Street Build ITFLYS - 1Gina2G

[Justin DuBois]

Switching to narrow band sim, the results are nearly the same.
Raising the RPM to 1050, similar result.
Open loop mode - no improvement.

 

[Justin DuBois]

I took a whack at a boost control setup - here is 2nd gear. 20psi at the peak. I think I can increase the slope of the duty cycle line (Blue) and hold 23psi.

29psi spike - sheesh.. If it could hold 29psi - that would be better. I'm probably going to increase the timing in the spool area, and see what happens, moving the low point from 2 to 5 deg. This will also put the lowest timing dip at the peak torque -as suggested- but not lower than what we see here.

 

[Justin DuBois]

I did adjust the timing in the cells right next to 5 at 750rpm. 10 and 12 deg cells knocked down to 8 and 10, and It did have a slight impact to the idle hunting. I also increased timing in the 6-10K range, from 16 to 18. Lets see if the knock sensor wakes up.

 

[Justin DuBois]

Hello Knock - 17 deg at 6k, is a little too much today.

The blue line is WGSDuty cycle for boost control, and 20psi is the max I can get, so I'm working on holding 20. Next time this should look smoother.

 

[Justin DuBois]

Smoother, but still a little too much timing where there is knock.

 

[Justin DuBois]

Merry Christmas, Happy Holiday, and a Happy new year!
Lovin this setup

 

[Justin DuBois]

This is also why you keep A/C - cause - in these things, you need it to defog the windscreen when it rains.

 

[Justin DuBois]

Idle hunting is almost gone - a timing adjustment in the DA tables for that worked well. I made some final tweaks to the WGDutyCycle table, and turned on error correction, with 20 psi targets. Nice to see it holding all the way to redline.

 

[Justin DuBois]

I went crazy go nuts on a side thread with Timing Tables, and ended up dropping an Evo8 table in for testing. I was impressed with the part throttle performance.

Too much timing was being pulled back at max boost for my setup, so I’m running with a hybrid of both 2G and Evo8 tables.

More work yet to go, the WOT HPbyAir numbers are low at the 20psi row

 

[Justin DuBois]

Just uploaded this adjustment:

The diff map is showing the difference between the EVO8 map and this latest adjustment. Pretty much anything that was more timing from last 2G adjusted map. Really just adding more timing back to where it has already worked previously.

I did get knock - I checked the AFR measured, and it was a touch leaner than targets, so I put in 2% more fuel there to see if that will quiet the knock sensor. Also, I didn't push more than previously good timing in that area that knocked in this attempt until the knock is resolved.

 

[Justin DuBois]

No knock this time - A little more fuel and smoother driving I think. HPbyAir improved with more timing.

This is showing where the Big16G with E25 is off the EVO8 map - so pushing the timing up in the higher boost area highlighted here is having the performance improvement I'm looking for.

 

[Justin DuBois]

Out of nowhere the WOT measured AFR results are 6-7% lean everywere, but idle fuel trims are 14.7, and close to 0%, so I don't think its an actual fueling issue. There is no knock - which I would expect to find if it was actually running 11.8 instead of 11. I wonder if the 02 sensor is on its way out.

 

[Justin DuBois]

I think the issue is with the ambient temps keeping the coolant below 190. The cold pulls read lean - then after the temps heat up past a threshold, the AFR goes richer.

Here is some evidence:
I started off with -43 global fuel, where it has been for weeks with really consistent results. Note BoostEst and MAP are aligned (5500 rpm) - but the AFR is lean, and the coolant temps are 174

After seeing 11.9 during the first full, I pull over make a global adjustment 3% richer to see if the shift was sensor failure or actual lean condition.

And now its gone 3% rich cross the board - the same 3% I just changed in global fuel - so there is another adjustment on top of this making it even richer than the pull moments earlier.

Look at the coolant temp - 206deg, and BoostEst is 2psi different than MAP (5500 rpm)

 

[Dusty Landrum]

This is also why you keep A/C - cause - in these things, you need it to defog the windscreen when it rains.
View attachment 717195

Don't need ac if you don't drive it in the rain... .

 

[Stapl3]

This happens to me when they switch the fuel blends for winter. Here in Michigan it's sold as e85, but its only 85 in june-august, then it's e78, then e70. I noticed no power difference, I could still run any timing and boost I wanted, but the second I fill up and pull out of the gas station in septemberish the global needs a 5% change.

 

[Justin DuBois]

Don't need ac if you don't drive it in the rain... .

