FIC Injector Data

[ferocity02]

Can someone look at this tune file and see if the injector data is entered correctly (disregard the rest of the tune, its from a stock Silverado, just want to see if I'm doing the injector data correctly). My fuel trims are going max negative on decel no matter how much I reduce the MAF table. My tuner thinks I'm hitting the minimum pulse width of the injectors/data. The IBPW PID shows ~1.77-1.81ms during decel, and ~2.15 at idle. My tuner said he's never had to adjust the minimum pulse width, small pulse adjust, etc of the injectors.

It's an '01 Silverado with the stock manifold referenced FPR (around 58psi with reference line disconnected) and FIC Bosch 60# injectors at 3 bar. The data they sent they scaled for me for 4 bar.

Thanks!

[O1Z06]

I think your hitting the PCM limit B9021 which you can't adjust without the .cax.

[ferocity02]

I think your hitting the PCM limit B9021 which you can't adjust without the .cax.

I have the cax and B9021 is set to 0.896ms, which hasn't been changed from the stock value.

[O1Z06]

I used the min pulse in B9021 and fixed the rich decel.

[ferocity02]

I used the min pulse in B9021 and fixed the rich decel.

Could you tell me a little more about what you did? What is yours set to?

[O1Z06]

I set it to the min pulse width from B4003. Have a look here http://forum.efilive.com/showthread.php?13364-IBPW-B4005-B3701-correlation/page2&highlight=b9021

[ferocity02]

I set it to the min pulse width from B4003. Have a look here http://forum.efilive.com/showthread.php?13364-IBPW-B4005-B3701-correlation/page2&highlight=b9021

Thanks! Read through that earlier today and talked to injector dynamics, and decided to set B9021 to 0.12ms, same as B4003 and B4003, and it helped a little bit. My pulsewidths on decel now get down to 1.55ms, but the fuel trims are still going max negative. The logged pulsewidths are still flattening out on decel so I think I'm still hitting a lower limit. Attached log is from this afternoon with this change.

[O1Z06]

You're running a MAF? Can you try OLSD COS3/5?

[ferocity02]

You're running a MAF? Can you try OLSD COS3/5?

Yes, still using the MAF within it's limits. I think the plan is my tuner will finish up my 1 bar MAF tune and then I'll go to 2 bar COS on my own. I hear going to COS is simple.

[BLK02WS6]

So, are you cutting down the VE table in the decel area and it won't lean out?

[ferocity02]

So, are you cutting down the VE table in the decel area and it won't lean out?

While I'm datalogging I have B0120 set to 400 RPM, so it should only be running off the MAF sensor, correct?

I have not been making changes to the VE table (not yet at least) because I was focusing on why my fuel trims were going max negative. Once I got that sorted I was going to get the VE table dialed in and then set B0120 back to 4000 RPM.

Is my thinking here off?

[BLK02WS6]

Yes, if you have B0120 set to 400 rpm, you are on the MAF table. I generally do it the other way around and tune the VE first and then the MAF - but, that's just me... Are you trying to lean it out by reducing the MAF table in the decel area? It would seem that the Hz that it sees during decel are the same ones that it will see at idle - so if you lean out the decel, you will lean out the idle. Is that what you are running up against? This is why I tune the VE first - if you look at the VE cells you are in during decel, they are different than the idle cells. After getting the entire VE right, then I put the MAF back on line and tune the MAF table (I keep the VE on line when tuning the MAF - if you got the VE right in SD, any correction needed now is from the MAF). To me, things just seem to fall into place a lot better doing it that way - but, like I said, that is just my way of doing things.

[ferocity02]

Yes, if you have B0120 set to 400 rpm, you are on the MAF table. I generally do it the other way around and tune the VE first and then the MAF - but, that's just me... Are you trying to lean it out by reducing the MAF table in the decel area? It would seem that the Hz that it sees during decel are the same ones that it will see at idle - so if you lean out the decel, you will lean out the idle. Is that what you are running up against? This is why I tune the VE first - if you look at the VE cells you are in during decel, they are different than the idle cells. After getting the entire VE right, then I put the MAF back on line and tune the MAF table (I keep the VE on line when tuning the MAF - if you got the VE right in SD, any correction needed now is from the MAF). To me, things just seem to fall into place a lot better doing it that way - but, like I said, that is just my way of doing things.

The decel MAFFREQ is around 3250, and idle is more like 2800, so they are not the same. The issue was that no matter how much I reduce the MAF table in decel it always goes rich. I think it was because my injectors were staying open too long hitting a minimum pulsewidth. I kept reducing the MAF table until there was a dip at 3250 and I said no more, something else had to be the issue.

I went back to the FIC settings, made B9021 equal to B4003 and B4004 (0.125ms) and then dropped B3701 by 10%, now the fuel trims are much better and aren't going max negative with the stock MAF table.

I'm still on the edge of going to 2 bar COS and ditching the MAF. Waiting to hear back from my tuner.

