ECMlink LinkV3: BoostEst and boost gauge not equal {attached log}

[lil corb]

Adjust your MAFComp as needed to get WBFactor and wideband as close to zero as possible. Just ignore that first 1/2 second or so after you mash the gas pedal.

If there is an airflow calibration issue there, it will show up somewhere else (in another log) instead of only when the throttle is going to 100%...so it should work itself out as you dial in MAFComp at different boost levels.

Got it! Makes sense!

I move to sticky this thread!

 

[snowborder714]

No offense, but this thread is not the only place that we have this info on this site. Yes, it's great that Craig is helping you out, but you aren't the first nor the last.

 

[Calan]

What Brian said. I'm basically just regurgitating and putting a bit different spin on what's been said many times before.

Most all of this (and much more) is covered in the ECMTuning wiki pages.... you just gotta read!

 

[snowborder714]

Most all of this (and much more) is covered in the ECMTuning wiki pages.... you just gotta read at least a few times!

Fixed

 

[razrman8]

Okay, new log. From what I am gathering, I need to do some MAFcomp adjusting from about 400hz on to bring my WBfactor down. Is this correct?

 

[snowborder714]

It looks like you need to start around 500Hz, but yes, most of the MAF Comp sliders from there on up need to be adjusted.

 

[Calan]

Is this correct?

Yes, but I still don't know if I trust your wideband. Part of the pull looks reasonable, but it's reading 14.2:1 at 20.7 secs...and at that RPM/load with 9* of timing in early summer heat, that seems pretty lean to not be getting any knock at all on pump piss.

If it's nice and cool after this recent cold front, maybe it really is hitting 14.2:1 with no problem (it does settle back down rather quickly). What is your ambient air temperature? How much boost are you running?

Go ahead and adjust MAFComp and do another pull, and see if that 14.2:1 area richens up like it should.

 

[razrman8]

It definitely is nice and cool out. It was just 90* out the other day and now it's 66* out. I went from shorts, to pants and a jacket.

I'm guessing this is probably subject to change once it gets warm out.

 

[snowborder714]

I think it may have been mentioned before, but did you try setting up the wideband as LinWideband and configure using the tech article on here?


Craig, is there a way to get people using link to only use Innovate (or one of the few others that work well)?

 

[Calan]

Craig, is there a way to get people using link to only use Innovate (or one of the few others that work well)?

That would be nice

OP - Per that article I linked to...set your AEMWBGauge and AFRatioEst properties to use the same Min and Max values. If you scale them the same and overlay them, it makes it much easier to see where your wideband deviates from what it should be.

I like these values, since it puts 14.7:1 at exactly 75% of the way up:

Min = -4.7
Max = 21.2

 

[MY1GDSM]

Op Set your wideband as a Linearwideband so that you can adjust the reading if need be.

There is a how to a lot of people link to on here that shows you hot to calibrate it while runing. The method I prefer for AEM gauges is.

I'd like to add something along these lines to it if a mod would post in that that how to (has been mentioned? didn't get a response)

Anyway, Basically while logging, remove the sensor from it's place in the exhaust and unhook it it will default to 14.7:1 and that will give you a voltage reading for lambada 1[/I place it into a break clean soaked rag then hook it up(you don't want a warm sensor hitting anything cold). You will be able to see the sensor warm up and Flat line to 10:1 and you can set you lambda to 0.68 for 10:1 and match the voltage up that way(using some common sense as well). this also is an easy test for those gauges to test if your sensor is on it's last leg as it will jump up lean if so after a bit of time if it is.

On top of that your not messign with your tune or dumping fuel at idle. FYI for anyone else You can also do it this way with out the car running so if it's fresh build you know your logs are right with out all the fueling at Idle.

Here is a screen shot

 

[razrman8]

I think it may have been mentioned before, but did you try setting up the wideband as LinWideband and configure using the tech article on here?


Craig, is there a way to get people using link to only use Innovate (or one of the few others that work well)?

I don't recall, I will look it up and give it a shot. I do remember reading it.

You got that right about the Innovate, should have been the one of choice. Guess you learn.

Ugh, was out tampering trying to get the LinearWB thing to work corectly per Gofer's video and I ended up just making things worse I think. I was able to get a idle log, I just feel like everything is off now.

 

[Calan]

I just found this thread again while doing some house cleaning. Did you get everything dialed in and running smoothly Keenan?

