TiredGXP wrapped the intake tube on his LS4 in some thermal wrap from a hardware store, and the AFR scatter got way smaller than without it. LS4 is troublesome in this aspect as it's a fwd v8 so the TB/intake tract is right above headers. So it's definitely something good to try out.
Looking at your #'s again just now Bluecat helped me understand why this makes sense--I see now that a lower IAT gets adjusted towards ECT more than a higher IAT. I was thinking it was more linear of an adjustment (for ex: 20 degrees applied to both vs 60 to lower temps and 40 to higher temps)--I guess I didn't look at the data closely enough before.
I though V 2.2.0.0 had the means to monitor estimated MAT (I would like to do this throughout the process), but I can't seem to find it in mine!?
Mine goes behind my radiator for about half the distance--I know that sucks, but it's the ONLY way to route it unfortunately.Originally Posted by redhardsupra
it's platform dependent, they implemented it for some model/years but not others. feel free to request it for more platforms, i think they're tired of hearing from me
If your in MAF mode, I'm not to sure how much changing the bias table in the upper airflow regions is going to help. It would be nice to see some results though. If a person stayed in MAF mode and did some WOT logs with the last few cells of the bias table at 1.0, then did some back to back logs with it at 0.0 - it would shed some light on if the bias table has any bering on MAF calculations.
Like I've said before, I don't think it does. The whole idea of the Bias table is to properly estimate the MAT so that when the SD calc moves from volume to mass, it knows the density of the air that fills the volume. The MAF is already a step ahead because it starts off reading directly in mass. Its not liters per second, it's grams per second. The MAF reading different via temp conditions is a whole other topic of discussion. Even at that the MAF wouldn't be scaled based on estimated or observed MAT, it would be strait IAT if anything.
The changes you made in the low part of the bias table will muck with things in MAF mode because of the dependancy on the SD at idle and other lower airflows, but I don't think it has much effect as the air flow climbs and the computer leans more towards the MAF for fueling.
The main HPT guys that reverse engineer the code could answer this in a split second, but have never seemed to take any interest in this thread. I think its because any answer they give will cause a new question which would cause more greif for them.
Example:
HPT: No, the temp bias table dosen't effect MAF calculations directly.
RHS: Then what effects MAF correction vs IAT
HPT: Another table.
RHS: Then why can't we access it.
HPT: Go away.
Last edited by Bluecat; 02-01-2008 at 08:04 AM.
hmm, I'll try to do some logs for the different bias #s at high airflow when I get a chance. 12.1 cold temp AFR vs 12.8 warm is unacceptable to me though--I hope I can get something to work!
It seems logical to me that the bias would still apply when you're running MAF because the air will still be prone to heatsoak, although that's just a guess of course. Changing the bias in the lower airflow range definitely leaned out my AFR. I raised the bias this morning for the higher airflow and it seemed to lean out the AFR at WOT (just watching my wideband gauge), but I could only hit it for a few seconds since the pavement was wet and I would start spinning pretty badly even at like 80.
not only it would be grief, it would literally obliterate all the current methods of tuning, especially AutoTune, which is what they've been making into a big selling point.
AMEN!Example:
HPT: No, the temp bias table dosen't effect MAF calculations directly.
RHS: Then what effects MAF correction vs IAT
HPT: Another table.
RHS: Then why can't we access it.
HPT: Go away.
Heat soak is still happening, but any heat up or what ever is happening to the air after the MAF is irrelavent. If 20 grams of air go through the MAF, then 20 grams got injested by the motor. Even if it got heated up and swelled to twice its volume between the MAF and combustion chamber, it's still the same amount of mass and requires the same mass of fuel.
Example:
HPT: No, the temp bias table dosen't effect MAF calculations directly.
RHS: Then what effects MAF correction vs IAT
HPT: Another table.
RHS: Then why can't we access it.
HPT: Go away.
2005 Grand Prix GXP - 5.3 LS4 - HP Tuned, MF catback, 1.8 rockers, K&N, Some day I'll finish putting the LS6 intake on
So is there not really much that can be done about MAF correction vs IAT if the bias isn't really affecting it, other than having a winter and summer tune?
I don't know about that. Jacking with the bias might effect MAF, I'm just assuming it wouldn't. Have to see what the test comarisons show if I or anyone else get a change to do some.
Tell you what I did notice the last little bit. When I do a tune I start in SD. After I get the SD lined out I take the last log and do a histogram of the dynamic airflow vs MAF Hz so that I can build a new MAF table that pretty much picks up where the SD left off. Usually it gives me a hell of a start on the MAF table and only requires 1 or 2 more logs before the car is ready for the dyno. Since its been cold I've noticed that my first try at the MAF table is pretty lean.