[joecar]

Sanity check: check the tables to make sure idle is not enabling (if for some reason idle is enabling while traveling, then it will alter the airmass and fueling).

[ferocity02]

Sanity check: check the tables to make sure idle is not enabling (if for some reason idle is enabling while traveling, then it will alter the airmass and fueling).

Max speed and throttle for idle B0107 and B0108 are both zero.

[5.7ute]

Log GM.VOLTS, GM.AFR, GM.INJFLOW, DYNCYLAIR_DMA, CYLAIR_DMA, IBPW1 or 2, as well as speed, TP etc & I will crunch some numbers when I get a chance.
Cheers Mick

[joecar]

Hey Mick, thanks.

[ferocity02]

Mick, I haven't gotten a chance to log that data for you, been busy getting COS5 2bar SD working. I was hoping it would help but I'm having a similar issue I think. Instead now the low MAP columns of my VE table are getting the shaft.

I made B9012, B4003, and B4004 all zeros. It helped a little, but it's still going rich around 1200 RPM on decel. Next thing would be to reduce the small pulse adjust.

I can still log that data for you. Do you want me to put the above calibrations back to what FIC recommended first?

[5.7ute]

Mick, I haven't gotten a chance to log that data for you, been busy getting COS5 2bar SD working. I was hoping it would help but I'm having a similar issue I think. Instead now the low MAP columns of my VE table are getting the shaft.

I made B9012, B4003, and B4004 all zeros. It helped a little, but it's still going rich around 1200 RPM on decel. Next thing would be to reduce the small pulse adjust.

I can still log that data for you. Do you want me to put the above calibrations back to what FIC recommended first?

Yes. Use the stock calibrations from FIC & set minimum transient pulsewidth (B9021) to the same 0.125ms. Make sure with the CYLAIR & DYNCYLAIR pids that you log the ones with the _DMA added. These are Direct Memory Access pids & let us know exactly what the pcm is using for airmass.

[ferocity02]

Got that data for you Mick. But, for some reason the CYLAIR_DMA and DYNCYLAIR_DMA would not validate on my COS5 OS 2bar OLSD, but I attempted to log them anyways and it looks like I got data for them... Not sure if its correct data or what...

I put all of the injector data back to FIC's and made B9021 0.125ms to match B4003 and B4004. It was running rich everywhere because I was tuning it with B3701 reduced 10% from FIC's data.

Let me know if you can draw any conclusions from this data. Thanks!

[ferocity02]

So I dropped FIC's B3701 offset table by 20%, retuned the VE table, and I believe the rich decel issue has been cured.

But now the question is, when does one stop modifying the injector data?

For example, I could drop B3701 which will lean it out across the board, and then raise the VE table to richen it back up, and vice versa. So at what point do you hold the injector data constant and use the VE table to do the adjusting?

If the injector data is way off, it could still be tuned with the VE table within reason, but the calculated airflow would not be representative of the real airflow.

Am I looking into this too much?

[5.7ute]

So I dropped FIC's B3701 offset table by 20%, retuned the VE table, and I believe the rich decel issue has been cured.

But now the question is, when does one stop modifying the injector data?

For example, I could drop B3701 which will lean it out across the board, and then raise the VE table to richen it back up, and vice versa. So at what point do you hold the injector data constant and use the VE table to do the adjusting?

If the injector data is way off, it could still be tuned with the VE table within reason, but the calculated airflow would not be representative of the real airflow.

Am I looking into this too much?

Injector data "should" be left alone as it is based on the mechanical properties of the injectors being used. However this depends on the reliability of the data you have been given. I am not saying that the data you have is erronous, but there have been some doubts as to where that data you have has been generated.
The only definitive way you can test the data without the neccesary equipment is to tune the vehicle with stock injectors & data. Swap them back to the new injectors & data & look for changes in the resulting AFR's.
The log you sent should answer some questions once I get an hour or so to go through it.

[ferocity02]

So when I swapped the injectors and didn't change anything else except for the injector data, the fuel trims went from +/-1% to -6%-max negative in decel. I tried adjusting the MAF table but it was always rich in decel. Reducing B3701 was the only cure I found.

Then I switched to COS5 2bar OLSD hoping it would help but had the same issue just with the VE table instead of the MAF table.

I am aware of the doubt about FIC's injector data, hence not trusting their data completely. My tuner is certain the injector data is incorrect as well.

Thank you for the help!

[5.7ute]

That is certainly pointing to bad data.
I posted a while ago on here a way I thought B3701 could be tuned in from a good known tune with accurate data. I will try find it & post a link.

[5.7ute]

Take a look here.
http://forum.efilive.com/showthread.php?10179-Injector-Base-Pulsewidth/page5&highlight=b3701
This is just the basics of it. There will be calculated pids to write to get it to work, as well as a lot of filtering to remove small pulse adder errors etc.
Also I should note that a offset issue should occur throughout the tune, not just in decel. If only decel is the culprit small pulse adder could be the problem.