Has anyone found a charge temp solution?

[erics_02_z06]

Just spent an hour reading old posts, but couldn't find an answer....has anyone found a way to avoid the running rich at night, and lean in the day issues?

[S2H]

Just spent an hour reading old posts, but couldn't find an answer....has anyone found a way to avoid the running rich at night, and lean in the day issues?

yes... tune the complex cylinder charge temp tables

I do it with a wideband...
drive around and log the table vs IAT plotting AFR error...
if it goes lean when IAT's go up then you need to increase the values in the cells it represents...if it goes rich... you need to decrease them...

make sure you are in SD when you tune them.... you can do it with the MAF... but it doesnt work nearly as good or as quickly

once you have changed your bias table.. retune VE...

then rinse lather repeat...do the same thing again.... if its still leaner or richer then move the table some more...and go back and retune VE again...
keep hitting it until its all good...
its best to do this by driving in town for a while so IAT's go up.. then get on the highway to get some airflow to cool it down, then go back to stop and go, then back to the highway..

you need a good solid 30-45 minutes worth of data every time....

takes several days to get it right...sometimes several weeks depending on how far you move it and how close you want it to be.

have to do it with the wideband...fuel trims off and reset to 0...
cant do it with fuel trims...they learn too slow and you get bad data

personally I also retune my VE with parameters telling it to only count data within a certain small temperature range...usually the range at which my car drives in most of the time when cruising down the highway on a warm day..example being I tell it to only do VE table data if IAT is between 95~105*(seems to be what my IAT settles at most of the time right now)

you will find that your VE values will be quite a bit higher if you have to raise the cylinder charge table and lower if you have to lower it...

then you can go back and tune your MAF tables once you are done...

[erics_02_z06]

ok, is this something that I can use off of someone elses tune since its IAT ECT and just g/cyl? Maybe someone who lives in a climate like Salt lake city (hot dry) can post thier table?

[S2H]

only use one from a vehicle that is set up exactly like yours and is in your area...
really needs to be tuned per vehicle...

mine starts up at a value of 1.3 and makes a pretty nice regular curve downwards and outwards... but a buddy of mine has a curve that starts lower and ends higher than mine.

partly dependent on IAT location and heatsoak issues.
I dont have any real heatsoak issues...yes mine gets hot, but the location of my iat allows it to cool down almost instantly when I open the throttle

[erics_02_z06]

ok, sounds good. Does anyone know the equations where these factors are used?? I might try to solve for a new correction factor to get a starting point to tune off of....I'm getting pretty tired of logging...

So how does this correction get used with the VE to calculate g/cyl. And this will sound really dumb coming from someone who says he wants to solve something, but if my AFR is 5% off, is it safe to say thats becausemy g/cyl is 5% off?

[S2H]

ok, sounds good. Does anyone know the equations where these factors are used?? I might try to solve for a new correction factor to get a starting point to tune off of....I'm getting pretty tired of logging...

So how does this correction get used with the VE to calculate g/cyl. And this will sound really dumb coming from someone who says he wants to solve something, but if my AFR is 5% off, is it safe to say thats becausemy g/cyl is 5% off?

its used in all fueling...
and it cant be blanketed as saying the whole table is 5% off... it may be off 30% in the lower cells and 2% in the higher cells

your x-axis is dynamic clyinder air, your y-axis will be IAT in 5*F steps


the table is basically lowest value = all IAT, highest value = all ECT, dead middle is an even blend of IAT & ECT.

have fun... took me about a week on my car.

[erics_02_z06]

so I tried to get the histogram right...dynamic air flow (PID) on x, IAT on the y, with the cells being populated by AFR....but I ended up with only 4 boxes....

[S2H]

so I tried to get the histogram right...dynamic air flow (PID) on x, IAT on the y, with the cells being populated by AFR....but I ended up with only 4 boxes....

heres the histogram I use..

be sure you log Dynamic Airflow in lbs/hr, and IAT in Fahrenheit

[Bluecat]

This is probally the most intelegent thread I've seen about the bias table in a long time. Good info. Good implentation.

Soundengineer, It's been to many years ago, but seems like back then you were the guy telling everyone to set the table to .5 across the board. My appologies if that wasn't you, if it was - thanks for reevaluating the bias table. There was a lot of people taking that as the gospel when I felt that was bad info.

