PI : DI Ratio Discussion

[steve99]

Originally Posted by solidONE

Any one else notice the fuel trims go crazy after loading these tables? How are you guys dealing with it?

That probably means your port and direct injection is not scaled correctly.

Assuming your maf scaling is ok and its not due to wide temp variation at lower rpm due heat soak.

I am assuming its gone all one way ie plus or negitive at the lower load rpm ranges where the big shift from direct to port was done

If your kean you could adjust the "injector flow scaling BRZ" (port injection) parameter up/down slightly to correct the imbalance between port and direct that easier than messing with direct injection pressure mutipliers. althout ztan did discover what he has called Multiplier C an overall direct injection multiplier

To do it correctly you would rearly need to go 100% port and correct that then go 100% direct and correct again.

[solidONE]

Originally Posted by steve99

That probably means your port and direct injection is not scaled correctly.

Assuming your maf scaling is ok and its not due to wide temp variation at lower rpm due heat soak.

I am assuming its gone all one way ie plus or negitive at the lower load rpm ranges where the big shift from direct to port was done

If your kean you could adjust the "injector flow scaling BRZ" (port injection) parameter up/down slightly to correct the imbalance between port and direct that easier than messing with direct injection pressure mutipliers. althout ztan did discover what he has called Multiplier C an overall direct injection multiplier

To do it correctly you would rearly need to go 100% port and correct that then go 100% direct and correct again.

You're right, fuel trims pretty much have all gone positive during low load CL operation. This end up increasing the LTFT in OL causing it to run rich in OL. So I'd assume I'd need to increase the "injector flow scaling BRZ" value?

[steve99]

Originally Posted by solidONE

You're right, fuel trims pretty much have all gone positive during low load CL operation. This end up increasing the LTFT in OL causing it to run rich in OL. So I'd assume I'd need to increase the "injector flow scaling BRZ" value?

OK if trims gone positive then its adding fuel to compensate when we switched in heaps more port , so we assume the port injectors were not adding as much fuel as the directs did.

So I think we need to make ports add more fuel.

In a petrol tune Inj scale BRZ (port) is 228
In a E85 tune its 171

So looks like lower number means more fuel.
Shiv added roughly 30% more fuel in E85 tunes so it looks like if we moved the injector scalar 2 it would equate to roughly 1% change in port fueling.

So if we want to add more fuel as the trims have gone up say +4% we need to move scalar down roughly 8. Its not going to be exact as we did not go from 0 port to 100% port but should be in the ballpark.

I would not go wildly adjusting the scalar though.

To do it correctly you would have to go 100% direct (whole map) make sure trims are good, then switch to 100% port and adjust scalar to get it as close as possible to what you got trim wise with 100% direct.

[thambu19]

Originally Posted by steve99

OK if trims gone positive then its adding fuel to compensate when we switched in heaps more port , so we assume the port injectors were not adding as much fuel as the directs did.

So I think we need to make ports add more fuel.

In a petrol tune Inj scale BRZ (port) is 228
In a E85 tune its 171

So looks like lower number means more fuel.
Shiv added roughly 30% more fuel in E85 tunes so it looks like if we moved the injector scalar 2 it would equate to roughly 1% change in port fueling.

So if we want to add more fuel as the trims have gone up say +4% we need to move scalar down roughly 8. Its not going to be exact as we did not go from 0 port to 100% port but should be in the ballpark.

I would not go wildly adjusting the scalar though.

To do it correctly you would have to go 100% direct (whole map) make sure trims are good, then switch to 100% port and adjust scalar to get it as close as possible to what you got trim wise with 100% direct.

Why do they not react appropriately when we change ratios. With more pfi the ecu should have added the right amount of pulse width I would assume

Sent from my HTC One_M8 using Tapatalk

[solidONE]

What's weird is that the LTFT seems to want to go positive, but if you subtract the amount of stft it's actually not that far off. It seems the afr will fluctuate pretty drastically under certain circumstances and somehow cause the ltft to go positive. While with ltft positive beyond a couple percent or so the stft will be just about the same amount negative canceling out the positive ltft. But once you go into OL operation the ltft just sticks at the positive trim causing it to run rich. Does that make sense? Lol

Originally Posted by steve99

To do it correctly you would have to go 100% direct (whole map) make sure trims are good, then switch to 100% port and adjust scalar to get it as close as possible to what you got trim wise with 100% direct.

I think this is what I'll have to do. Since the previous di/pi ratio table I had it set to fully Di in the first 3 columns (.10, .20, .30 load) Whereas the new table has it set to 70% PI from .20~.40 at low rpms. This is the area I'm getting the most trouble I think.

[Kodename47]

I would avoid going for the big multipliers as it's unlikely the whole lot is off.

If you can log fuel quantity, I have a way to tie in the fuel between injectors. If not then you can try running 100% DI, 100% PI or 50% and then compare the trims. I would then adjust the DI Flow Scalar as it is the most adjustable of the lot without affecting large areas unnecessarily .

