Increase torque split in auto??

[TomsSVX]

Just a quick idea... would one be able to fit a resistor inline to increase the torque split but not induce binding?? I understand its a signal not a voltage reading but what do ya think??

Tom

[Manarius]

Originally Posted by TomsSVX

Just a quick idea... would one be able to fit a resistor inline to increase the torque split but not induce binding?? I understand its a signal not a voltage reading but what do ya think??

Tom

As in set the solenoid C from 0% duty to 100% duty? I know of guys that have done it with a switch, but not with some sort of resistor.

You're going to get binding if you try to put in too much torque without enough slip.

[Nomake Wan]

No... removing power from the solenoid causes it to go to 0% duty, which fully engages the clutch, causing 50/50 split.

Full signal (not power... signal) causes the solenoid to go to 95% duty (there is no 100%) and thus give you 100/0. This is when you put in the FWD fuse.

What Tom is suggesting is a way to make the car start off with something other than 90/10 (if that is indeed what the base split is; I intend to dig deeper about that). For instance, starting with 70/30.

To answer your question... since it is a signal, I'd think you'd need a way to create your own signal instead by creating a circuit designed specifically to translate the signal based upon a ramp. So that you still get 50/50 when need be, but that any signal which goes past a certain duty level is sent as the duty level for your specified rating instead. I understand the concept at hand, but don't look it me when it comes to designing the actual circuit. I don't know anything about that sort of thing. Someone with more experience will have to chime in and tell you what would be required.

Of course we'd also have to know exactly how Solenoid C and the TCU operate together before messing with it...

[Trevor]

Originally Posted by TomsSVX

Just a quick idea... would one be able to fit a resistor inline to increase the torque split but not induce binding?? I understand its a signal not a voltage reading but what do ya think??

Tom

Tom, I do not have data on the US system incorporating a clutch and have read confusing comments. Is the clutch engaged or is it released, when fluid pressure is applied.

When I can be absolutely sure in this regard, I will comment further.

Thanks, Trevor.

[Nomake Wan]

95% duty cycle of Solenoid C results in the clutch being 100% disengaged. Though I don't know what that means, perhaps you do.

[crazyhorse]

Couldn't a variable pulse wave generator make this happen?

If so, I think I can lay hands on one in an industrial relay form.

[Trevor]

Originally Posted by Nomake Wan

95% duty cycle of Solenoid C results in the clutch being 100% disengaged. Though I don't know what that means, perhaps you do.

The "C" solenoid valve in 4AUT transmissions is normally closed. The valve is shut when the solenoid is not energised and opens to a drain when energised, thus reducing the fluid pressure delivered to the clutch.

Therefore if a 95% duty cycle of Solenoid C results in the clutch being 100% disengaged, the solenoid is open and there is low pressure delivered to the clutch when it is open and pressure would close the clutch.

However this must be absolutely confirmed before any suggestions can be examined. I hope Tom has hands on knowledge of the mechanics involved.

P.S. If the above is correct, Tom's idea of a resistor would appear feasible and worthy of experiment.

[Crazy_pilot]

So the idea is to have the car operating at a standard torque split with more rear bias than it does now, right? Obviously we can't exceed 50:50, but more rear than I have now would be awesome. After the first few winter days in the SVX, I can say that it definitely doesn't swing the tail as easily as the Foresters when I step on it in a turn. It also doesn't do donuts as tightly.

[b3lha]

I don't think a resistor will do the trick. The solenoid will still pulse open and closed at the same rate.

I think you need a circuit like this to generate your own duty cycle. Then you need a switch to select whether the transmission uses the TCU's duty cycle or the one from your circuit.

http://www.doctronics.co.uk/555.htm#more%20astables

You could replace R1 and R2 with a single potentiometer which you would turn to vary the duty cycle.

(Please check this with an electronics expert before you try it out. I've repeatedly proved in my memory dump thread that I don't know much about electronics.)

The ultimate solution is to reprogram the TCU. But there is a lot of reverse-engineering to be done before that becomes possible.

[SilverSpear]

Originally Posted by b3lha

I don't think a resistor will do the trick. The solenoid will still pulse open and closed at the same rate.

I think you need a circuit like this to generate your own duty cycle. Then you need a switch to select whether the transmission uses the TCU's duty cycle or the one from your circuit.

http://www.doctronics.co.uk/555.htm#more%20astables

You could replace R1 and R2 with a single potentiometer which you would turn to vary the duty cycle.

(Please check this with an electronics expert before you try it out. I've repeatedly proved in my memory dump thread that I don't know much about electronics.)

The ultimate solution is to reprogram the TCU. But there is a lot of reverse-engineering to be done before that becomes possible.

That is the most confusing reply I ever read

[Trevor]

Originally Posted by b3lha

I don't think a resistor will do the trick. The solenoid will still pulse open and closed at the same rate.
.

Yes but, a resistor could limit the opening of the solenoid such that it did not open fully for the complete duration of each voltage pulse. Granted, this is a handyman's idea which will require proving.

Most certainly it is a crude approach and reliable calibration of any control arrangement would be unlikely. There are other ways of reducing the on time, but the low impedance of the solenoid increases the difficulty.

In all of this, the prospect of increased clutch wear will have to be taken into account.

[b3lha]

Originally Posted by SilverSpear

That is the most confusing reply I ever read

Sorry to confuse you Danny. Let me explain. What I am proposing is a knob that the driver can turn to vary the torque split. Something like the DCCD (Driver Controlled Centre Differential) on some STi models.

The TCU controls the torque split using a duty cycle. It pulses the solenoid on and off. A 95% duty cycle, for example, means that the power is on 95% of the time and off for 5% of the time. That is why I am saying a resistor might not work - the power will still be on 95% of the time but at a reduced voltage.

The circuit I posted is a duty cycle generator. The duty cycle is selected by the two resistors R1 and R2. Replacing them with a potentiometer would allow the driver to vary the duty cycle by turning the knob. A two pole switch would be fitted to connect the solenoid either to the TCU or to the duty cycle generator.

My final point was an alternative solution. That the TCU program could be modded to produce a different duty cycle. For example, to change the function of the manual button such that it changes the torque split rather than its current function of locking out first gear. This will become possible once somebody reverse-engineers the program in the TCU. It's on my website waiting for anybody who wants to look at it. I need to finish looking at the ECU before I start on the TCU.

[b3lha]

Originally Posted by Trevor

Yes but, a resistor could limit the opening of the solenoid such that it did not open fully for the complete duration of each voltage pulse. Granted, this is a handyman's idea which will require proving.

Most certainly it is a crude approach and reliable calibration of any control arrangement would be unlikely. There are other ways of reducing the on time, but the low impedance of the solenoid increases the difficulty.

In all of this, the prospect of increased clutch wear will have to be taken into account.

Forgive me for asking this Trevor, but it was my understanding from your discussions with Harvey that a solenoid could not be made to partially open. Have I misunderstood?

[SVXRide]

Originally Posted by b3lha

Forgive me for asking this Trevor, but it was my understanding from your discussions with Harvey that a solenoid could not be made to partially open. Have I misunderstood?

Phil,
as a mechanical engineer that has worked with plenty of electrical engineers over the years, that is my understanding of solenoids too (i.e., they're either all the way open, or all the way closed).
-Bill

[Tuck]

Did I not read somewhere here that the JDM transmission and TCU provided an initial 35-65 torque split? If so what makes them different?

Who needs throttle steering?