PDA

View Full Version : HELP ME PLEASE!!! Minimum pulse width?



LS1FREEK
June 19th, 2017, 02:09 PM
Guys this has been killing me, literally... So I'm gonna do my best to try to explain whats going on here.

I'm working on my brothers 1971 Nova, LQ4/4L80e/Turbo swap. 2000 Z28 harness and ECM. 02020005 Custom OS (based it off of a 12212156 2002 Z28 file).

On his first build we had some what we believed to be Siemens Deka 60# injectors, thats where it all started SMH... I got inj data from FIC for those, typed it all up in the tune, but right off the bat, tune seemed rich everywhere in light throttle areas and lean in heavy throttle areas, as if the injectors weren't very linear. Called FIC, they were maybe assuming i didn't know much about tuning, after i explained i had been successfully tuning cars for over 9 yrs he started to think the injectors might be ebay knock offs. We sold those and bought some new 95# FIC's.

Well after installing the new FIC 1000cc Bosch injectors, I started from scratch with a new tune WITH FIC injector data just how I've normally done successfully with all my builds. Started the car and immediately noticed it was rich at idle and light throttle just sitting there, so started taking fuel out of the VE table, but it made no difference, i even tried drastic numbers like -200%, no difference. I sized the injectors bigger in the tune and that did the trick. Finally got the STFT close to 0% at idle and very light throttle, but then any moderate throttle it would go extremely lean, even with the VE at like over 120% at 12psi map and it was still lean.

I moved the injector flow rate back to where they were supposed to be and noticed that the minimum pulse width the PCM would ever command was around 2.319ms. So at idle and and light throttles it would never command less than 2.319ms, EVEN if i commanded leaner AFR in the DVT section of the scantool.

DATALOG ATTACHED

So here's whats weird, IF i turn fuel trims on, it'll pull fuel and idle/cruise 14.6 AFR... some times itll pull upto -70%, the problem with that is then when you tip into the throttle it will hesitate and go lean for a sec before it recovers.

SOOOOO FRUSTRATING!!!!!

Ive been tuning Gen3 LS' stuff for over 10 years, never had this problem before. In fact, Ive tuned multiple trucks also with FIC 95's and never had this problem. I figured there must be an underlying issue, but today I was tuning another car with nearly the same setup (except 5.3 and truck harness) and it is doing the exact same thing to me...

MY BROTHERS TUNE AND A DATA LOG IS ATTACHED


21224

21225

LS1FREEK
June 19th, 2017, 02:12 PM
Also, some stuff may look a bit wacky in the tune and thats because those settings seem to help a little.

ProperTuningOG
June 19th, 2017, 02:47 PM
Wow, those must be some kind of trick/magic injectors to have minimum PWs that low. Care to share the supplied injector data? Why no wideband in the log?

LS1FREEK
June 20th, 2017, 12:11 AM
Now I messed with B4006 (small pulse threshold) and set it real low, seemed to help a little. It's supposed to be at 2.500. Also B4003 and B4004 are both supposed to be .125 across the board. I also messed with B3701 a little. But like I said that doesn't help at all.

Site wont let me upload the inj data sheet from FIC.

joecar
June 20th, 2017, 05:34 AM
Hi John,

Questions:
a. is the FPR manifold referenced (expected answer = yes)...?
b. are you getting an immediate MAF DTC (expected answer = yes)...?

joecar
June 20th, 2017, 05:42 AM
( BTW: set B3609 to 0.050 s )

joecar
June 20th, 2017, 05:48 AM
Commanded AFR is doing strange movements...

ProperTuningOG
June 20th, 2017, 07:22 AM
Do you understand what all the injector setup does? You need to eliminate the error in the opening and closing times below sustainable accurate pulse width. The way you have things setup is not correct, you have eliminated all small pulse correction. This is why I asked for the injector data. Seeing an injector that size with a sustainable non corrected minimum PW of 0.151975 would be straight magic. I know you are gonna say where the data came from and its been a decade worth of tuning. Proper setup of base calculations kills majority of tuning issues. Understanding what the ECM does with this data is just as important as using accurate external devices to collect the needed data for tuning.

You have not changed the minimum transient setting B9021, its still set as stock 5.7.

LS1FREEK
June 20th, 2017, 08:13 AM
Hi John,

Questions:
a. is the FPR manifold referenced (expected answer = yes)...?
b. are you getting an immediate MAF DTC (expected answer = yes)...?

Yes and yes


Commanded AFR is doing strange movements...