[steve99]

Originally Posted by thambu19

Why do they not react appropriately when we change ratios. With more pfi the ecu should have added the right amount of pulse width I would assume

Sent from my HTC One_M8 using Tapatalk

Yes you would think it should. but others have found it not always so.

their was a heap of discussion on it here

http://www.ft86club.com/forums/showthread.php?t=53064

http://www.ft86club.com/forums/showp...5&postcount=12

http://www.ft86club.com/forums/showthread.php?t=3172


jamesm who is a tuner, no longer on this forums suggested this proceedure

http://www.ft86club.com/forums/showp...&postcount=195


ztan found some pi/di fueling error
http://www.ft86club.com/forums/showp...1&postcount=37

[ztan]

Originally Posted by thambu19

Why do they not react appropriately when we change ratios. With more pfi the ecu should have added the right amount of pulse width I would assume

Sent from my HTC One_M8 using Tapatalk

You would think, but the PI and DI fuelling algorithms are not dead on. There are differences with the scaling in both systems which mean that if you alter stock settings, the balance between PI and DI may change final fuelling quantity. There is also the issue of port fuelling running out the exhaust in high overlap areas which should not happen as much with DI fuelling. We are reliant on our lambda sensors to give us information of what is happening just after the cylinder, but don't have the ability to find out what is actually happening in the cylinder...

I set the car at 100% DI and got the MAF rescaled to flat, then set to 100% PI and set PI scalar and latency to get the PI fuelling in line - made everything run much smoother. You can work it any way you want.

[ztan]

Originally Posted by solidONE

What's weird is that the LTFT seems to want to go positive, but if you subtract the amount of stft it's actually not that far off. It seems the afr will fluctuate pretty drastically under certain circumstances and somehow cause the ltft to go positive. While with ltft positive beyond a couple percent or so the stft will be just about the same amount negative canceling out the positive ltft. But once you go into OL operation the ltft just sticks at the positive trim causing it to run rich. Does that make sense? Lol



I think this is what I'll have to do. Since the previous di/pi ratio table I had it set to fully Di in the first 3 columns (.10, .20, .30 load) Whereas the new table has it set to 70% PI from .20~.40 at low rpms. This is the area I'm getting the most trouble I think.

There is something in the fuelling algorithms that makes OL operation much richer than we would expect - I have not been able to unpick it and the logic is very complex, but there is much written on NASIOC and RomRaider on "rich dip" on throttle application or CL/OL tranistion. A lot of people have tried to get rid of it and have ended up accepting it or deciding to run full time OL. I've partially addressed it by applying more CL fuelling using the EL Comp table at the higher MAP cells - this smooths the transition a bit, but is a bit of a messy band-aid fix.

[ztan]

Originally Posted by solidONE

Any one else notice the fuel trims go crazy after loading these tables? How are you guys dealing with it?

I had done quite a bit of work to try to balance my DI and PI systems prior to loading a similar table - found my trims got slightly better clustering if anything.

[solidONE]

Originally Posted by ztan

I had done quite a bit of work to try to balance my DI and PI systems prior to loading a similar table - found my trims got slightly better clustering if anything.

Are your ratios similar in the .10 (0% pi) and .20 (70% pi) columns? Since idle loads after the engine is up to operating temps tend to float between .14~.20 g/s, I noticed this is where AFR will fluctuate quite a bit. I'm assuming the fluctuation in the di/pi % has a lot to stay about the fluctuating AFR thus the fluctuation in the fuel trims.



fluctuating fluctuations have fluctuated...

[ztan]

Originally Posted by solidONE

Are your ratios similar in the .10 (0% pi) and .20 (70% pi) columns? Since idle loads after the engine is up to operating temps tend to float between .14~.20 g/s, I noticed this is where AFR will fluctuate quite a bit. I'm assuming the fluctuation in the di/pi % has a lot to stay about the fluctuating AFR thus the fluctuation in the fuel trims.



fluctuating fluctuations have fluctuated...

I run 0% up to 0.3 load and 70% at 0.4. Full DI at low load for me stops a bit of erratic behaviour when AC cuts in and out. I also have bigger port injectors which have a larger non-linear region than stock which accentuates the split fuelling problems at low pulsewidths.

I do see quite wide variations in AFR trims at idle speeds but not at low load whilst cruising - there is a lot of intake temp variation under the hood when the car is not moving and I largely ignore these now.

[solidONE]

Originally Posted by ztan

I run 0% up to 0.3 load and 70% at 0.4. Full DI at low load for me stops a bit of erratic behaviour when AC cuts in and out. I also have bigger port injectors which have a larger non-linear region than stock which accentuates the split fuelling problems at low pulsewidths.

I do see quite wide variations in AFR trims at idle speeds but not at low load whilst cruising - there is a lot of intake temp variation under the hood when the car is not moving and I largely ignore these now.

Same here, but the change in AFR and fuel trims due to IAT change is predictable and not quite as drastic for me while running 0% Pi up to .30 load. At most my LTFT will go up to +4% previously, but after changing the table to the one I posted above it will go up to as much at 8~9% varying from 0%~9% at idle speeds depending on load and temps. I think the more Pi you run, then more LTFT will vary due to IAT change versus DI.

I'm going to scale my MAF as close to 68* as I can the entire range in 100% DI then take it from there. Good thing it's getting colder in socal, so I shouldn't have too much trouble doing this.

[ztan]

Originally Posted by solidONE

At most my LTFT will go up to +4% previously, but after changing the table to the one I posted above it will go up to as much at 8~9% varying from 0%~9% at idle speeds depending on load and temps. I think the more Pi you run, then more LTFT will vary due to IAT change versus DI.

I get idle speed trim variation of +/- 11 at times. I have chased this in the past and have given up on getting them stable as long as my fuelling is good whilst the car is moving. Seemed to me to be quite weather dependent and I have put it down to temp/humidity shifts whist the car is stationary and the oven under the hood heats up (I also have the Greddy Turbo intake which sits in the engine bay consuming the hot air).