• Hello Guest, welcome to the initial stages of our new platform!
    You can find some additional information about where we are in the process of migrating the board and setting up our new software here

    Thank you for being a part of our community!

AW30-43 torque management between shifts?

T

Turbosundance

New member
Joined
Oct 11, 2007
Location
Toronto
I want to go through the transmission in my '98 S90 and shim the accumulators and see if there is a way to up the line pressure.

I know a number of other manufacturers use something that often called Torque Management to drop engine torque while the transmission is shifting to soften the shift and improve transmission life.

Does anyone know if this practice is done with the 9060/S90/V90 cars? The shifts are very smooth and it almost feels like the engine output is being reduced during shifts.

Chrysler cars in particular use the can bus communication between the ECU and TCU on the older cars to tell the ecu to pull ignition timing while the TCU is commanding a shift.

I'm concerned that if I speed up the shift timing in the transmission it will throw off some carefully choreographed shift setup between the ECU and TCU.
 
Well that kind of sucks. It look like I can't simply cut the torque limiting wire and shim the accumulators. Do you have any idea how the TCU arrived a value for how long to command the ECU to limit torque? Is the torque limiting signal based off the actual shift time or is just a look up value based on how long the TCU "thinks" the shift will take? If it's based on actual shift time than a quicker shift will make for an equally shortened torque limiting time.

I was considering purchasing one of these kits: http://www.ebay.ca/itm/A340E-A340F-...1-A343-TOYOTA-JEEP-solenoid-tcc-/330719393950
 
Last edited:
I plan to play with the same thing. What I want to do is to cut the request lines from TCM->ECU, and set the ECU->TCM acknowledge to be permanently set. Or, if the permanently set acknowledge will be problem, I will use the request signal as a source for the acknowledge signal - so the request will be immediately acknowledged.
If you would have some progress, put the info here ;)
 
I dont mind the torque management being active since it will increase the lifespan of the transmission. I mainly want to firm up the shifts and shorten shift time. Im going to buy one of the commonly available shift kits for the a341e/a343e and install it and see what happens. People seem to have no problems with the trans computer when people install them into the lexus version.
 
Played with the logical analyzator today. Torque reduction takes usually between 100 and 250ms (quite a long time IMHO), but there were some extremely long shifts where the reduction was up to 0.7sec - the second screenshot.
I was even able to feel the reduced power output in the car acceleration behavior.
Going to remove this reduction logic in following weekend :nod:

full_seq.png


long_shift.png
 
tomasss said:
Played with the logical analyzator today. Torque reduction takes usually between 100 and 250ms (quite a long time IMHO), but there were some extremely long shifts where the reduction was up to 0.7sec - the second screenshot.
I was even able to feel the reduced power output in the car acceleration behavior.
Going to remove this reduction logic in following weekend :nod:

full_seq.png


long_shift.png
Click to expand...

Greetings from Finland!

This is super interesting topic, cannot wait for the result!

I'm building a 960 TDI (D5252T engine from 850) and I'm using AW30 transmission. I'm using slightly modified D5/AW50-55 torque converter, original D5252T "flywheel", modified torque converter housing and original TCU.

MSA 15.7 has electronic gas pedal which gives 0-5V signal and AMM. The problem is that MSA15.7 engine management doesn't have input pins for torque limiting signals nor output pins for feedback back to TCU. If this case will be solved somehow, my project will be a success! In addition, it basically means that original TCU can be used with every engine management system which has electronic throttle position sensor and has load signal from AMM or similar!
 
Last edited:
Well, the torque reduction exists no more in my car :cool:
In fact it was pretty straightforward:

- there are two outputs from TCM to request the torque reduction
- these signals goes to ECU, torque is reduced and acknowledge is generated
- the acknowledge goes back to TCM and based on it the gear is changed
- when shift is completed, the TCM ends the request, ECU ends the acknowledge and reverts the torque back

To remove this feature, it is needed to keep the ECU inputs in inactive state (which is +5V) which means no torque reduction; but as the TCM has some backward checks between REQ and ACK, it is not possible to simply cut the wires. So, based on this thread:

http://forum.savarturbo.se/viewtopic.php?t=36768

I have done what is shown on the picture below:

case_scheme.GIF


The 74AHC1G08DBV is simple 5V AND-gate with A and B inputs and Y output (Boolean function Y = A and B). It can be done also with simple transistors, but I wanted to try this way. So, whenever there is a TC1 or TC2 request, the acknowledge is asserted immediately (7ns the datasheet says) and the gear is changed, but with unrestricted engine torque :oogle:

The "magic box" looks physically like this (the AND gate is from the other side so it is not visible)

case_inside.jpg


and mounted on the TCM like this

case_mounted.jpg


The difference is pretty noticeable, the larger the throttle opening the bigger the difference. With the throttle opened just a bit I would say the smoothness of the shifts is the same, but WOT means really nice quick shift without the power decrease just before the shift happens.

Please note that this solution may be used for the Motronic 1.8 ONLY as it uses 5V logic. The later Motronic 4.3 or 4.4 uses 12V logic, so the "AND" function has to operate in different voltage level.
 
Last edited:
blkaplan said:
wonder if something like this should be used if manual swapping an auto 1.8 car to manual?
Click to expand...
Swapping to manual means the TCM may be removed(?), so the only thing you should ensure is to connect the torque limiting inputs of the ECU to +5V (Motronic 1.8) or +12V (Motronic 4.4).
 
Can you just connect 12v to the correct line? It seems like there are some "Checks" as you stated above?

tomasss said:
To remove this feature, it is needed to keep the ECU inputs in inactive state (which is +5V) which means no torque reduction; but as the TCM has some backward checks between REQ and ACK, it is not possible to simply cut the wires. So, based on this thread:
Click to expand...
 
The checks are in TCM, not in ECU. So if you swap to manual and remove the TCM, you have no problem with the checks (but you should power the ECU input not to limit torque all the time).
 
tomasss said:
The checks are in TCM, not in ECU. So if you swap to manual and remove the TCM, you have no problem with the checks (but you should power the ECU input not to limit torque all the time).
Click to expand...

wheres the best place to get a 5v signal?
 
tomasss said:
Check my schematic above, I've used the 7805 stabilizer connected to switched +12V.
Click to expand...

I see, Thanks!!!

I sent this thread to Pete, he has a 960 that he converted to m90 but never changed the ECU. I suspect his car is in torque limited mode. It feels slow to me.
 
You must log in or register to reply here.
Back
Top