Playing on a wet, sandy, rocky slope — some data

[SpaceEVDriver]

Someone in a different forum asked about what the various modes mean for off-road play. I ran a few experiments on a nearby trail.

This first one was meant to give a baseline of how the two different motors behave in normal driving mode.
Everything before the spike is on a 2-ish mile gravel road. No need for any kind of 4x4 behavior. I leave the truck in normal mode with one pedal drive on.

It takes very little power to get the truck moving and keep it moving on the gravel road at 20 mph. There are a couple of turns, so that’s where you see the little jumps in the HV EV Battery Power graph. It’s difficult to make it out on this graph because of the spike, but the peak draw on the gravel road was about 42 kW. The average was something more like 5-10 kW. For a 20 mile/hour speed, that translates to 20 mph / 5 kW = 4 miles/kWh to 20/10 = 2 miles/kWh. That’s pretty typical for me on gravel roads: 3 miles/kWh is my expectation when I’m on forest service roads, etc.

Okay, now we get to the turn onto the highway. That’s at about 11:01:15 or so on the x-axis. There’s a short stop at 0 kW as I checked for traffic, and then I pulled onto the Rt. 66. There was traffic coming up behind me, so I didn’t stop on the highway. I was going about 30 kph (19 mph) when I punched it. This drew about 475 kW of power and gave me an acceleration of about 6.5 m/s^2 (about 0.66 Gs), sending my speed up to 110 kph (68 mph) in about 2.5 seconds. Then I let off the accelerator and regen took over and fed energy back to the battery. Shortly after, I turned off the highway onto another dirt road. The graph ends there.
Note the 475 kW of power from the battery translates to about 637 HP, but that’s at the battery. There are some losses as well as some multipliers to get to the brake HP most people think of when they think of HP.

Comparing the Primary and Secondary motors:
Commanded torque is just how much torque has the computer told the motors to put out.
Let’s compare the torque from the primary and secondary motors:
They’re nearly identical in shape and in magnitude. Mostly where they differ is when the vehicle is turning. I overlaid the graphs here so you can see just how close to the same they really are. The max torque commanded was about 650 N-m, to each motor. That’s about 1300 N-m total (about 960 ft-lbs), but again, that’s at the electric side of the motor. There’s a step-down gear in the motor, and you have to multiply by that ratio, then you have to do the multiplication by the tire radius, etc.

Okay, moving on to comparing the motor speeds:
These match very closely as well. Why does the rear motor turn ever-so-slightly more slowly at times? I think it’s to do with when I’m slowing down slightly. I believe there’s a slight bias toward higher regen on the front tires than on the rear. But I’m not 100% certain of that.

More to come in the next several posts.

Sponsored

 

[SpaceEVDriver]

For this next post, I’ll talk about Off-road mode.
A reminder: Off-road mode turns off one-pedal drive, disables traction control, and engages the rear locking differential.

For those with color vision issues, I make my apologies: I don’t know how to tell CarScanner to plot with different color lines. In the following plot, the primary (rear) motor torque in this plot stands above the secondary (front) motor torque whenever there’s a separation between the lines, except in very few cases.

Torque:
This plot shows that in Off-road mode, the front and rear motors are commanded to have different torques (about 58% rear to 42% front), depending on what’s going on. When the torque is less than 0 N-m, it’s regenerating power. And you can see here, most of the time both motors are contributing about equally to that regen.

When climbing the hill, the rear is commanded to push more than the front is commanded to pull. This is a pretty standard profile for most 4x4 / AWD vehicles. A common split is (something like) 60% rear, 40% front. Different transfer cases are designed differently for different purposes.

View attachment 104893



Motor speed looks different. The plots are much closer together and in only a few instances are they wildly different. This is what you would expect for most vehicles. There are a couple of times the front tires slipped and you can see those clearly where the secondary motor speed spikes while the primary motor speed does not.

The biggest instance of a non-slip deviation is just before 11:20:10 (when everything goes flat because the truck was put into park at the top of the hill). You can see the plots diverge, again with the front motor speed higher than the rear motor speed. But without the spike that is characteristic of slippage you might usually think of when climbing a hill. This occurred while I was turning the vehicle around on the trail. Because the rear differential was locked, the inside rear tire controlled the rear motor speed. That tire dragged because of the locked differential, and the rear motor had a lower RPM.