Where’s the fun in that! GSX love rain

 

[Justin DuBois]

The rich pull at 206 temp with -40 global, is the the one with 2psi separation in BoostEst to Map.
If I had left the global fuel alone at -43, everything would have snapped in.

I’ll prove this out tomorrow

 

[Stapl3]

2psi separation in AFRest to Map

What does this mean?

 

[Justin DuBois]

What does this mean?

That's a typo - What I was trying to say - At 5500rpm when you have your global fuel calibrated correctly, the BoostEst is aligned with MAP. I was trying to point out that the pull where I added 3% fuel was bad with 2psi difference in boostEst to MAP

Thank you for pointing that out.

 

[Stapl3]

That's a typo - What I was trying to say - At 5500rpm when you have your global fuel calibrated correctly, the BoostEst is aligned with MAP. I was trying to point out that the pull where I added 3% fuel was bad with 2psi difference in boostEst to MAP

Thank you for pointing that out.

That's impossible as you target 5000rpm and do not have a cell at 100.
View the VE table as a %. At some point you will be not at 100% VE, rather 100% _OF_ your peak VE. There has to be a value of 100 at peak VE on the table which is widely accepted to be ~5500 or higher (I'd really love to learn more how to find that exact target RPM) and at that point boostest would match boost because it's at 100 which is what you typed in so it's redundant to look at it or display it in datalogs. I've literally had fighting words over it hah but I digress you do you and I won't press on it [much].

Did you get fuel from a different place very recently? If not, that would eliminate that variable.

 

[CrackedDSM]

Where’s the fun in that! GSX love rain

Definitely gonna have AC in mine even without factoring the rain or etc.

 

[We're on Boost]

Out of nowhere the WOT measured AFR results are 6-7% lean everywere, but idle fuel trims are 14.7, and close to 0%, so I don't think its an actual fueling issue. There is no knock - which I would expect to find if it was actually running 11.8 instead of 11. I wonder if the 02 sensor is on its way out.

The log you posted in this post, there's a good 100 seconds from 231sec to 331sec where it looks to me like your wideband is giving bad numbers.
In that span your narrow band is cycling pretty as a textbook illustration. But your wideband never reads anything less than 15.0 afr. It's showing usually 15.2 to 15.5 around in that range. I don't think that's possible unless you have an air leak someplace in-between your narrow band and your wideband.

Know what I mean? Whenever your narrow band is over 0.45 volts, your actual AFR is richer than stoic, so the wideband should show a number less than 14.7 although it will lag by a fraction of a second behind the narrow band. But there, when your narrow band is 0.8 volts your wideband is still showing lean, 15.x volts! I know, you know, actual AFR numbers with ethanol are different but we talked about that before so I know you know how that works.
Your narrow band looks in good health to me. 0.92 volts sometimes, pretty healthy!

At the end of your log when coolant is at 203 deg, it's less clear if this is still happening because there aren't long idle periods there.
Anyway what I'm saying is your wideband should agree with your narrow band at warm idle when the narrow band O2 is cycling, and that is regardless of anything that might be off with the tune. So I think there is something off with your wideband.

 

[Stapl3]

It's off the whole time, are you looking at log 2024.01.09-01?

It kinda tightens up at the end, but I wonder if that's a learned adjustment. Locking the ECU in openloop eliminates that variable.

 

[We're on Boost]

are you looking at log 2024.01.09-01?

Yes.

 

[Justin DuBois]

The log you posted in this post, there's a good 100 seconds from 231sec to 331sec where it looks to me like your wideband is giving bad numbers.
In that span your narrow band is cycling pretty as a textbook illustration. But your wideband never reads anything less than 15.0 afr. It's showing usually 15.2 to 15.5 around in that range. I don't think that's possible unless you have an air leak someplace in-between your narrow band and your wideband.

Know what I mean? Whenever your narrow band is over 0.45 volts, your actual AFR is richer than stoic, so the wideband should show a number less than 14.7 although it will lag by a fraction of a second behind the narrow band. But there, when your narrow band is 0.8 volts your wideband is still showing lean, 15.x volts! I know, you know, actual AFR numbers with ethanol are different but we talked about that before so I know you know how that works.
Your narrow band looks in good health to me. 0.92 volts sometimes, pretty healthy!

At the end of your log when coolant is at 203 deg, it's less clear if this is still happening because there aren't long idle periods there.
Anyway what I'm saying is your wideband should agree with your narrow band at warm idle when the narrow band O2 is cycling, and that is regardless of anything that might be off with the tune. So I think there is something off with your wideband.

Yeah - I think the sensor is not reading right all the time. It was reading better when hot hot.