d2903 factory set at 10?

[Ritch]

I have a 99 5.3 truck and the factory set d2903 (minimum TCC PWM) is set to 10. From what I have read this should be way higher, like 75-90. Maybe this is an OS issue?

Also, you might notice base pressures are set low. I have adjusted because the shifts were too hard after Trans-Go shift-kit. Maybe there easy an easier way, but it helped alot.

[joecar]

Hi Ritch,

You only need to bump up D2903 if the TCC is not being locked all the way... i.e. if you find TCCDC stuck below 90% for more than 5-10 seconds at a time... if you see that happening then set it to 90 (but make sure that it is lower than D2904).

The shift kit physically increases the apply rate (increased line pressure, increased flow to apply component)...

The three things that you can use to reduce the level of hard:
- reduce shift pressure tables (like you did),
- increase shift time tables (but maybe leave these at zero with the shift kit),
- increase shift torque reduction tables (you already have this).

If you're getting a hard shift at low throttle, then increase the low end of the TR tables.

After reducing the shift pressure tables, how different is the shift feel...?
Does each shift still give a positive/definite feel (across the throttle range)...?

[Ritch]

Thanks! The truck is not back on the road yet. The snow has finally gone and the streets are are being swept. I will may try the shift time factor. The kit increased accumulator pressure significantly too, making the 1-2 shift a real chirper.

[Ritch]

TCC seems to be slipping a bit while cruising up hill. TCCSLIP shows this in the attatched file. You are saying increasing D2903 should illiminate this?

BTW shifts feel good. Easily felt. The torque convertor is now a 3200 from 2500, I think that helped soften things too.

[joecar]

What is TCCDC when it is slipping...?

If the TCCDC is not above 90% when you detect slipping, then trying setting D2903 higher (but make sure you keep the relationship D2903 < D2904 for the same cells).

[joecar]

In your log, when TCCMODE=Lock, the most amount of TCCSLIP is 95 rpm... is that what you mean...? In this area the TCCDC is 30-40%...

I then see the TCCDC go to to 58% and TCCSLIP goes down to 2 rpm.

Try bumping up D2903 and see what happens.

[Ritch]

TCCDC seems to be about 35% and TCCSLIP peeked at 70rpm while TCCmode is LOCK

D2905 was set from factory to be 0 when TCCslip is 0. I will change that aswell.

[joecar]

Note that TCCSLIP is valid only when TCCMODE = Lock.

[joecar]

Those ripples are strange...

It looks like the PCM lowered TCCDC until those TCCSLIP ripples occurred...

I can't tell if the PCM is performing a test (why else would the PCM command a ripple in TCCDC) or what.

Those ripples also show up on the RPM waveform...
I don't think it is due to misfire...
if you're running with a MAF then also log MFTOTAL and MAF...

you can get rid of some pids like these:
- PRND
- PRNDSW-xxxxx
- SHIFT32
- SHIFTxx
- SHIFTxxERR
- TFMCUR
- TFMDES
- TFMDC
- TSHIFT
- TURBINE
- TSTATE01

Keep TSTATE00 and the remaining pids.

Interesting log, thanks.

:)

[joecar]

The tun file in post #7 shows D2903 to be at 75% or higher in all cells... so look at D2901 (zero)... try setting that higher and see what happens.

[joecar]

TCCDC seems to be about 35% and TCCSLIP peeked at 70rpm while TCCmode is LOCK

D2905 was set from factory to be 0 when TCCslip is 0. I will change that aswell.

D2905 immediately around zero slip won't make much difference since this is the rate at which TCC pressure is ramped up (i.e. psi per second of ramp up)... if the TCCSLIP happens to be zero (usually is not) when the PCM decides to command TCC lock, then the TCC will lock easily without requiring the ramp up to be faster... but it won't hurt to try it.

[Ritch]

moving d2904 and d2903 made a huge difference.

[joecar]

That does make a lot of difference... :cheers: ...that will improve the life of the TCC.

Take some more logs: get to 4th gear and get more throttle.

[Ritch]

newbie question

How to you set autoscan to display entire log on one screenshot? :doh2:I'll look it up.

Anyway, I found the slipping test parameters, but previous log showed ripple at 20% not 10?

This is my progress thus far, still too rough on up-shift, and TCC lock-up is sharp enough to cause drive-line flap at light throttle. Especially important in quiet neighbour hoods.