View attachment 104894



So why is torque different when the speeds are the same? Because the computer can detect wheel slippage >1000 times per second and can modulate the commanded torque in an attempt to keep the wheel speeds the same. Sometimes it fails and the tire slips, usually because there’s literally no traction to be had. Those two spikes in the graph are when one of the front tires stopped touching the ground entirely. Remember that the front axle is an open differential, which means that whichever tire has the least traction spins more than the other. And because there’s no traction control engaged, the only thing to stop those tires from spinning is the ABS or my manual braking. I didn’t brake, but ABS did apply and that’s why the spike was very short. You’ll also notice that the commanded torques did drop for the front motor when the wheel spun.

 

[SpaceEVDriver]

Sport mode, no differential lock, no one-pedal drive, no traction control.

Torque: Again, the rear motor stands above the front motor for most of the drive (more torque was commanded to the rear motor).
The shapes are similar, but the values are different. This mode sends more torque to the rear motor—about 30% more than in off-road mode, from my understanding.

Motor speed: Here, it was the rear tires that spun. The front tires were providing more of the traction pulling the truck up the hill, but only in minor instances. That rear tire spin is because of the extra torque being sent to the rear tires and no traction control reducing slip. This is not the mode you want to be climbing hills in.

 

[SpaceEVDriver]

This is the Normal mode, no differential lock, but with traction control and one pedal drive on.

Torque:

For the most part, the rear motor is being commanded with more torque, but not at the same ratio as with sport mode nor off-road mode. These are almost at a 50:50 torque ratio, probably something like 52% rear, 48% front. I could do the math, but precision here is not a big deal.
Note however, there are a few instances where there’s a larger deviation. Just after 11:14:14 (a couple of times), there’s a huge spike in torque on the rear motor compared with the front. And again at 11:14:54.

If you look at the motor speed, you can see that there was slippage of the front tires just after 11:14:14 and that’s why there was more torque sent to the rear motor: to overcome that slippage.

But that spike at 11:14:54 is different. Let’s look at the motor speeds to understand that one.

Motor speed:
For the most part, the front and rear motors have the same speeds. But look at that jagged blip at about 11:14:54. The rear motor not only dropped speed dramatically, it went in reverse. But the front motor did not. What the heck is going on there?!

This is where traction control comes into play. On a conventional ICE vehicle, the tires have to all turn in the same direction. There’s no way to decouple the front from the rear without some fancy differential/transfer case hardware linkages. It’s just not an option in normal driving. But in an EV, the motors do not have to be commanded to turn in the same direction. And when there’s a reason to command the rear to turn in a different direction from the front, the EV can be programmed to do that.

Some kind of slippage happened, but not just wheel slippage. The entire truck slipped down a deep rut in the trail here and the truck tried to recover control by reversing the rear motor while still allowing the front motor to turn forward. This allowed it to regain yaw control since the vehicle was not only slipping forward, it was yawing as well. As far as the sensors were concerned, the truck was out of control in yaw and the way to recover it was to reverse the rear motor and turn the front motor forward.

Fascinating!

 

[SpaceEVDriver]

Summary:
Without an ICE-like transfer case with a set torque ratio between front and rear axles, a dual (or more) motor EV can do some incredibly fun things with the power ratios.

The traction control behavior is also fascinating when the vehicle slips in certain ways (yaw, for example), the vehicle can use novel approaches to correct it. I’ve felt this on ice as well, but I didn’t have the OBD2 & CarScanner running at the time.

 

[Timeless Epoch]

Okay, moving on to comparing the motor speeds:
These match very closely as well. Why does the rear motor turn ever-so-slightly more slowly at times? I think it’s to do with when I’m slowing down slightly. I believe there’s a slight bias toward higher regen on the front tires than on the rear. But I’m not 100% certain of that.

I’ve read in a few places that the rear motors is geared a little taller; relying more on the front motor for efficiency at speed and the rear motor for more torque when needed.

What is weird is, that would have the opposite effect of your observation.

Very cool analysis, BTW!

Sponsored