Why such low torque?

As it makes a significant difference to available torque/power whether you are driving forward or reversing, it was worth finally giving your response in post #16 to the question Tombukt2 first asked you in post #2.

 

There was no first time you answered that question before post #16. Notwithstanding your "as I explained" effort to muddy the history of the thread:

You had not explained that you "backed into the slot" until post #14, and even in that post you still did not say whether you were trying to drive forward or reverse from that position later, after having been asked that specific question numerous times.

While "backed up to a shop door" would clearly imply driving away forward if it was a closed door, it's equally clear that a door you could back through far enough for your rear wheels to settle into the door track probably wasn't closed. So whether your next move was forward or reverse from that point was still never answered until post #16.

 

Anybody who can read the posts in this thread can verify both my reading comprehension and your misdirection.

 

Pushing the throttle all the way to the floor would allow enough power to the front wheels
to allow it to get out of the ditch....

 

In the closest I could easily get to "pull rear wheels up 1 inch deep gutter edge", I had two large solid, square-edged blocks that measure 1 inch thick. They are the same hard, solid nylon blocks seen in this old pic:



I chocked those just ahead of the rear tires, on hard pavement, and watched the MG2 torque as I slowly applied the go pedal until the car just drove over them, and then again in reverse to back over them to the starting position. I did both a few times and the readings were fairly consistent. The engine was stopped the whole time, so only MG2 torque was involved.

It took MG2 torque in the mid-seventies (newton meters) to get over the 1" blocks going forward, and a smidge less (closer to 70 Nm) going backward. The difference may reflect that the blocks are about 10" farther aft when backing over them, so farther from the car's center of gravity.



With gen 3 gearing, 76.5 Nm torque from MG2 is 202 Nm at the PSD ring, or 659 Nm at the differential. Using 314 mm as my tire effective radius, that's a forward thrust of 2099 newtons, or 472 pounds.

These numbers are lower than what I had estimated on paper (923 pounds thrust, or 150 Nm MG2 torque), but then for my paper estimate I had assumed the Prius loaded to its maximum rear Gross Axle Weight Rating, and my Prius at the moment is surely not.



I don't have an easy way to know my current rear axle weight, but I could easily believe it's closer to half the rear GAWR (the back is pretty empty, but for my tools and stuff that's always there), which would make these numbers roughly in agreement with my theoretical ones.

The gen 3 MG2 is rated for 207 Nm torque, so this test with my not-fully-laden Prius felt like no sweat at 76 Nm. Fully laden to rear GAWR, the needed 150 Nm would be closer to the 207 Nm limit but still comfortably below.

Is there any chance the depth of the gutter was really more than one inch? That can make a large difference in the numbers (assuming the width was also more than "a few inches", enough for the tire to hit bottom).


Edit: I cadged that door-plate image from somebody else's PriusChat post, so the VIN on it isn't mine

 

Repeated the exercise watching a couple different Live Data items instead of the MG2 torque.

"Accelerator degree" (that's what it's called in the Autel AP200 app) isn't an angle in degrees as you might think, it's %. Mine does read 0% when I'm off the go pedal, and, like, 99.5% when I floor it.

I had to go around 17.5% to pull the rear tires over the 1-inch blocks.

"Power resource Ib", the battery current, went to about 5 or 5.3 amps just before going over the blocks. This with "Power resource Vb" around 212 V, so say 1.06 to 1.12 kilowatts.

That wasn't all going to MG2; the traction battery was also powering the DC/DC converter and thus all the 12V car stuff, which I usually consider around half a kW. So, the remaining half-kW or a bit more going to MG2.

This is a special circumstance for a motor, essentially locked-rotor condition. It is making torque but not able to turn (until finally the torque is enough to climb the blocks). Because it is not turning, there is no back-EMF, and the impedance of the windings pretty much boils down to their DC resistance.

