Using LTFT gives you only the low load parts of the VE table.

In your situation, it might be best to use wideband-only with CL/LTFT/STFT disabled, and MAF disabled also (i.e SD).


You have to fail the MAF (force a MAF DTC to show up) otherwise the PCM will still use the MAF...

the PCM uses the MAF as follows:
- above B0120: MAF only,
- below B0120: MAF for steady state (airmass changes less than some threshold), VE for transient air (i.e. where airmass is changing faster than the threshold),
where B0120 is typically 4000 rpm.

As you can see, with MAF enabled, the MAF will be used as much as possible.

[RADustin]

anyway to fail MAF and not be on the backup ve table?

can I change the DTC to do not report? I guess that wouldn't throw it in SD though..

[joecar]

No, unfortunately not...

when Backup VE exists, PCM will use that when MAF is failed.

[RADustin]

I understand. Thanks.