 

[razrman8]

Yes, and no. The Gsx was down for quite some time with transmission problems, damn welded differential. Today literally marks probably the fifth day that the car has been up and running. I'm on a break in period for new parts in the transmission, so I can't be too hard on it for now. It's sitting in the garage until I can tend to checking out the brakes in the rear. They need replaced, and him getting some clunking noise. So I want to make sure of everything before I continue driving it all the time.

 

[boost97gst]

I gotta say, this thread almost single handedly, taught me how to tune. Ive got a pretty nice running car right now at 23 psi thanks to you craig, and all others who weighed in. Good luck razr with your trans. If its not one thing its another, these are dsms afterall!! They break!!!!

 

[razrman8]

Thank you sir, everything is going good so far.

 

[ZenkaiRacerDsM]

After reading this thread I almost don't want to switch to speed density.

 

[Calan]

After reading this thread I almost don't want to switch to speed density.

Why is that?

There really is no fundamental difference between tuning a MAF setup and an SD setup. It's basically just a matter of calibrating airflow with sliders in one (MAF), vs. calibrating with cell values in the other (SD). Once that is done, the actual "tuning" process is identical.

When it's all said and done, you're still just trying to get your WB to match the AFR that you are telling the ECU to hit.

 

[ZenkaiRacerDsM]

Hmm, you are correct. I just haven't messed with SD at all yet as I'm still trying to find someone to weld my IAT bung.

So I guess this thread has helped me understand SD calibration just as well! Thank you good sir!

 

[VisionPaintball]

Calan- I have fic 750 injectors and my global is set as if I were using 697cc injectors to get my afr's closer. Is that ok or should I really use the calculated global for 750's (set it and forget it)and just adjust sd table to get afr's lined up?

 

[Calan]

Calan- I have fic 750 injectors and my global is set as if I were using 697cc injectors to get my afr's closer. Is that ok or should I really use the calculated global for 750's (set it and forget it)and just adjust sd table to get afr's lined up?

Ideally, you would want to know what your injectors actually flow, and then set the ECU for that. If the numbers you plug in for fuel are real, correct values (injector static flow rate, fuel pressure, fuel type, etc), then any error in the AFR has to be due to uncalibrated airflow or inaccurate WB readings...at least in theory.

If true fuel flow isn't known, then it's a balancing act between fuel values and airflow calibration, and you just have to hope you hit on a good combination that works at all operating points. You can get decent starting values for FIC 750s from other people that run them, or contact Jens at FIC. They won't be exact since every set has their own personality, but they should be fairly close. If you catch Tom over at ECMTuning on a good day and bribe him properly D), you may get him to test your injectors on a dedicated rig...which will give you the best data available.

IMHO, one of the most common problems with fuel values that seems to fly under the radar is due to the type of fuel. Unless you test what you are putting in the tank every now and then, you really have no idea how much ethanol and other crap is being mixed with the gas these days. Gasoline that contains 10% ethanol requires a surprisingly large adjustment to fuel delivery, especially with larger injectors. If you don't know what fuel blend you are really running, your fuel delivery could be off several percentage points right out of the box.

 

[razrman8]

Hey Craig. Just wanted to let ya know I'll be back in the game here soon. Break in period is pretty much over for the most part. Looking forward to getting this ti.e dialed in.

 

[Calan]

Cool beans. I've got this thread subscribed, so I try to check in when it gets updated.

FWIW - You can still be doing a lot of open loop tuning without thrashing on the car and tranny. In fact, you could lock the ECU in open loop and use the WB exclusively for the entire range of airflow calibration if you really wanted.

This is actually a good way to double-check your fuel trim-based closed loop calibration. If your airflow is correct in the closed loop areas, then your WB error (WBFactor) should closely match your fuel trims at those same points if you lock the ECU in open loop. Of course your CombinedFT will read zero in open loop, so you'll have to either compare logs or switch back and forth between normal closed loop operation and locking open loop in one log.

Long highway trips with periods of varying speeds is ideal for this.

 

[razrman8]

Alrighty. That sounds like a pretty good plan. From what I can recall, WBerror was starting to look really good. I actually had the laptop hooked up to the car today doing a couple logs ( checking coolant temperatures, etc.). I need to go back through this thread and get back up to date on everything. I hate being out of the game for so long.

 

[Calan]

Just remember that in closed loop, the ECU is always targeting 14.7:1, but in open loop it is targeting whatever value is in the OpenLoopMaxOct DA table for any given RPM/Load point.

AFRatioEst will allow you to track the DA table for the most part, although it isn't a direct mapping of it. A more accurate method is to click the "Track datalog" button on the DA table view, and then scroll through the log and watch which cells are highlighted.