Troubles with scaled tune for 95# injectors

[GNXClone]

2006 GTO with twin turbo (E40 ECM)
LS2 block, LSA heads, LS9 cam, 9.5:1 compression
FIC 95# injectors
Returnless fuel system @constant 58 PSI
2.5 BAR OS
MAF DISABLED
Wideband installed
Injector parameters received from FIC (IFR, Min PW, short pulse adder, offset, etc)

I made a 50% scaled tune for this setup and it only barely runs. When I crank it, it's very rich and I have to push the pedal to the floor to get it to start (flood mode). I can keep it idling by feathering the accelerator and eventually LTFT kicks in at -20% and it will idle. As soon as I try to give it any throttle, it goes very lean and sputters...sometimes...other times it goes very rich, and still sputters. At these times, LTFT is zero as it has never had the opportunity to learn. It seems like the more I attempt to fix it, the worse it gets.

I finally gave up, took my original 60# injector tune and adjusted the Stoich AFR only. The injector parameters in this tune are from the SD 60's, not the FIC 95's. Yet, this tune starts up immediately, idles smooth and runs without sputtering in all throttle positions. I know this isn't the preferred method of tuning for large injectors, but it runs good on the first attempt.

Does anyone have a 50% scaled E40 tune for FIC 95's they would care to post to the repository that I could use as a template?

[mowton]

Post your 60 and 95 tunes and any logs you may have for help. Remember, the 60 tune (IFR and all other airflow related parameters) must be scaled to start like 25% or if IFR is straight-lined at 63.5 then its even worse. May have to go back to a stock file and scale by 50%.

Ed M

[DSteck]

Injector parameters received from FIC (IFR, Min PW, short pulse adder, offset, etc)

I can tell you right now they are wrong. That's a pretty significant part of your problem.

[GNXClone]

Post your 60 and 95 tunes and any logs you may have for help. Remember, the 60 tune (IFR and all other airflow related parameters) must be scaled to start like 25% or if IFR is straight-lined at 63.5 then its even worse. May have to go back to a stock file and scale by 50%.

Ed M

FIC1000 is the scaled tune which barely runs. The other is the SD60 tune which has the const maxed out at 63.5 all the way and absorbs the rest into the VE. However, the SD60 tune ran perfect with the 60's, and runs pretty decent with the just AFR scaled with the FIC95's.

[mowton]

FIC1000 is the scaled tune which barely runs. The other is the SD60 tune which has the const maxed out at 63.5 all the way and absorbs the rest into the VE. However, the SD60 tune ran perfect with the 60's, and runs pretty decent with the just AFR scaled with the FIC95's.

You mean IFR...right.

Ed M

[edcmat-l1]

I would bet your math for the % to scale isn't correct. It's all based on what it ran before as, and what the IFR was before, relative to actual correct IFR of the injectors used.

If you were using 60s, you couldn't actually get the proper IFR in for them, as the table is capped at 63.5. So, you were starting with a somewhat scaled VE. To think you can just plug in the IFR as 50% of 1000cc (or whatever you're using) scale the VE by 50% and you're good, will not work.

If you're trying to do it that way, you'd be better off installing the original injectors, mapping the non boost VE to them, then scaling by 50% and plugging in the IFR.

I can tell you right now they are wrong. That's a pretty significant part of your problem.

I agree with this here too.

[mowton]

It is 50% (actually 51%) and he scaled the MAF from a stock tune for his 50% file but the VE doesn't look scaled properly. Agree with Ed, as I said above, the straight line 63.5 is a scaled tune to start...just not linear which is what makes it a bad place to start.

Also the 60's tune has open loop set to stock and 1000 tune is set all to 1.0 for tuning. Not sure where you are in temp when you say it won't run good....too many variables I would think. As the OP states, just by changing the IFR in the 60's tune, it runs. That leads me to believe it is the balance of the scaled tune which isn't correct.

I would start over with the stock file and tune it accordingly. Go with MAF only if it is easier to start.

Also I have seen many statements on the FIC data. I have used several and not seen any huge issues? They have been all E38 types....don't know if they are more forgiving.

Have a great Holiday weekend all.

Ed M

[GNXClone]

On the 50% scaled tune:

I started with my 60# tune, but first I set the MAF and VE tables to the stock 06 GTO values (non-boost VE obviously). So the scaling was from the stock values, not my tuned 60# values. There are so many changes in this 50% scaled tune, I'm tempted to start over.


On the modified 60# tune:

I left the IFR at 63.5 across the board. Setting only the Stoich AFR to 17.64 has the car running with very little LTFT, on the order of 5% or less. What's interesting is the VE for this tune was setup when it was a standard LS2. Now it has a ZR1 cam and LSA heads, but the VE seem to be pretty close still.

Now I have a third tune:

I computed the percentage difference from 17.64 to 14.7 and applied that to both VE tables in the second tune mentioned above. I then set the AFR back to 14.7. This also runs well. I do realize that I have absorbed all of the injector difference into the VE and that this is not preferred, but at least it's drivable and so far, no stumbling, stalling or popping on decel.

