Injector Offset ?'s vs LS1 edit

[Tordne]

I created a custom PID some time ago that turns the LTFT values into a factor similar to BEN that I could use as a multiplier.

Could that work? Using a sort of multiplier against the current Voltage Offset table as sort of a error %?

[jfpilla]

Next file.

[jfpilla]

another one.

[jfpilla]

I created a custom PID some time ago that turns the LTFT values into a factor similar to BEN that I could use as a multiplier.

Could that work? Using a sort of multiplier against the current Voltage Offset table as sort of a error %?

I'd try it. The biggest problem is logging volts. There's no way to control voltage and log them sufficiently. It looks like lots of extrapolating.

[hpcubed]

I created a custom PID some time ago that turns the LTFT values into a factor similar to BEN that I could use as a multiplier.

Could that work? Using a sort of multiplier against the current Voltage Offset table as sort of a error %?

You will be combining all of the data for each map range and condensing it into one cell so there could be some error as your BEN is probably not 1 across the board. But it will give you an idea. One thing I looked at was the difference between the BEN factors between 13.5V and 14V for the same map values. Not how far the BEN was from 1.00. Only the difference. I think this is useful and got my BENs a lot closer.

Initially i used the maf and scaled the offset table so I had similar BEN values that i had with my stock injectors. I had to increase mostly around idle and not much at WOT.

[TAQuickness]

You will be combining all of the data for each map range and condensing it into one cell so there could be some error as your BEN is probably not 1 across the board. But it will give you an idea. One thing I looked at was the difference between the BEN factors between 13.5V and 14V for the same map values. Not how far the BEN was from 1.00. Only the difference. I think this is useful and got my BENs a lot closer.

Initially i used the maf and scaled the offset table so I had similar BEN values that i had with my stock injectors. I had to increase mostly around idle and not much at WOT.

Now that's a good idea...

[jfpilla]

Now that's a good idea...

It looks like you got your wish. The attached was done with the MAF so it's not accurate. I was interested in seeing the values it would produce. It should be done in SD. Any thoughts on what comes first?

-VE's back to stock.
-IOS adjusted.
-VE's

[hpcubed]

It looks like you got your wish. The attached was done with the MAF so it's not accurate. I was interested in seeing the values it would produce. It should be done in SD. Any thoughts on what comes first?

-VE's back to stock.
-IOS adjusted.
-VE's

Those look pretty good already just need a little tweeking. I would adjust the IO's with the maf in open loop, then go to speed density and adjust the VE's. Don't really need to put the VE's back to stock.

One question is how accurate is the stock offset table for the stock injectors. Assuming that it is very accurate, I think the way to do it would be:

1. Log BENs with maf in open loop using stock inj. If you have old files saved with this data you could use that.
2. Log BENs with maf in open loop using new inj.
3. Adjust inj. offset so that new BENS match old BENS. Obviously everything is not going to match perfectly. This effectively divides out the error in the maf.
4. Go to speed density and start the tuning process.

[jfpilla]

Those look pretty good already just need a little tweeking. I would adjust the IO's with the maf in open loop, then go to speed density and adjust the VE's. Don't really need to put the VE's back to stock.

One question is how accurate is the stock offset table for the stock injectors. Assuming that it is very accurate, I think the way to do it would be:

1. Log BENs with maf in open loop using stock inj. If you have old files saved with this data you could use that.
2. Log BENs with maf in open loop using new inj.
3. Adjust inj. offset so that new BENS match old BENS. Obviously everything is not going to match perfectly. This effectively divides out the error in the maf.
4. Go to speed density and start the tuning process.

1. Log BENs with maf in open loop using stock inj. If you have old files saved with this data you could use that.
>>>>That's out.

2. Log BENs with maf in open loop using new inj.
3. Adjust inj. offset so that new BENS match old BENS. Obviously everything is not going to match perfectly. This effectively divides out the error in the maf.
4. Go to speed density and start the tuning process.
>>>>I don't get the point of logging in open loop then tuning in SD?

I've established an offset table using ltft's vs volts and manvac and the stock VE table. You mentioned using Ben's so I tried it to see what kind of values would be returned. The results seem accurate but whenever the AFR changes enough it's obvious that trims affect these results making the results look decent since the AFR is being maintained by trims. It looks as if there are two ways to do this. 1) Set up the offset table by ltft's. This is very tedious. 2) Go into mafless mode and use BEN's.
What I don't know is if the VE table should be stock when doing offsets. VE's should be done with correct IFR's. If the IOS's are wrong then the IFR's are wrong. It sounds like I'm going around in circles and that's how I feel. Any opinions on my thinking are encouraged since I'm a little lost as to what to go at first. I guess I can do the offsets using BEN with a stock VE table in SD and see what I get. It would seem that if the offsets come into line then the VE table should also be close. I'd better save my current file since the car is running great with it.

[hpcubed]

1. What you suggest should work. But I would rather use a wideband than trims so that is why I suggest using the BENs (and open loop).

2. My suggested steps utilized the stock inj and maf because we need a known to compare to. And by doing it the way I suggest, the error from the maf is divided out. Since you can't do this in your situation. So you have to either accept that what your maf is saying as a known or the VE table. I would think the maf is more correct right out of the box - so to speak.

3. You set up your table as you suggested either using BENS or trim. However the value in your table for 1 cell is condensed down from the values of trims from all rpms for a given map. If they are all the same value then bingo if they are all over the place you have to use some judgement.