ECMlink Help me learn to tune injectors

[dwb]

Today I installed a set of new 1000cc FIC injectors, replacing stock injectors. I'm new to tuning but very seasoned when it comes to DSMs. I would like some guidance on how to tune these injectors. I feel like I have it running OK-ish right now, but need to fine tune it. Prior to new injectors, everything was running great with stock injectors, so there is a solid starting point with no known issues. I need to perform another boost leak test, but it wasn't long ago that I did so with very minimal leaks that *should* be all fixed now.
My build thread should have everything up to date: https://www.dsmtuners.com/dsm-profiles/1991-tsi-awd.16075/

A brief rundown of my current relevant mods:
1991 TSi AWD - daily driver
EVOIII 16g @ ~15 PSI (less than 1000 miles since new)
FIC 1000cc HI z injectors with resister delete plug (brand new, spec sheet below)
91 octane fuel (high altitude)
43.5 PSI base fuel pressure
Walbro 255, rewired
EVO 8 MAF
AEM Wideband (sensor is downstream from stock narrow band)
2g pistons running 2g timing maps
ECMLink v3
Stock exhaust

I spent a few hours getting it to idle and some light throttle while in the garage, playing only with global fuel and global deadtime. I referenced the data in this chart as a baseline but through trial & error and I'm currently at -57 Global Fuel and 637 Global Deadtime. Deadtime seems a little high but it ran better there than dropping global fuel instead.

baseinjectordata [ECMTuning - wiki]

www.ecmtuning.com

I took it out for a short drive and it seems to be a little rich but is idling decent. My latest log is attached.
I tried to get the AEManalogWB AFR close to the AFRatioEST. And WidebandFactor somewhat close to 0%.
What do you guys think my next adjustment should be? What should be my goal as far as ECMLink logs and things to watch out for?

You must be logged in to view this image or video.

 

[Tony1094]

From what I can see, the reason your global deadtimes are so high is it looks like you didn't put the injector data in to the Direct Access InjBatteryAdj table. You'll have to interpolate some values but for example, your log shows @ 12v your injector baseline deadtime is it 840usecs (microseconds), but your data sheet shows your injectors @ 12v (assuming 42.5psi base fuel pressure) should be 1.23milliseconds (ms) which is 1230 microseconds, 14v 940ms, etc.. making that change alone should cut your global deadtime nearly in half.

Can't really offer assistance with anything else as I've only been doing speed density tuning for mine.

 

[steve]

Global Deadtime and InjBatteryAdj are the same thing the two are summed together to arrive at the injector deadtime the ECU uses. I prefer once things are dialed in to update the DA table and leave the global at 0 but that's to eliminate confusion and maybe save a few ECU clock cycles.

Unless Tom and Dave rewrote the way the fuel global works it's not very precise once you get into larger injectors.

I always started out getting the car to start well, get it up to temp and start measuring the AFR or O2 feedback at moderate RPMs under a little load to see how far off the global is and then start looking at idle closed loop to see if the deadtime was making it rich or lean at idle and adjust as needed. One nice thing about MAF cars is the airflow is somewhat calibrated to a standard so you're not fooling yourself playing with the VE table while trying to get fueling right.

 

[Tony1094]

Global Deadtime and InjBatteryAdj are the same thing the two are summed together to arrive at the injector deadtime the ECU uses. I prefer once things are dialed in to update the DA table and leave the global at 0 but that's to eliminate confusion and maybe save a few ECU clock cycles.

Unless Tom and Dave rewrote the way the fuel global works it's not very precise once you get into larger injectors.

I always started out getting the car to start well, get it up to temp and start measuring the AFR or O2 feedback at moderate RPMs under a little load to see how far off the global is and then start looking at idle closed loop to see if the deadtime was making it rich or lean at idle and adjust as needed. One nice thing about MAF cars is the airflow is somewhat calibrated to a standard so you're not fooling yourself playing with the VE table while trying to get fueling right.

When you make the DA change to 0 out the global deadtime, are you offsetting for the voltage differences as well? Just curious how you set it up, I'm still super new to ECMLink and trying to learn it and the theory behind it. I was only trying to let him know why his global deadtime seemed higher than he was expecting.

 

[steve]

Before V3 you only had the global to work with and it was applied to whatever the internal map was interpolated to. Those of us who worked directly with the factory binary would go and modify what ECMLink calls InjBatteryAdj.

It's my understanding in V3 the global is still added to what ever the ECU interpolates from the InjBatteryAdj table for the current voltage but since you have full control now of the deadtime by voltage (InjBatteryAdj and the individual injector adjustments) you can tweek things as needed. So the simple answer is yes. If you have good data from the vendor that you know was derived using DSM hardware, you can just plug those into the DA table and be really close. Generic specs are only going to put you in the ballpark.

This is really a something I'd consider a final pass at tuning, same for adjusting some of the other DA tables that effect fueling and driveability. Kind of like machining something, a couple of rough passes to get in the ballpark and then sneek up on the final dimensions.

 

[dwb]

Thank you @Tony1094 & @steve. Here's a new log just a light cruise around the neighborhood. I adjusted InjBatteryAdj per the injector spec sheet. I kept global fuel at -57 and was able to drop Global Deadtime to 350. I still feel like I'm running rich, very similar to yesterday. I'm not sure if I should continue tweaking the global deadtime or what exactly should be my next move. Steve mentioned going at it from the DA tables, which I haven't messed with yet.

The previous tune from yesterday started and idled fine on cold start today, btw.

Thanks again for your help guys!