Investor Engineering Discussions

[Oil4AsphaultOnly]

Yah, but regen is running the motor in reverse and you can rely on a (properly working) powertrain and cooling system. That's where you lost me on the quoted section.

Actually regen is NOT running the motor in reverse, it's switching the direction of current flow (it is reversing the FUNCTION of the motor though).

Running the motor in reverse meant putting the vehicle into reverse, which was one of the solutions that would NOT work (due to excessive heat generated), still has current flowing from battery to inverter to motor. It was a theoretical way to consume some of the charge in the battery WHILE also slowing the vehicle down, which we now know it could not.

Edited for clarity.

 

[Oil4AsphaultOnly]

Like I said avoid chargers at the top of a grade. I think a more likely scenario is a cold battery where someone parks overnight at the top and then heads down the grade in the morning.

Does anyone know how much energy the engine brakes actually absorb on a diesel. It is the majority of braking on a downhill or just a small % to save the normal brakes. Visually they seem to just make a lot of noise and not really stop the truck.

If it's a cold pack issue, that's easily solved by the same mechanism for the model 3/Y, run the motor against itself to generate heat to warm up the battery pack, so it's really a trivial problem to solve. Once the pack is warm enough, full regen would be available. I think this is where dhanson865's suggestion of putting the vehicle in reverse (wording it funny like this to distinguish this from regen) came from.

 

[MP3Mike]

It was a theoretical way to consume some of the charge in the battery WHILE also slowing the vehicle down, which we now know it could not.

Why can it not? 3 motors can apply ~1200kW of force to go up a hill why couldn't they do that to slow down? Obviously time is limited due to heat rejection, but Plaid Model S seems to do pretty good at that.

 

[Oil4AsphaultOnly]

Why can it not? 3 motors can apply ~1200kW of force to go up a hill why couldn't they do that to slow down? Obviously time is limited due to heat rejection, but Plaid Model S seems to do pretty good at that.

the 3 motors can apply 1200kw of force into going uphill, but being 97% efficient, it's only generating 36kW's of heat. The rest of the power went into raising the potential energy of the semi truck. Going downhill, you're converting potential energy to kinetic energy + heat. If you don't want that kinetic energy to get out of hand, it needs to be converted to heat, ergo melt your motors/powertrain.

 

[MP3Mike]

the 3 motors can apply 1200kw of force into going uphill, but being 97% efficient, it's only generating 36kW's of heat. The rest of the power went into raising the potential energy of the semi truck. Going downhill, you're converting potential energy to kinetic energy + heat. If you don't want that kinetic energy to get out of hand, it needs to be converted to heat, ergo melt your motors/powertrain.

So, you can't supply 1200kW to provide reverse force, still only generating a small percentage of that as heat? Why does it work differently applying reverse torque?

You can even do things like regen ~220kW/each on the acceleration motors, apply 400kW of reverse torque on the cruise motor, and use the remaining 40kW of energy for cooling, DC-DC converter, etc. so that you essentially aren't using the battery pack.

Or taken to the extreme, 400kW of regen on each acceleration motor, 400kW of reverse toque on the cruise motor, ~40kW of cooling/loads, and ~280kW of regen power into the battery pack. (Figuring ~20% energy loss on the regen that doesn't make it into the battery.)

From the presentation it looked like they had two independent cooling packages, probably essentially the same as in the Plaid S/X. So they can probably only use with ~10kW of power each. I don't know how much power you need for the DC-DC and other loads on the Semi.

 

[mongo]

Actually regen is NOT running the motor in reverse, it's switching the direction of current flow (it is reversing the FUNCTION of the motor though).

Running the motor in reverse meant putting the vehicle into reverse, which was one of the solutions that would NOT work (due to excessive heat generated), still has current flowing from battery to inverter to motor. It was a theoretical way to consume some of the charge in the battery WHILE also slowing the vehicle down, which we now know it could not.

Edited for clarity.

Regen IS putting the motor into reverse.
Or, in the case of tires still having traction, shifting the motor commutation such that it would be running in reverse. Since it isn't, the power flow is motor to pack.
Stalling the motor is also regen, but dumping all the energy into the stator windings and electronics.

Consider an AC motor is also a generator, where a generator spins faster than the line frequency (positive slip) and a motor spins slower (negative slip).

Going downhill, you're converting potential energy to kinetic energy + heat.

And electrical current that can be used to charge the battery (if warm and not 100% SOC).

 

[Oil4AsphaultOnly]

Regen IS putting the motor into reverse.
Or, in the case of tires still having traction, shifting the motor commutation such that it would be running in reverse. Since it isn't, the power flow is motor to pack.
Stalling the motor is also regen, but dumping all the energy into the stator windings and electronics.

