pcmhacking.net

Hi all.

I'm a little confused about some tuning results that I've been doing.

I am using a direct copy of a tune I had done by a dyno tuner who used $A5 and burnt the tune to a memcal.
I am using $12p real time with all the same values in the VE tables, spark tables etc...

After a few long drives I think I have found and issue with consistency... I have noticed that after a decent warn up period, if I then turn the motor off and let the heat soak through to the inlet manifold (it's runs pretty cool when the car is on) My MAT reading goes up to the 40's and it seems to be making the tune very rich. Alternatively if I take it out to the open road and give it a boot full its excessively (dangerously) lean. I would have thought the reverse should be occurring, ie. Cold air = more fuel, Hot air = less fuel. 36 seems to be the sweet spot.

Also the car seems to be happy to cruse at 100km (3000RPM) with and AFR well into the 16's without bucking, missing or running hot??? is this normal, os is it a grenade waiting to blow up?

I've attached the Bin and 3 log's

there are so many settings that work together its near impoossible to transfer a tune over and have it run perfect especially when 12p has extra settings and does things a little different.
you just need to start there and tweak the air temp related tables to get it right

the other thing to take into account as well, which i've found with 202's having the MAT in a totally different location to a commodore with mean heat soak is nothing like it might be on a commodore so like delcowizzid says hard to pick the one thing effecting it ...

Where are the settings controlling the MAT readings?

Charge temp- Inverse Temp Term vs charge temp?

You will need to play with "Inverse Temperature term vs Charge Temperature" and "% Coolant contribution for calculating charge temp Vs RPM & MAP". The inverse temp is the easier of the two and is sort of an overall adjustment, if you have little lean/rich spots that vary with temp (make sure they are not from incorrect VE though) then you can adjust the coolant contribution value - this is a difficult table to adjust and get good results.

Thanks,

It looks to me that if I can get the tune to run as if the MAT was in the mid 30's the whole time it would be pretty sweet.
I know the VE table and Target AFR table should be around the money because they are the ones used on the other tune. (Providing $12p doesn't screw with it too much as mentioned above)
Adam

Just looking at the table now....

there are 2 columns. 1 fixed (deg C) and one I can adjust (deg K)

What effect does raising or lowering these numbers have? (I'd rather have some clue before I dive in and play)

The ecu uses the table to convert between kelvins and celsius. The kelvins side is converted to degrees C by tunerpro, so in its default state the numbers should be the same in both colums. Changing this table is a bit of a hack, but it works. So if you want the car to think 30 degrees when its reading 40 degrees, put 30 in the 40 row. As said, its hard to get good results this way, but this is what I know on the subject.

just for your information, your Target AFR table from what i can recall of the A5 tune was screwed with!

i've been thinking about it and how to get it right, only thing i can think to do is get an area, say for example 3600rpm 100kpa, its 11.90 AFR .... i'd think more like 12.6-12.8 would be what you want, so you need to change the table to that .... BUT if you do this you'll lean out that point, so what i'd then do is 12.8 / 11.9 = 1.08 ... so then go to the same point in the VE table and multiply its number by 1.08 (add 8% more fuel).

this is a pain in the ass process, but if you look at the original OSE12P AFR table its nothing like yours.

and i have a feeeling your tune will not be actually commanding the AFR's that are in the table, more so its been fudged to get the AFR's reading right on a dyno or similar.

The inverse temp table is part of the charge temp calc and where you want to play to make the AFR correct for all temps. A MAP sensor based ECU works on the ideal gas law, but for the processing power of the CPU its too slow to calc on the fly, so a table is used instead. There have been a few posts on it, but its prob more info than you need to know ..
https://pcmhacking.net/forums/viewtopi ... gas#p17969
https://pcmhacking.net/forums/viewtopi ... gas#p16809