If I don't encounter any odd driveability issues, I may just stick with the "raped" tune.

[GNXClone]

You mean IFR...right.

Ed M

I meant AFR. Stoich AFR. I took that tune, which already had the IFR table maxed out, and altered just the Stoich AFR to make up the injector flow difference. I left the IFR maxed out at 63.5.

The SD60 tune itself was not the proper way to tune given they really flow on the order of 73# @ 4 BAR. But this non-scaled SD60 tune ran perfect for me for over two years.

[mowton]

On the 50% scaled tune:

I started with my 60# tune, but first I set the MAF and VE tables to the stock 06 GTO values (non-boost VE obviously). So the scaling was from the stock values, not my tuned 60# values. There are so many changes in this 50% scaled tune, I'm tempted to start over.


On the modified 60# tune:

I left the IFR at 63.5 across the board. Setting only the Stoich AFR to 17.64 has the car running with very little LTFT, on the order of 5% or less. What's interesting is the VE for this tune was setup when it was a standard LS2. Now it has a ZR1 cam and LSA heads, but the VE seem to be pretty close still.

Now I have a third tune:

I computed the percentage difference from 17.64 to 14.7 and applied that to both VE tables in the second tune mentioned above. I then set the AFR back to 14.7. This also runs well. I do realize that I have absorbed all of the injector difference into the VE and that this is not preferred, but at least it's drivable and so far, no stumbling, stalling or popping on decel.

If I don't encounter any odd driveability issues, I may just stick with the "raped" tune.

Not be be argumentative, but you can't adjust a tune with AFR...I thought you had changed the IFR. If you want to get this right I still think you need to set the IFR (and assoiciated injector data right) according to The FIC data and start at like 2K rpm and tune the car according to your WB delta error whether it is MAF or VE. Then move down to the idle regions. There are way too many differences from your 60 tune and the scaled 50% tune to even copmpare in my book....sorry

To see the differences, setup an excel file showing the scale factor differences from the 63.5 straight line approach for the 60 lbr's values (58 psi fuel pressure) and percentage change from the linear scaled factor of the FIC 1000. You will see there are huge disparages over the MAP range.

I have just finished a 2007 TT 427 with 850 cc injectors (based on FIC parameters) and the car drove beautiful and made 950 rwhp on a conservative (AFR-11.5 timiung -12.5 degrees at wot) tune.

Ed M

[GNXClone]

No argument here, I agree completely. I actually have already tried what you suggested.

I scaled the entire non-boost VE by the difference seen when I set the AFR in the other tune, which turned out to be .8333%. This normalized the VE to match what I needed to run the 95's with 63.5 in all IFR cells. Then I entered the FIC IFR differential numbers and calculated the diference between those and the previous 63.5. I then took those for each column and applied them to the VE table columns. Where VE 105 is IFR 0 kPa, so on so forth.

What has me confused is that it runs so poorly. Not even well enough to get any valid WB data.

I need to start from scratch with scaled tune, and I suppose I should re-enable the MAF and begin with that since MAF tuning is so much simpler. Then move on to the VE tables.

You're right, I changed almost everything in the scaled tune. I should've known better and should've used the "change one thing at a time" approach.


Several have mentioned the injectors parameters I got from FIC are a problem? Is there a better source for these?

[DSteck]

Yea, buy injectors with real data aka Injector Dynamics. That spreadsheet FIC gives out with their redrilled bullshit is a ripoff of ID data, meaning it is data for a totally different injector.

[edcmat-l1]

Yea, buy injectors with real data aka Injector Dynamics. That spreadsheet FIC gives out with their redrilled bullshit is a ripoff of ID data, meaning it is data for a totally different injector.

While I agree the ID injector data is great to have, it's not an issue here. Besides, some of us were tuning the big injectors way before the ID data was even available. Not everyone is going to drop the coin on the IDs. To say it can't be tuned without swapping out the injectors for IDs is a cop out.

[DSteck]

I never said can't. However, nobody else dynamically matches their injectors. I've seen what FIC gives out with their injectors, and it is just LS9 data or ID's spreadsheet with anything mentioning ID removed.

The issue here is lack of proper scaling. The injector scenario just adds another very important element to a puzzle that's already hard for most to solve. The best course of action IMO for the OP is to just start with a stock file and small injectors, tune a little, then scale it while keeping the small injectors in. Then swap injectors and IFRs and work on the offsets/short pulse adders.

Just for the record, the stoich AFR trick works great and makes it very simply for most users. The problem is E40s are still limited to 67.75 lb/min even with the custom OS. E38s have a much easier time. What this means is E40s have to be scaled the long way.

[edcmat-l1]

I never said can't. However, nobody else dynamically matches their injectors. I've seen what FIC gives out with their injectors, and it is just LS9 data or ID's spreadsheet with anything mentioning ID removed.

