Subaru diff control

[RDLperformance]

Hi all, I run the electronic center diff control off the g4x, 6 speed sti subaru box, via a motec half bridge.

Currently i can only have one map as per pic and just use the 0 app line for braking lock. i would like to run a seperate table for braking activated by the stop light switch to account for left foot braking. I have tryed a couple of options with a second table GP PWM2. It operates on that table but wont transfer that to the first table when i set gp pwm2 as the axis.

 

Anyone got a better plan????

 

[Vaughan]

I haven't tested this but something like the below should give you the ability to use one table or the other depending on whether or not the brake is active. You could also simplify your setup if you used a formula instead of a table or if you want to use one table and modify it with a math block when the brake was active.

Your first table looks like it is just more 4wd with more speed or aps, what are you wanting to do with the torque split under braking? I'm interested as I plan to control the torque split in my 4wd Stagea in the future.

[kaptainballistik]

Any other motec side Diff controller inputs or a driver/controller used? ( Because this REALLY interests me, I was discussing it with my tuner.....!) 

So are you directly supplying the diff with a  PWM output it then controls the diff with? (Its a lot of power, the ECU wouldn't be able to do it directly.. Well I didnt think it could but that MoTec 1/2 bridge thing....imteresting)  , or its supplying the Analogue inputs to a controller and it converts to PWM??

I Would say you still want some lockup on the center diff on braking, and add a knob to allow multiple maps

[Adamw]

5 hours ago, Vaughan said:

what are you wanting to do with the torque split under braking?

In rally cars you generally want much less lock in the diff during braking as you want to be able to semi-lock the rear wheels without affecting the front wheels, typically same for hand brake also.  In road race cars you typically dont need to change the torque split during braking unless it is messing with your brake bias.   

 

4 hours ago, kaptainballistik said:

So are you directly supplying the diff with a  PWM output it then controls the diff with? (Its a lot of power, the ECU wouldn't be able to do it directly.. Well I didnt think it could but that MoTec 1/2 bridge thing....imteresting)  , or its supplying the Analogue inputs to a controller and it converts to PWM??

The DHB is basically a solid state relay, so the ecu is sending a low current PWM to the DHB, the DHB is connected directly to the diff, just like how you would control a fuel pump etc. 

 

6 hours ago, RDLperformance said:

I have tryed a couple of options with a second table GP PWM2. It operates on that table but wont transfer that to the first table when i set gp pwm2 as the axis.

Vaughan's option above is a nice way to do it.  But your original plan will work too - you just need to put the output DC parameter from the 2nd table on the axis.  For example if GP PWM2 table is assigned to CAN aux 1, then you put CAN aux 1 on your 1st table axis, not GP PWM2.  

Example below pressing brake:

 

 

[kaptainballistik]

So if have this right, you could also add into the GP PWM config as one of the parameters to Turn off the PWM when a switch is engaged (Handbrake..) OR  lock the diff When Launch control is active AND handbrake is on  ( Switch logic conditions)...

I presume the PWM output from the Link is a Square wave? ( I think the resident electrical engineer has a project to build a driver!  )

[Adamw]

14 hours ago, kaptainballistik said:

So if have this right, you could also add into the GP PWM config as one of the parameters to Turn off the PWM when a switch is engaged (Handbrake..) OR  lock the diff When Launch control is active AND handbrake is on  ( Switch logic conditions)...

You would have to something like the above examples to lock the diff for launch as you would need 100% DC.   The unlock for handbrake slide could just be done via the conditions on the GP PWM. 

You dont need to reinvent the wheel, a common solidstate relay like the Hella H41773001 or the Holley 15620 would drive it fine, Im pretty sure they only pull about 5A.  

Note, the good diff controllers dont just use a fixed PWM, they do a closed loop current control which gives more consistent torque split as the diff temperature varies.  We are developing an accessory that will be capable of doing constant current via CAN from the ECU this but it is a little while off yet.  I was hoping to get some sort of proper diff control strategy in G4X when we have that device closer.  