Consider an AC motor is also a generator, where a generator spins faster than the line frequency (positive slip) and a motor spins slower (negative slip).


And electrical current that can be used to charge the battery (if warm and not 100% SOC).

No it isn't. If it was, why does regen put energy INTO the batteries, but driving in reverse consume energy?

 

[Oil4AsphaultOnly]

So, you can't supply 1200kW to provide reverse force, still only generating a small percentage of that as heat? Why does it work differently applying reverse torque?

You can even do things like regen ~220kW/each on the acceleration motors, apply 400kW of reverse torque on the cruise motor, and use the remaining 40kW of energy for cooling, DC-DC converter, etc. so that you essentially aren't using the battery pack.

Or taken to the extreme, 400kW of regen on each acceleration motor, 400kW of reverse toque on the cruise motor, ~40kW of cooling/loads, and ~280kW of regen power into the battery pack. (Figuring ~20% energy loss on the regen that doesn't make it into the battery.)

From the presentation it looked like they had two independent cooling packages, probably essentially the same as in the Plaid S/X. So they can probably only use with ~10kW of power each. I don't know how much power you need for the DC-DC and other loads on the Semi.

If my first explanation didn't make it clear, then let's go back to basic physics:

Energy is neither created nor destroyed, it can only be converted. Electrical energy in the batteries is being traded for kinetic energy (by the electric motors, the fact that so little heat is generated is a sign of how efficient they are in the conversion). The kinetic energy is then converted into potential energy (if the motors stopped spinning, then the momentum will carry the truck higher, but slow down in the process). So the 1200kW of power (and X kWh of energy) is being traded for elevation gain as the vehicle drives uphill. As long as there's energy in the batteries and the motors continue to spin, the energy conversion cycle continues. The 36kW of heat produced is whatever energy that couldn't be efficiently converted into kinetic or potential energy. The cooling system has to remove this heat or your motors would melt (for context of scale, your average 4-burner electric cooktop pulls only 7kW when ALL burners are on high).

When going downhill, potential energy is being converted to kinetic energy at the rate of 550kW (see previous chart for conversion rate for a 6% grade). This means unless you continuously remove 550kW of energy out of the system, the truck will ACCELERATE. So to answer your 3rd question. In regen, all that energy is being converted back into electrical energy and going back into the batteries.

But WITHOUT regen, where does that energy go? And that's why I quibbled with mongo about the distinction between regen (reversing the FUNCTION of a motor into a generator) and putting the motor in reverse (motor is still functioning as a "motor"). So to answer your 2nd question, the motor efficiency becomes different (wheels spinning forward, while motor is trying to spin reverse), because it's unable to convert the 550kW of electrical energy into kinetic energy - it's now 100% inefficient (since NONE of applied electrical energy was converted to kinetic energy) and all energy must become heat OR the tires get shredded (if the motor is robust enough to actually spin the tires in reverse against the direction of motion).

Edit: added an explanation for the 100% INefficiency

 

[MC3OZ]

When going downhill, potential energy is being converted to kinetic energy at the rate of 550kW

The other relevant question is how long is it storing energy at the rate of 550kW? i.e. How many kWh.

Assuming it is travelling down a steep 6% grade at an average speed of 30 Miles per hour, it would travel 30 miles.

I could not find stats on the length of the 6% downhill grade sections.

The longest section of downhill road in the world is apparently 72 miles long, but it drops about 2.5 miles over that distance.
That makes the average grade about 3.5%.

 

[Oil4AsphaultOnly]

The other relevant question is how long is it storing energy at the rate of 550kW? i.e. How many kWh.

Assuming it is travelling down a steep 6% grade at an average speed of 30 Miles per hour, it would travel 30 miles.

I could not find stats on the length of the 6% downhill grade sections.

The longest section of downhill road in the world is apparently 72 miles long, but it drops about 2.5 miles over that distance.
That makes the average grade about 3.5%.

550kW is the potential energy to kinetic energy conversion for a 82,000 lb semi truck driving down a 6% grade at 56mph. Going at 30mph, the potential energy converts at a slower pace (293kW), so over 30 miles (it would take 1hr) you can regen less than 293kWh. Rolling resistance, aero drag, and regen efficiency losses will mean you won't recover anywhere near that amount of energy. Just out of curiosity, if you ran the numbers for going down the same 30 mile section of downhill at 60mph instead, you'd get 586kW to convert, but only over 1/2 hr, so the regen limit is 293kwh! Conservation of energy maintained! So going slower downhill will reduce frictional and aero-drag losses, meaning potentially more regen energy! Unless the regen efficiency is lower at slower speeds?!

As for the "longest section of downhill road" example, at a 3.5% grade (see transpondster's equation for how to calculate this yourself), assuming 60mph, you'd need to dissipate 342kW for 1.2 hrs - so about 410kWh. Accounting for aero and friction losses, I'm going to guess (I don't have mongo's equation for those) less than 2/3rds of that will go into the batteries?

