My 5AT starting behaving oddly after a recent reflash to eliminate some DTCs and I've noticed that my base fuel schedule would go to 0 on WOT and calculated load was also dropping to zero. I am using Haltech's MAF output table and I can see MAF volts increasing.

Anyone have any idea how MAF voltage can be ok and the BFS can just drop to zero?

 

I think the BFS is TPS and MAF based. Do you get any new codes? What codes have you disabled?

Seems like an unusual situation which would warrant something popping up.

 

That is helpful to know - thank you. Perhaps I have created an invalid MAF voltage / TPS pairing? I was just trying to move away from being dependent on the MAF because I think mine is dirty. Moving to copy through on the Haltech fixes the situation but my MAF readings are all lower than usual and calculated load gets up to 80% but not 100%. I have a spare MAF - will swap that in or clean my old one and go back to copy through MAF on the Haltech - and then data log TPS and MAF voltage to get a nice stock map. Then use that and go back to the MAF output table and see if that does the trick.

I don't think this is related but this might help others trying to pass OBD - this disable list gives one "not ready" and makes evap "not supported" and allows the others OBD sensors to set quickly for passing emissions. I get no CELS and can pass inspection with this set. The evap ones are the new ones I set this time around.

P0011 intake cam timing
P0021
P0171 system too lean bank 1
P0172 system too rich bank 1
P0174 2 lean b2
P0175 2 rich b2
P0441 evap ctrl sys incorrect purge flow
P0442 evap ctrl sys small leak detected
P0456 eval small leak detected
P1148 closed loop b1
P1168 closed loop b2
P1276 af sensor high voltage b1s1
P1278 af sensor slow response b1s1
P1279
P1286 af sensor b2s1
P1288
P1289

 

UPDATE: Problem solved - I will leave this thread up as it may help others. So in the Haltech I had MAF output type set to "MAF OUTPUT TABLE" and I left copy through enabled inadvertantly. That is not allowed. The Haltech puts the option in red and says "NOT ALLOWED" but I didn't pay attention. I figured it would automatically disable copy through. Anyway, after I manually disabled copy through (which got rid of the big red warning), the MAF output table started working fine.

It is really nice to have MAF output based on MAP rather than TPS. The shifting strength appears to be much more appropriate for actual engine load being transferred through the drivetrain. It is also important to note how relevant calculated load is for the 5AT to be able to hold torque. Back to quick, crisp solid shifts and no trouble holding 18 psi of boost!

 

good deal cuz I didnt get any of those codes, but my MAF is not referenced at all since I cant do pass thru

 

^You are a 6MT, it wouldn't matter for you anyway

 

So problem is not solved. Weird. Base fuel schedule started going to zero, and the calculated load along with it. 5AT just revs under load like the clutch packs are gone. Switched back to MAF copy through on the Haltech, and it's working fine again. I really wanted to use the MAF output table though. Might have to get Hal's help on this.

 

Have you tried just messing with the calc load table instead of playing with maf input to get what you want? Could you log TPS1/2, APP1/2, and MAF V during this issue? I don't see what else could cause this if the inputs are appropriate unless the ECU is going into failsafe or just plain failing.

 

^ I've done that now. I had to reduce calculated load values by nearly 40% to get 100% calculated load on a throttle tap in all gears. God knows what damage I've done to my 5AT in the interim. The 3-4 shift doesn't appear to be holding at 18 psi anymore even at 100% calculated load and even bumping up the torque management tables a bit

Looks like I will rebuild this winter. I still don't know why this changed occured in the first place? I have tested calculated load before and it was always at 100% on 1/4 throttle. Bogus. :/

 

According to the OP you reflashed to avoid some DTC's, right? Maybe you didn't scale the calc. correctly load before you reflashed the new ROM, and ended up with stock (non 100%) values at that point...

 

No I modified my saved ROM. I double checked and it was the correct file. It may be the Haltech 1.10 upgrade change the MAF pass through readings.

I have changed the K value from 29k to 40k and increased all of the torque managment cells above 3200 rpms by 150%. It feels like a VB on top of a VB. A little harsh but she's holding good at nearly 600 wtq power levels .

 

hmmmm 600wtrq and auto, ya i'm def not heading out to the hwy with you anytime soon. LOL

 

Torque falls off pretty quick with my tiny turbos! You worry too much.

 

I was alerted to this thread by my Ninja's. Is it under control now? Is your goal in doing MAF emulation via MAP, to get realistic load signals to the TCM?

 

LOL, i granny shift as well. I'll keep making excuses all day

 

Sorry for the late reply - on the other side of the planet at the moment. My goal was to fix a problem I never used to have. It may have been coincidental, but after I moved to the Haltech v1.10 firmware, my 5AT starting behaving oddly. At least I only noticed odd behavior after the firmware upgrade. Hal is looking into it.

Anyway I have managed to completely resolve the sluggish shifts via Osiris by logging the BFS that matches zero boost and above, changing the K multiplier to ensure that the BFS gets into the far end of my maps, and increasing the torque table values above 3200 rpms by multiplying all those cells by 1.5.

EDIT: During my testing, I did try use an emulated MAF signal entirely and that did NOT work. The BFS would just go to zero in Cipher at WOT. I have no idea why. Perhaps the TPS and MAF signal need to jive for the BFS to be calculated correctly? I have no idea why the stock ECU freaks out honestly. Hence the title for this thread. The BFS will only go to zero IF using MAF output table with the Haltech. If you use pass through and enable copy through, this problem does not occur. Again, this really only affects 5AT users since you need valid load reporting from the stock ECU to the TCM.

 

Haltech Australia has confirmed the bug in v1.10 firmware.

http://forums.haltech.com/viewtopic.php?f=24&t=9202

This does not explain why the stock ECU does not like the fake MAF signal, but I'll be happy with just the copy through working again.

 

turbo 400

/thead


just messing with you raj. that really sucks when the new "upgrades" tend to mess something up that worked. Hope everything works out.

 

^ Ha!

Many thanks to Hal and the Australia team to get a new firmware out so quickly. It works!!!

Any other 5AT Haltech users out there, contact Hal to get the updated firmware. (Also relevant if you use MAF based tuning I suppose).

 

Any idea about the simulated MAF function? Do you have a multi-meter (or preferably a scope) available to test what's going into the haltech vs what's going into the ECM?