[joecar]

You mean to see the whole log in the scantool chart... see button "All" circled in red in attached pic.

[joecar]

You mean the slipping test parameters E0801,2,3,4...?

Those are for the transmission slipping in gear, not the TCC slip...

Altho I think for the 4L60E the PCM has to lock the TCC in order to perform the trans-in-gear slipping test.

[Ritch]

That was easy, thanks again.

I think I need to calibrate the force motor current to match actual line pressure. That way might help. I assume line pressure may be different than what the force motor setting are adjusted for. I really noticed a big change when I leveled out the trans press/temp adjustment. For now, I am adjusting the force motor tables by 10% at a time...using the old seat-o-meter.:angel_innocent:

[Ritch]

TCC does freeup when I let off the thottle. I wish it wouldn't.

[joecar]

In post #14, that's a good log, it shows:
- 2 rpm or less TCCSLIP when locked,
- TCC locked during upshifts,
- TCCMODE going to Coast,
Very interesting log...:cheers:

Reduce your pid channel count to 24 or less to get faster pid update (10 Samples/second instead of 5 S/s)...
[it seems to me that a rapid TCCMODE pulse may have been missed because of 5 S/s]
Remove the following pids (they don't tell us anything):
- TRQTRANS,
- SHIFT32,
- MFCSTAT,
- MFABNORMAL,
- TSHIFT.

Good log, thanks for posting... :cheers:

[joecar]

...

I think I need to calibrate the force motor current to match actual line pressure. That way might help. I assume line pressure may be different than what the force motor setting are adjusted for. I really noticed a big change when I leveled out the trans press/temp adjustment. For now, I am adjusting the force motor tables by 10% at a time...using the old seat-o-meter.:angel_innocent:
Have a look at the Torque Signal hydraulic circuit in this diagram (bright yellow, see detail 1c and 1a): showpost.php?p=39472&postcount=4 (http://forum.efilive.com/showpost.php?p=39472&postcount=4)

the TFM (aka PCS) regulates Torque Signal pressure which biases the Boost Valve which (via the Isolator Spring) assists the Pressure Regulator valve in regulating Line Pressure...

i.e. the Torque Signal pressure range 0-96 psi maps to a Line Pressure range of something like 55-230 psi (or so)...

it's not easy trying to calibrate the TFM/PCS Current table(s)...

I assume you have not altered the screw setting in the back of the TFM/PCS...

[joecar]

TCC does freeup when I let off the thottle. I wish it wouldn't.The 2001/2002 F-car (Camaro/Firebird) files will keep the TCC locked when throttle is released (and MPH is above TCC Release curve)...

If your PCM is compatible with either of these, then try flashing one of these in (you will need to copy all your tables over by hand):
- OS 12202088 from 2001 F-car with A4,
- OS 12212156 from 2002 F-car with A4.

First copy all your tables over to the F-car file, then do a full flash.

[ScarabEpic22]

Yep, only car OSs can keep the TC locked with 0% TPS. Same goes for E67s so I assume E40 and E38s follow suit. But you can flash a car OS in, Im going to play with that this summer when I have a chance.

[joecar]

...
Remove the following pids (they don't tell us anything):
- TRQTRANS,
- SHIFT32,
- MFCSTAT,
- MFABNORMAL,
- TSHIFT.
...Since your log indicates the MAF has failed (SD mode), you can also delete MFTOTAL (since misfire detection doesn't operate when MAF is failed).

[Ritch]

pc valve was turned down a little, last time the tranny got rebuilt. I think the accumulator springs were changed with Trans-go.

Thanks for the tip about no misfire with SD. I actually thought I might have a lo rpm miss, but trusted the ecm to tell me about it.

I can't remember why I have not switched to the car OS.

test/tune run

[Ritch]

as long as I keep TCC above 35% the torque won't slip.

Joe in SD you say misfire detection in disabled. Is that true for "rough road" TCC release still active? or is that the same thing?

obvious intentional slip at the end of this log

[joecar]

...
Joe in SD you say misfire detection in disabled. Is that true for "rough road" TCC release still active? or is that the same thing?
...Ritch,

This is what I read in the service manual...