 

[Oil4AsphaultOnly]

So, you can't supply 1200kW to provide reverse force, still only generating a small percentage of that as heat? Why does it work differently applying reverse torque?

You can even do things like regen ~220kW/each on the acceleration motors, apply 400kW of reverse torque on the cruise motor, and use the remaining 40kW of energy for cooling, DC-DC converter, etc. so that you essentially aren't using the battery pack.

Or taken to the extreme, 400kW of regen on each acceleration motor, 400kW of reverse toque on the cruise motor, ~40kW of cooling/loads, and ~280kW of regen power into the battery pack. (Figuring ~20% energy loss on the regen that doesn't make it into the battery.)

From the presentation it looked like they had two independent cooling packages, probably essentially the same as in the Plaid S/X. So they can probably only use with ~10kW of power each. I don't know how much power you need for the DC-DC and other loads on the Semi.

Another idea came to mind, but please read my previous reply first for the physics.

Although we can't apply the full 1200kW electrical power to the task of slowing the vehicle down, we CAN apply a fraction of it! Putting the motor in reverse would make it 100% inefficient and thus converts all that electrical energy into heating the motor. At 6kmph down a 6% grade, the potential energy to kinetic energy conversion rate is only 36kW (the same heat produced as when all 3 motors are at full throttle going uphill, so the cooling system should be able to manage this and possibly more). At this speed, the motors, running in reverse can slow the truck down enough to keep it from accelerating downhill without exceeding the cooling capacity of the powertrain cooling system. If the cooling system can handle more than 36kW, then this method can be used at higher speeds.

But obviously, it would be more effective to just warn drivers to NOT 100% charge a fully-loaded semi truck at the top of a mountain. Or just locate the superchargers no more than 1/3 of the way from the top of a mountain.

 

[MC3OZ]

But obviously, it would be more effective to just warn drivers to NOT 100% charge a fully-loaded semi truck at the top of a mountain. Or just locate the superchargers no more than 1/3 of the way from the top of a mountain.

Yes, this is the point I was suggesting.

If they have entered their route into a route planner, then the software knows where they are going and can recommend the right charge level.

They should mostly be running fixed scheduled with replanned charging stops.

Someone getting to the top of a mountain and charging to 100% at a charger located on top of the mountains seems unlikely to me.

It is best to have chargers located at the foot of mountains. There is more flat land and that is very handy for driver climbing the mountain.

 

[Stretch2727]

Do we think applying the Semi clutches to other vehicles is a possibility? Were all 3 motors permanent magnet so this is one reason to disengage the motors? I believe the Model 3/Y use an induction motor in the front so I understand these can freewheel with limited drag so probably not much benefit. The model s/x now have permanent motors on both axels so it could have some benefit on the front axel to freewheel? If we do see it on other vehicles they will probably start with the CT.

 

[mongo]

No it isn't. If it was, why does regen put energy INTO the batteries, but driving in reverse consume energy?

But WITHOUT regen, where does that energy go? And that's why I quibbled with mongo about the distinction between regen (reversing the FUNCTION of a motor into a generator) and putting the motor in reverse (motor is still functioning as a "motor"). So to answer your 2nd question, the motor efficiency becomes different (wheels spinning forward, while motor is trying to spin reverse), because it's unable to convert the 550kW of electrical energy into kinetic energy - it's now 100% inefficient (since NONE of applied electrical energy was converted to kinetic energy) and all energy must become heat OR the tires get shredded (if the motor is robust enough to actually spin the tires in reverse against the direction of motion).

If the wheel are not slipping, a motor cannot spin opposite from the vehicle direction.
Applying reverse commutation to the motor's direction results in regen.
Moving forward with reverse commutation -> regen (removing kinetic energy)
Moving backwards with reverse commutation -> drive (adding kinetic energy)

There is no situation by which you can remove energy from the system by putting more energy into it. The premise of more braking in a full battery is inherently flawed. There is no power flow into the motor (unless purposely trying to heat it with a non-motive waveform)

A plugged/ shorted motor is 100% electrically inefficent, but is actually in regen with the inverter and stator as the load resistors. So kinetic -> electrical -> heat.

 

[Krash]

…The model s/x now have permanent motors on both axels so it could have some benefit on the front axel to freewheel? …

Sounds like a question. I had assumed that the new 2022 S/X motors were induction in the rear, like the previous models, but it appears that is not the case. So unlike 3/Y which started IPMSynRM rear and added induction front that can freewheel, and the mid release (Raven) S/X which added IPMSynRM front so the rear induction can freewheel, the 2022 S/X has no induction motors. I think since no motors can freewheel on the latter, that if one of the two (or three in plaid) fail, you’re likely getting towed.