[LSxpwrdZ]

I still wish we just had access to the physical IAT adjustment/correction factor.

[S2H]

This is probally the most intelegent thread I've seen about the bias table in a long time. Good info. Good implentation.

Soundengineer, It's been to many years ago, but seems like back then you were the guy telling everyone to set the table to .5 across the board. My appologies if that wasn't you, if it was - thanks for reevaluating the bias table. There was a lot of people taking that as the gospel when I felt that was bad info.

I was telling people to start in the middle of their bias values and work outwards...
ls1's start at 1.0...ls2's start at 0.5

[Higgs Boson]

I am running OL only now but have had much better luck running the current release instead of the beta.....

Any chance you could post the histogram info so i can see it too? :-) Thanks!

[knyght4]

I got interested in checking this out, but something isn't right. I got this table set up in my Scanner, and went into the Editor to see what my current bias table looks like.

The Scanner table set up as 3D, but my bias table is only 2D and I found a bias filter table.

I'm running the latest 2.23 Beta, and I was looking in Engine > Airflow > General Airflow > Cylinder Charge Temp. Is this the right place for the table?

[Higgs Boson]

yes it is a 2d table. i don't think you can use copy/paste on this one. i think you have to interpret the hist and manually adjust the bias table then recheck it. that's why it takes so long.

[knyght4]

yes it is a 2d table. i don't think you can use copy/paste on this one. i think you have to interpret the hist and manually adjust the bias table then recheck it. that's why it takes so long.

Interesting. I'll simply have to play around with it some and see what happens.

[LSxpwrdZ]

To give a little better understanding mathematically, this is what I have found to be the calculation used. The GM code calculates the charge temp in degree's Kelvin and then uses this function:

273.15+IAT+((ECT-IAT)*factor)

factor is the value lookup on the Bias table.

If you want to do everything in Fahrenheit then this is the same function:
32+IAT+((ECT-IAT)*factor)

[redhardsupra]

So to summarize: mock up a histogram mirroring the Bias table and tweak it accordingly to what AFR%error tells you to do.

Let's think about it for a second. What does this method achieve?

Bias is a part of the temperature estimator. Temperature, in combination with VE and MAP is used to create airmass. That would only work, if your VE is a known quantity. But it isn't. You need VE to calibrate Bias, and you need Bias to calibrate VE. It's a vicious circle.

For the more enginering-minded, the entire SD calibration process is described with:

AFRwb*IFR*IPW=GMVE*MAP/TEMP

You cannot solve this equation if you have two unknowns, because it's an underdetermined equation. This has nothing to do with tuning, that's just math.

So dear SoundEngineer, your idea of holding VE constant, adjusting Bias, then swap, rinse, repeat approach yields nothing better than endless running in circles.

Love,
--Marcin

[S2H]

I totally agree with you that it should run you in circles...
but if you move in small steps...and use only data from a specific temperature range.. then you have set an IAT temp...and you can get VE for that IAT temp...plus yuor ECT should be within a fairly stady range at cruising speeds...
so you have just solved for the variables needed....

you can then proceed to solve the values above and below your variables...

it will be vehicle dependent because one guy does it in 70* IAT's and another guy does it in 120* IAT's.... but you can solve it for your vehicle.

the Key is to tune the VE with those small IAT/ECT ranges and use the bais to solve VE for temps above and below...

either way.. if it goes lean as your IAT's go up, then you need to add to the bias temp table.. if it gets richer.. then you need to decrease the values in the bias temp table.

if you want to get really picky...
solve these values with a stock GM table..where VE is less than 100%...find what temps you see no difference in values from the stock table... then you can make that the starting temp to build from....but honestly...it may be a full year of weather to determine that...or a really good wind tunnel with good temperature control like GM uses.

[erics_02_z06]

so how would you propose this gets fixed? Also, how much of a variance is acceptable I go from +6 to -5 at night on my LTFT's

[S2H]

Also Marcin... if it didnt work.. then I wouldnt have my fueling consistent from 60* in the morning after a cold front and 95* in the middle of the afternoon.......
granted there are minor variances... but it basically stays within 3%~4% overall thru the range.. I consider that extremely acceptable in any book of EFI tuning