For example, say running down the road at light cruise the SD tune was coming up with 14 g/sec dynamic airflow - wide band shows 14.6. Even though the MAF is disabled I still log the MAF Hz. In this same frame of the log that it shows 14 g/sec the MAF hz shows 3375. When I build the starting MAF table from the SD tune I end up with 14 g/sec in the 3375 cell. After I switch the MAF back on under the same flow/load I get 15.6 and have to incease the MAF talbe accordingly. In the summer I never noticed this, but I tuned 4 pre 02 LS1's last week in the cold and they all exibited this same trend.
I need to take one of those logs/tune and do what I did earlier in the thread and compare dynamic airflow vs maf hz and the actual table in the tune that was being used at the time. My 04 5.3 didn't show any variance but maybe a car ls1 might. For all I know the truck MAFs with the IAT might be already be correcting themselves electronically and not require a table to correct for temp. (far fetched but possible since those MAF electronics have access to a temp sensor)
I wonder if this could have anything to do with it--the OS/BIN/whatever that is on my PCM is from a 2002 V8 Chevy Silverado, whereas the motor I have is out of a 2003 C5 Z06. The PCM was initially configured for this motor by Jesse Bubb, but I wonder if there is other hidden tables and behind the scenes stuff that is throwing off some things?
Well apparantly the bias doesn't have an effect on WOT when running the MAF sensor. I set my bias to .5 at 150 grams (originally it was .0898, then Bluecat's recommendation) and it still had no effect whatsoever on the WOT AFR (still running at around 12.1 instead of the 12.8 I was getting when it was hot outside).
My next step is to insulate the intake tract and see if that helps at all.
here's a small but important find in the fight for proper temperature accounting:
http://redhardsupra.blogspot.com/200...important.html
it's not quite done yet, i'll have to generate some more graphs tomorrow, i need to nap. for now i have enough that you should see the main point.
Hey, Marcin.
Cool results, I look forward to the "expanded" version.
2005 Grand Prix GXP - 5.3 LS4 - HP Tuned, MF catback, 1.8 rockers, K&N, Some day I'll finish putting the LS6 intake on
Is the "speed" the air speed, the engine speed, or the vehicle speed?
I'm curious about this since I'm trying to get my car to run right when the engine is cold, with a VE table that's dialed in when it's up to normal operating temp. I suspect the bias table is the only knob I have to work with, and I have been.
It runs much leaner than commanded when cold, even though the bias table has become very flat (biasing more toward IAT), on the low airflow end of the table.
The IAT's are typically in the 45-60F range, recently. The WB shows 16.x:1 when commanded is around 12.5:1 and ECT is 60F. As the engine warms up, the commanded and the actual start getting closer to each other, until they are tracking very closely at 180F ECT.
Big changes in the bias table had little effect on this divergence I'm seeing when the ECT is cold.
vehicle speed
remember that when the car is cold, multiple things are out of whack: WB probes are cold, thus cannot be trusted, OLFA is commanding richer than normal, and then there's the cold fuel not atomizing normally, throwing off AFRwb too. it's really too many variables to account for, that's why i tune only on a fully warmed up car (15+mins of normal driving usually does it)
Your going the wrong way. You need to be increasing the number towards the ECT.
My statement of moving towards IAT to increase the temp "compensation" is only relative when we are talking about a fully warmed up engine. ECT being more of a constant and IAT being the variable.
Its just the opposite for what your wanting. If you want the ECT to have a bigger effect, you need to move towards it. But you can't look at it like trying to richen it up when cold. Because at cold start the IAT and ECT will (should) be identical and the Bias dosen't have any effect. What will essentially happen is you've got to let the bias lean it out when the engine is warm and then increase your VE to richen it back up everywhere.
Also, like RHS said the OLFA is commanding rich AFR's when cold. They don't really want a 10:1 AFR when the engine is very cold, but that is what it takes for the engine to run right due to things he stated above. Its one time that you need to ignore the AFR % error. Ignore the number, just command it fat enough that the wideband says 13-14 range while warming up, even if you have to command somthing like 10 to do it. Then when it warms up try to make the commanded match the observed.
If you don't want to have all the time tied up in changing the bias and redoing the VE table, just crutch it in the OLFA table. Its not right, but thats how most people would do it anyway.
Makes perfect sense now. I started to stumble onto that yesterday when I was posting, when I put in some more text to support what I was saying, but the additional text ended up (after reviewing it) being contrary to what I thought, so I took it back out until I could think about it some more...
Good to know. I was thinking the WB needed to show the commanded afr when cold. I will shoot for the 13-14 range, should be enough to eliminate hesitation when getting the car to first start moving. I usually let it warm up all the way anyway because of the aftermarket valve springs...
That was going to be my next move if I couldn't get the bias figured out. Now I think I'll concentrate on a bias that is best for cold/hot ambient air at normal ECT. If that happens to work out right for cold start too, then great. If the cold start is still off then I'll hit the OLFA table.
Reply With Quote