 

[Stretch2727]

Sounds like a question. I had assumed that the new 2022 S/X motors were induction in the rear, like the previous models, but it appears that is not the case.

Yes this is why the question. I thought I saw somewhere where there would actually be a larger % gain in efficiency than the Semi as the drag due to the permanent motor is a larger percentage total consumption. I guess it depends on how expensive and complex the solution for the clutch.

 

[Krash]

According to this article, which claims to be a simplified explanation of Sandy Monro’s findings, all three plaid motors are the same. And all those plaid stators (outside) are the same as the 3/Y. But all the plaid rotors (inside) are the high compression carbon fiber wrapped copper.

 

[Oil4AsphaultOnly]

If the wheel are not slipping, a motor cannot spin opposite from the vehicle direction.
Applying reverse commutation to the motor's direction results in regen.
Moving forward with reverse commutation -> regen (removing kinetic energy)
Moving backwards with reverse commutation -> drive (adding kinetic energy)

There is no situation by which you can remove energy from the system by putting more energy into it. The premise of more braking in a full battery is inherently flawed. There is no power flow into the motor (unless purposely trying to heat it with a non-motive waveform)

A plugged/ shorted motor is 100% electrically inefficent, but is actually in regen with the inverter and stator as the load resistors. So kinetic -> electrical -> heat.

No, commutation implies current flowing from the battery into the stators to generate a magnetic field. Regardless of whether that magnetic field is leading or lagging, it's still the electric current INDUCING a magnetic field. That generated magnetic field would be repelling the magnetic field of the permanent magnets in the rotor. Having them work against each other produces heat in the permanent magnets. In regen, there's no applied current flow, instead the magnetic field from the permanent magnets in the rotor INDUCES an electric field in the stator and generates an electric current in the OPPOSITE direction thereby charging the batteries. It's Maxwell's right-hand rule.

Looked at another way, you can NOT have current flowing both directions AT THE SAME TIME in the same wires.

Having said that, I just realized that you're probably thinking of how AC Induction motors regenerate.

In AC Induction motors, current has to flow to generate a magnetic field. No magnetic field, then no regen. But the key is how electrons flow with AC. The generated magnetic field on the stator is ALWAYS switching polarity, at whatever rate the frequency is set to. Whether or not the motor is in regen mode or power mode is dependent on the phase difference between the frequency of the rotor versus the inverter. But the same exclusivity still applies, the AC current (that's generating the induced magnetic field) can't be leading AND lagging at the same time. So no. A plugged/shorted motor is indeed 100% electrically inefficient, but it's NOT in regen mode.

And in my "remove energy by putting more energy into it" scenario, that's a misunderstanding - within the system's view there was no net gain/loss of energy. All energy simply got converted to heat, from which the cooling system had to dump it out to the ambient air.

 

[MC3OZ]

Do we think applying the Semi clutches to other vehicles is a possibility? Were all 3 motors permanent magnet so this is one reason to disengage the motors? I believe the Model 3/Y use an induction motor in the front so I understand these can freewheel with limited drag so probably not much benefit. The model s/x now have permanent motors on both axels so it could have some benefit on the front axel to freewheel? If we do see it on other vehicles they will probably start with the CT.

Seems like a trade-off to me.

Pros:-

• Greater range
• Perhaps a smaller battery pack, reduced cost and weight.

Cons:-

• More drive unit weight
• More complexity, including manufacturing complexity.
• More cost.

So IMO it is worthwhile if the additional range can be achieved with a smaller battery pack, the cost equation balances out to at least even, and the additional manufacturing complexity is not significant.

Anything thing means fewer cells, also means higher production with the same amount of battery raw materials.

If Tesla does it, it will be because they are chasing fewer cells, not more range.

 

[Stretch2727]

Seems like a trade-off to me.

Pros:-

• Greater range
• Perhaps a smaller battery pack, reduced cost and weight.

Cons:-

• More drive unit weight
• More complexity, including manufacturing complexity.
• More cost.

So IMO it is worthwhile if the additional range can be achieved with a smaller battery pack, the cost equation balances out to at least even, and the additional manufacturing complexity is not significant.

Anything thing means fewer cells, also means higher production with the same amount of battery raw materials.

If Tesla does it, it will be because they are chasing fewer cells, not more range.

Thanks. Yeah hard to say as we don't know the cost or the efficiency improvement it achieved. The semi is also likely to spend more time in a cruise mode on the highway where disengaging the motors is beneficial.

One side note. I have had my dual motor model 3 on jack stands in tow mode. Tow mode basically releases the emergency brake to allow it to be pulled onto a truck. The resistance from the rear permanent magnet motor is quite strong and you can feel the pulsing of the magnets. The front induction motor freewheels very easily.