The ORNL report on 2010 Prius gives the MG2 winding resistance as 77 mΩ, phase to neutral (Table 2.7, p. 46). The Prius inverter has no connection to the wye-wired winding neutral, only to the three phase connections, so current must always pass through one phase winding and some combination of the other two, a path whose resistance seems bounded below by 77 + (77 ∥ 77) or 116 mΩ, and above by 77 + 77 or 154 mΩ.

Half a kilowatt through a resistance in that range implies (P = I²R) a motor current of 57 to 66 amps at that moment.

Considering how casually I did this analysis (oh, half a kW here, half a kW there!), I'm pretty pleased that on ORNL's locked-rotor MG2 torque plot (p. 60), looking between the 50 A and 75 A curves, you're pretty much in the 70 to 75 Nm torque ballpark fitting the readings I got in post #28.

 

In any case, posing the question presents a learning opportunity for all readers interested in the capabilities and limits of the car.

I suppose it's possible, since you weren't planning to post about it, there may be some margin for error in recalling the dimensions of the gutter you dropped into.

I used exactly 1-inch-high blocks. Because of the geometries in play, the needed force goes up pretty swiftly if the gutter may have been deeper than that. Also, how heavily the back of the car was loaded makes a (linear) difference.

 

Yep, life is full of situations like that. It was some time ago, and whatever details didn't get measured or datalogged at the time are details we don't have now, and so there may be aspects of that particular past situation that never get fully explained.

At the same time, posting about it is useful as a springboard for discussion on PriusChat that can lead to insights into the forces and limits involved in that kind of situation. Several of the above details about the power flow in low-speed or locked-rotor conditions, for example, I had never had a reason to sit down and work through before. It'll simplify thinking about future questions of the same kind. So in that way it's an interesting thread even if a full explanation of that past experience doesn't emerge.

 

When you’ve dropped a pair of wheels down into a trough, to get going you’re taking on gravity head-on, dead-lifting, with precious little ramp effect. Guess that’s why wheel chocks are so effective.

 

The height of the vertical-faced obstacle (front edge of the trough, or a block) makes a difference.

You're taking on gravity (rear-axle weight) ✕ cos(θ) where θ is the angle to where your tire contacts the obstacle. If you take the angle 0° to be straight forward from the axle, and -90° to be straight down from the axle, a 1" obstacle hits at about -67° (for a 315-ish mm radius tire) and so cos(-67°) or about two fifths of the axle weight bears against you. A 2" obstacle hits at -57°and more than half the axle weight is in play, and so on. In the extreme case, a 12-inch obstacle would hit near zero degrees and to turn that wheel is to try to dead-lift the back of the car.

It gets worse because the rear wheels aren't driving wheels. The only way you can apply torque to them is by pulling the car forward ... against the obstacle ... again making contact at angle θ, which again acts against you even more strongly as the obstacle height increases. Only |sin(θ)| of your forward thrust goes toward turning that rear wheel. So the forward thrust needed to pull you over has sin(θ) in the denominator. As sin(θ) approaches zero (obstacle gets closer to mid-height of wheel) there comes to be no force whatsoever that will pull you over it.


If I haven't done this wrong, here should be two curves showing the forward thrust needed (in pounds) to pull the rear tires over different-height vertical-faced obstacles (height in inches). The steeper one is for the car loaded to the full rear GAWR (2175 pounds), and the other one is for a car loaded like mine with not much in the back right now (something like 1100 pounds on the rear axle).

Loaded to full rear GAWR, it should take no more than a 1.7" obstacle to require more thrust than MG2 can supply.

Loaded more lightly, like my car right now, it looks like I might be able to manage 4¼".

I used gearing and specs for gen 3. Similar plots for other gens ought to look a lot like this, but the numbers won't be the same.

 

When I pulled up to the 1” lip of my garage floor going in, it would take a good goose of the accelerator to get over it unless I hit it running.
Usually enough to fire the engine

 

Another fun thing to do with these results might be to flip them around and use 'em as a poor-man's way to estimate the loading of the car, by watching the MG2 Torque reading while slowly pulling over blocks of a known height.