[RDLperformance]

18 hours ago, Adamw said:

In rally cars you generally want much less lock in the diff during braking as you want to be able to semi-lock the rear wheels without affecting the front wheels, typically same for hand brake also.  In road race cars you typically dont need to change the torque split during braking unless it is messing with your brake bias.   

 

The DHB is basically a solid state relay, so the ecu is sending a low current PWM to the DHB, the DHB is connected directly to the diff, just like how you would control a fuel pump etc. 

 

Vaughan's option above is a nice way to do it.  But your original plan will work too - you just need to put the output DC parameter from the 2nd table on the axis.  For example if GP PWM2 table is assigned to CAN aux 1, then you put CAN aux 1 on your 1st table axis, not GP PWM2.  

Example below pressing brake:

 

 

Perfect! ill give both options a try next week, booya!

1 hour ago, Adamw said:

You would have to something like the above examples to lock the diff for launch as you would need 100% DC.   The unlock for handbrake slide could just be done via the conditions on the GP PWM. 

You dont need to reinvent the wheel, a common solidstate relay like the Hella H41773001 or the Holley 15620 would drive it fine, Im pretty sure they only pull about 5A.  

Note, the good diff controllers dont just use a fixed PWM, they do a closed loop current control which gives more consistent torque split as the diff temperature varies.  We are developing an accessory that will be capable of doing constant current via CAN from the ECU this but it is a little while off yet.  I was hoping to get some sort of proper diff control strategy in G4X when we have that device closer.  

Interestingly,most other solid state relays dont last, hence we have gone to motec.... expensive but doesnt stop mid rally;). 

I assume you are talking about a solid state relay like the Kaizan? i am pretty keen to try it for diff control over CAN....

Ps, yes please to the diff strategies!!!!

[Adamw]

8 minutes ago, RDLperformance said:

I assume you are talking about a solid state relay like the Kaizan? i am pretty keen to try it for diff control over CAN....

No, Kaizan doesnt have any current monitoring so cant do constant current control, it would not work any better than any other SSR.   

[RDLperformance]

6 minutes ago, Adamw said:

No, Kaizan doesnt have any current monitoring so cant do constant current control, it would not work any better than any other SSR.   

Gotcha, Actually failed to read your whole comment,duh. A pdm (obviously more expensive) that does PWM would be able to set up with that type of strategy. Interesting... I might have to experiment...

[kaptainballistik]

11 hours ago, Adamw said:

 

You dont need to reinvent the wheel, a common solidstate relay like the Hella H41773001 or the Holley 15620 would drive it fine, Im pretty sure they only pull about 5A.  

 

Ill stick to my transistors  

You have me interested on the temperature correction aspect .. I'll get back to you on that....

 

[kaptainballistik]

I've really gone down the rabbit hole on this one as I was discussing using the Link for DCCD with my tuner before I got it! ( It was in the Get the damn thing running first basket... )

Is there a preferred CanBus Accelerometer to use with the Link? (Personal Preference on this one, I didn't like the Third party DCCD controller without it). I would attempt using the Factory Subaru RA/R unit but I'm running out of inputs! 

[Adamw]

The G4X plug-ins have a 3 axis accelerometer already built-in.  >Analog inputs>Accelerometer

[kaptainballistik]

3 hours ago, Adamw said:

The G4X plug-ins have a 3 axis accelerometer already built-in.  >Analog inputs>Accelerometer

OHHHH, This just gets better  Any particular orientation for the ECU? (Currently on my work laptop and they haven't approved the Link Software as a Business app so I cant check if its in the help file )

 

[Vaughan]

29 minutes ago, kaptainballistik said:

OHHHH, This just gets better  Any particular orientation for the ECU? (Currently on my work laptop and they haven't approved the Link Software as a Business app so I cant check if its in the help file )

Any orientation, you will need to calibrate it to first find level and then forward/backwards

[kaptainballistik]

Excellent!