Misfire Monitor Diagnostic Operation (http://127.0.0.1:9001/servlets/BlobShtml?ShtmlFile=569739&evc=sm&pubid=151&cellId=32568&mspsdsubkey=1227#ss16-569739)

The misfire monitor diagnostic is based on crankshaft rotational velocity (reference period) variations. The PCM determines crankshaft rotational velocity using the crankshaft position sensor and camshaft position sensor. When a cylinder misfires the crankshaft slows down momentarily. By monitoring the crankshaft and camshaft position sensor signals, the PCM can calculate when a misfire occurs.
For a non-catalyst damaging misfire, the diagnostic is required to monitor a misfire present for between 1000-3200 engine revolutions.
For catalyst damage misfire, the diagnostic responds to the misfire within 200 engine revolutions.
Rough roads may cause false misfire detection. A rough road applies sudden torque variations to the drive wheels and drivetrain. This torque can intermittently decrease the crankshaft rotational velocity. The ABS system detects uneven speed between the vehicles wheels and sends data via the serial data bus to the PCM to disable the misfire monitor until the rough road is no longer detected.
On automatic transmission equipped vehicles, the Torque Converter Clutch (TCC) will disable whenever a misfire is detected. Disabling the TCC isolates the engine from the rest of the drive line and minimizes the effect of the drive wheel inputs on crankshaft rotation.
When the TCC is disabled as a result of misfire detection, the TCC will re-enabled after approximately 3200 engine revolutions if no misfire is detected. The TCC remains disabled whenever the misfire is detected, with or without a DTC set. This allows the misfire diagnostic to reevaluate the system.

(I wonder how many man-years GM spend developing this distributed algorithm, that required some major R&D)


The misfire monitor is disabled when rough road is detected... so misfires cannot be detected... so the TCC won't be disabled (since no misfires are detected)... that's my simplistic understanding of it... :doh2:
TSTATE09 or TSTATE10 mentions "TCC off due to misfire", and I don't see any flags for rough road... so it's hard to tell when the PCM is aware of rough road.

[joecar]

I see in your log when TCCDC is below 35% that slip appears and has a ripple to it.

Good log, thanks for posting.

[Ritch]

Thank you Joe

One day, I'll have to pay some more money and upgrade EFIlive so I can change OS.

[joecar]

You're welcome... :cheers:

To upgrade a V1-Personal to V1-Commercial is $50... see here (http://www.efilive.com/index.php?page=shop.product_details&flypage=flypage_EFILive.tpl&product_id=54&category_id=8&option=com_virtuemart&Itemid=56).

The Commercial lets you do full flash (i.e. so you can flash the OS).

[Ritch]

One more thing,

Will my PCM accept the new OS? It appears to be the 3 plug device. Picture attatched.

Rich

[ScarabEpic22]

That appears to be a regular LS1 99-02 512kb PCM, so yes you can flash a 2002 OS in and use a COS.

[Ritch]

:cucumber:

[Ritch]

Warning flashing a truck os12212156 that had a 4l80 will not work well :doh2:

[joecar]

Hi Ritch,

TIS reads zero always.
No shifting above 2nd.
TSHIFT pattern is toggling between 2WD and 4WD.
Looks like is applying the TCC in 2nd but TCCSLIP is not dropping down.

Some things are all wrong... was this a full flash or a cal-only flash.

Can you flash it back to original...?

Can you post the tune from the truck prior to flashing, and the tune flashed in.

That PCM might not be compatible.

[Ritch]

I did flash the original back in. I had copied an OS with 4L80 trans, and 4x4 apparently. I'm halfway through converting a 2002 Camaro 5.7 A4 bin. Get back to it tomorrow.

Edit: The Camaro tune cause issues with theft,gear selection and TCC upshift. Now using 2002 truck OS much more successfully

[Ritch]

OK not that successfully

http://forum.efilive.com/showthread.php?t=13664

[joecar]

Warning flashing a truck os12212156 that had a 4l80 will not work well :doh2:Was this PCM previously 4L60E...?

[Ritch]

Joe, I assume it was because 4L80 is in the Vehicle Identifiers under Calibration. I could change the VIN, but the rest stays the same.

[joecar]

Sanity check: you did do a full flash...?

[Ritch]

yep, did the full flash with 02 camaro, 02 truck and 01 Tahoe. The truck and Tahoe ones caused the TFMS to drop to zero when throttle was zero.