That was me commanding it to in the DVT controls. As you can see injector pulse width does not drop, even when afr is commanded really high, however when I richen it up, it responds fine.

LS1FREEK
June 20th, 2017, 08:17 AM
Do you understand what all the injector setup does? You need to eliminate the error in the opening and closing times below sustainable accurate pulse width. The way you have things setup is not correct, you have eliminated all small pulse correction. This is why I asked for the injector data. Seeing an injector that size with a sustainable non corrected minimum PW of 0.151975 would be straight magic. I know you are gonna say where the data came from and its been a decade worth of tuning. Proper setup of base calculations kills majority of tuning issues. Understanding what the ECM does with this data is just as important as using accurate external devices to collect the needed data for tuning.

You have not changed the minimum transient setting B9021, its still set as stock 5.7.

Like I said, all these problems were and still are problems even when I had the set the way they're supposed to be 2.5ms on the small pulse threshold. And minimum pulse width is supposed to be .125ms, according to data. For some reason my search isn't coming up with anything when I search B9021.

LS1FREEK
June 20th, 2017, 08:23 AM
I can email the data, if you wanna give me your email. You can also text me @ 6206755626

5.7ute
June 20th, 2017, 05:54 PM
0.125ms is the same as what ID injectors specify IIRC. FIC have reportedly "borrowed" this data from ID and passed it off as there own.
You will need to load the cax file for B9021 to be able to modify this parameter. Look here: https://forum.efilive.com/showthread.php?17699-need-B9021-for-01250003/page2&highlight=B9021

ProperTuningOG
June 20th, 2017, 06:08 PM
I'd be interested in having a look at the borrowed data. Tuning@team208motorsports.com

ProperTuningOG
June 20th, 2017, 06:11 PM
0.125ms is the same as what ID injectors specify IIRC. FIC have reportedly "borrowed" this data from ID and passed it off as there own.
You will need to load the cax file for B9021 to be able to modify this parameter. Look here: https://forum.efilive.com/showthread.php?17699-need-B9021-for-01250003/page2&highlight=B9021

In regards to a PW that low. The short pulse info would have to be within a few percent accurate. I know most places stop testing around 1ms and fudge below that part.

5.7ute
June 20th, 2017, 06:28 PM
All the info can be found here http://injectordynamics.com/articles/the-same-as-injector-dynamics-but-cheaper/
What you need to remember is that the 0.125ms is the fuelling portion of the pulsewidth which when added to the offsets etc give the final pulsewidth. There are some great articles on that site that breakdown the GM injector model, making things much clearer.

ProperTuningOG
June 20th, 2017, 06:38 PM
I totally get that but my point was, in the tune in question, small pulse offsets are not even being used.

5.7ute
June 20th, 2017, 07:09 PM
From what I understand he incorrectly lowered the short pulse adder threshold to bring the final pulsewidth down, without realising the pcm is hitting a pulsewidth floor that lowering B9201 will address.

LS1FREEK
June 21st, 2017, 06:19 AM
Thanks guys, I'm gonna look into the B9021 settings and see what that does

Edit: B9021 lol

joecar
June 21st, 2017, 06:43 AM
B9021


( I'm sidlexyc too )

ProperTuningOG
June 21st, 2017, 09:54 AM
I'm finding various data sheets on the bosch 1000s. Which version do you have height wise?

ProperTuningOG
June 21st, 2017, 09:57 AM
21228

I changed the transient pulse and short pulse threshold. Rest is the same. Try that and see what it does.

LS1FREEK
June 21st, 2017, 02:54 PM
That did it!!! I've been struggling with this for the last 6 months! You guys don't even know my excitement levels right now!

5.7ute
June 21st, 2017, 03:04 PM
Common problem with big injectors with large offsets, glad you got it sorted.
I had a mate call with a strange issue 2 days ago where idle fuelling was fine, but put any throttle into it and it would peg rich no matter what changes were done to VE etc. In this instance it was the stock default minimum pulse width settings causing the issue. Once my car is repaired (cam and lifter failure) I will be doing some more testing to see if I can replicate his issue and update my earlier findings on IBPW composition.

LS1FREEK
June 21st, 2017, 03:23 PM
I really appreciate the help! On my brothers car, while idling in open loop with commanded AFR at 14.6, it was roughly 12-13 on the wideband, I could go into DVT's on the scan tool and command 20.0 AFR and the wideband wouldn't budge. Now it idles at 17 afr on the wide band when commanded afr is 14.6, if I command anything leaner with the DVT's it responds great, if I slide it to far over it'll die. So it's finally working! Now I've gotta build the VE! Yehaw lol