The issue here is lack of proper scaling. The injector scenario just adds another very important element to a puzzle that's already hard for most to solve. The best course of action IMO for the OP is to just start with a stock file and small injectors, tune a little, then scale it while keeping the small injectors in. Then swap injectors and IFRs and work on the offsets/short pulse adders.

Just for the record, the stoich AFR trick works great and makes it very simply for most users. The problem is E40s are still limited to 67.75 lb/hr even with the custom OS. E38s have a much easier time. What this means is E40s have to be scaled the long way.

I'm well aware of the qualities of the ID data, and also aware of the LACK OF quality data with the FICs. I would much prefer to use the IDs myself. BUT, in this case, the IDs would NOT make the issue o scaling any easier for the OP. If his math is off, it's not going to want to run regardless.

And just for the record, the IFR cap in the E40s is 63.5 lb. per hr.

[mowton]

I never said can't. However, nobody else dynamically matches their injectors. I've seen what FIC gives out with their injectors, and it is just LS9 data or ID's spreadsheet with anything mentioning ID removed.

The issue here is lack of proper scaling. The injector scenario just adds another very important element to a puzzle that's already hard for most to solve. The best course of action IMO for the OP is to just start with a stock file and small injectors, tune a little, then scale it while keeping the small injectors in. Then swap injectors and IFRs and work on the offsets/short pulse adders.

Just for the record, the stoich AFR trick works great and makes it very simply for most users. The problem is E40s are still limited to 67.75 lb/hr even with the custom OS. E38s have a much easier time. What this means is E40s have to be scaled the long way.

I could see AFR trick working if both tunes IFR's were in a linear format instead of the source tune being flat lined at 63.5. That flat lined tune has a ever changing scale factor for each MAP KpA value. That is why I also recommend going back to stock tune and starting over using the data he has, scaling the IFR by 50% and ballparking the MAF, tune with MAF and even use a histogram to calculate the VE for a starting table. Of course, all the other airflow parameters need to be scaled including the timing as well.

Tuning with the smaller injectors (within the 63.5 limit) is also a good idea and would show how good or bad the FIC data was..... would be interesting. That is the way Greg shows how scaling works on his 2nd video.

As I stated above, I have used the FIC data on 3 occasions in mild/highly boosted applications and have been pretty good success.

Have a great weekend

Ed M

[edcmat-l1]

I could see AFR trick working if both tunes IFR's were in a linear format instead of the source tune being flat lined at 63.5. That flat lined tune has a ever changing scale factor for each MAP KpA value. That is why I also recommend going back to stock tune and starting over using the data he has, scaling the IFR by 50% and ballparking the MAF, tune with MAF and even use a histogram to calculate the VE for a starting table. Of course, all the other airflow parameters need to be scaled including the timing as well.

Ed M

That's almost exactly what I recommended to him over on the GTO board. Go back to the original injectors (AND original IFR), dial in the non boost VE, then scale both by 50%, then throw in the FICs (With their IFR scaled 50%) and go from there.

[DSteck]

And just for the record, the IFR cap in the E40s is 63.5 lb. per hr.

Yes, but the airflow cap is 67.75 lb/min. My units were a typo in my other post.

[GNXClone]

Lots of great advice here, thank you. I won't be buying ID injectors since I already made the FIC purchase, so I will just have to deal with that. I drove the car about 60 miles last night and it idles and drives good on the hacked tune, so I think the injectors will work out. Once the LTFT correction kicks in at idle (550 RPM), the AFR reading on the WB is 14.7, so I know they are capable of a clean idle, which was a concern of mine being as large as they are.

Interesting idea about using smaller injectors to tune, then switching and scaling. Coincidentally, I will have the opportunity to try this soon. I bought some stock ZR1 injectors (52# or 56#) and can use those as my 'small' injectors. These should support crank 650HP which means I'll be able to tune most of my boost VE table as well. I'll also re-install and re-enable the MAF. I do recall that tuning with the MAF was much easier when I setup the initial tune three years ago.

[DSteck]

Lots of great advice here, thank you. I won't be buying ID injectors since I already made the FIC purchase, so I will just have to deal with that. I drove the car about 60 miles last night and it idles and drives good on the hacked tune, so I think the injectors will work out. Once the LTFT correction kicks in at idle (550 RPM), the AFR reading on the WB is 14.7, so I know they are capable of a clean idle, which was a concern of mine being as large as they are.

Interesting idea about using smaller injectors to tune, then switching and scaling. Coincidentally, I will have the opportunity to try this soon. I bought some stock ZR1 injectors (52# or 56#) and can use those as my 'small' injectors. These should support crank 650HP which means I'll be able to tune most of my boost VE table as well. I'll also re-install and re-enable the MAF. I do recall that tuning with the MAF was much easier when I setup the initial tune three years ago.

If the airflow model is correct, it should idle stoich without requiring fuel trims. I'd be more concerned about the fact that it requires closed loop to idle stoich.