[Anaerin]

Looking through local used car listings, I find there are a lot of 4WD trucks going cheap. I was considering a direct drive system with twin/siamese motors (like EVWest sells in a complete kit here), but I'm looking at 4WD systems and thinking I could do one of those. I have some questions, though.

GM trucks use a torque split device, which looks like it outputs to 2 shafts with spicer yokes on them. This would work well, but do the two shafts turn at the same speed? Do they turn in the same direction, or is one flipped?

The kit in question uses 2 controllers (obviously) in a master/slave configuration. Do they require the motors to turn at exactly the same speed, or is there some slip (so it would work like a differential)? How would performance be with two motors joined "through the wheels"? Will the setup support flipping one motor's output, or would I have to effectively wire it in reverse?

[edit]To clarify, I'm suggesting buying 2 motor/controller kits, setting them up in master/slave configuration like they do for their siamese kit, and having 1 motor power each drive shaft/differential (one to the front, one to the back). Though I suppose if the diffs turn the right way, I could get the siamese "joined" motor and drive the front wheels from the tailshaft and the rear wheels from the main shaft, effectively having a rigid connection from front to back. That, however, would lead to issues with transmission wind-up and the like, I think.[/edit]

 

[brian_]

GM trucks use a torque split device, which looks like it outputs to 2 shafts with spicer yokes on them.
...
Do they turn in the same direction, or is one flipped?

A classic 4X4 transfer case turns both front and rear outputs in the same direction - the front output is linked to the main (rear) output by a chain or two pairs of parallel gears, so the direction of rotation is preserved. While not technically necessary, the front and rear output shaft speeds are normally the same, allowing the same final drive ratios (ring-and-pinion gear sets at the axles) to be used front and back.

This would work well, but do the two shafts turn at the same speed?

In a traditional "part-time" system, the front and rear outputs are locked together (in 4WD mode), so they must turn at the same speed... leading to this problem (your thinking is correct):

G... I could get the siamese "joined" motor and drive the front wheels from the tailshaft and the rear wheels from the main shaft, effectively having a rigid connection from front to back. That, however, would lead to issues with transmission wind-up and the like, I think.

In a "permanent 4WD" or "AWD" or "full time 4WD" system (terms vary by manufacturer and model), there is either a differential between front and rear outputs or a controlled-slip clutch (of various types) in one of the outputs (the front output in a typical truck system) to allow the required speed difference for turning corners.

All of this applies equally well to GM and to other major manufacturers.

 

[brian_]

The kit in question uses 2 controllers (obviously) in a master/slave configuration. Do they require the motors to turn at exactly the same speed, or is there some slip (so it would work like a differential)?

...To clarify, I'm suggesting buying 2 motor/controller kits, setting them up in master/slave configuration like they do for their siamese kit, and having 1 motor power each drive shaft/differential (one to the front, one to the back).

The linked kit uses Curtis controllers; typically linked Curtis controllers do not require the motors to run at the same speed, and are capable of handling two motors which are not mechanically connected. In this case, the kit is for motors which are not just connected but built on the same shaft, so the ability to handle separate motors is not being used; the two motors will likely be run by the controllers at matching torque outputs.

How would performance be with two motors joined "through the wheels"?

The Curtis "Dual-Drive" system (and some others) are designed to take in a steering input, so the master controller knows (when properly configured) how the two wheel/motor speeds should be related, and commands torque to each motor (directly by the master and via communication to the slave) to keep the speeds appropriate; if one tire slips and spins significantly, torque to it will be reduced. I believe that this could be readily adapted to a front-rear split and would work well.

Without steering input (and corresponding configuration) the two motors will likely get equal torque, which is far from ideal but not bad for the two driven axles of a 4X4: it's like having an open centre differential. How well it works depends on how the controllers are programmed: hopefully they will reduce torque to avoid overspeeding the motor when a tire spins, but they will not prevent massive spinning. Essentially no vehicle off-roads with a centre differential unless it is lockable, limited-slip, or controlled by brake-based traction control. Even ordinary street-driven AWD vehicles have all had some sort of control like this for many years.

 

[Anaerin]

So the question is, does the Master/Slave Curtis setup try to match speed, or torque (or some combination of both). If it tries to match speed over torque, it would end up like a limited slip differential (AIUI) - both axles would be moving at similar speeds, and the axle with the slipping wheel would get less torque (as it would need less to maintain that speed). If it tries to match torque, it would act as an open differential, allowing the slipping wheel's axle all the speed it can take trying to get the torque up.

 

[brian_]

So the question is, does the Master/Slave Curtis setup try to match speed, or torque (or some combination of both).

When properly set up, it matches speed (or more correctly, sets the speeds of the two motor to the correct ratio for the current radius of turn). Without using Dual-Drive, controllers are generally set up to take a torque command so if both are given the same command the motors will put out approximately the same torque (within whatever other limitations are designed and configured into the controllers).

If it tries to match speed over torque, it would end up like a limited slip differential (AIUI) - both axles would be moving at similar speeds, and the axle with the slipping wheel would get less torque (as it would need less to maintain that speed). If it tries to match torque, it would act as an open differential, allowing the slipping wheel's axle all the speed it can take trying to get the torque up.

I agree, although in the equal-torque scenario the motor driving the tire with traction seems not likely to be limited to the low torque output of the spinning tire, so it could still be much better than a strictly open diff. Letting a tire spin wildly is still bad...