Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

One motor per wheel: pros/cons

This site may earn commission on affiliate links.

Saghost

Well-Known Member
Oct 9, 2013
8,224
7,110
Delaware
(mod note: split from Multi-gear EVs are probably the future)
More importantly I think is the differential 'problem'. It stands to reason that the technology will advance to the point where the propulsion motor in an EV could be direct drive, but you'd then need a motor per wheel to manage the differential wheel speeds. Very feasible of course, but a second motor on the same axle increases the overall complexity that one is otherwise trying to drive down by implementing direct drive.

An interesting potential solution is an integrated dual-ish DD motor, where you have only one unit that has mostly one set of components and is thus mostly one motor (as opposed to two motors bolted together), but it doesn't push its power through a traditional-ish mechanical differential like the model S does now.

One motor per wheel is desirable for a bunch of other reasons, both traction and stability/handling benefit greatly from being able to independently apply arbitrary amounts of torque to each wheel. All the serious EV supercars are doing it already, and I rather expect that it'll become either standard or an upgrade option on higher class EVs in the future - possibly even on all of them.
Walter
 
Last edited by a moderator:
One motor per wheel is desirable for a bunch of other reasons, both traction and stability/handling benefit greatly from being able to independently apply arbitrary amounts of torque to each wheel. All the serious EV supercars are doing it already, and I rather expect that it'll become either standard or an upgrade option on higher class EVs in the future - possibly even on all of them.
Walter

Could be, but I don't think so. Torque vectoring is a thing of the present in ICEs, so its not like you need multiple motors like the supercar EVs...and after vectoring, there really aren't many other reasons why you'd want more motors than less motors other than the technological limits of what's currently available. The beauty of an EV is simplicity, and less is...umm...more simple.

If hub mounted direct drive motors become light, efficient, and powerful enough to make sense, I could see individual motors making their way into production EVs.
 
  • Helpful
Reactions: Brando
more motors is not a panacea. It reduces max hp/torque available at each wheel.

for a fixed total hp and torque for a vehicle spec, use 500hp/600Nm torque for an example
One motor and the right diffs can send all its torque to any wheel. The latest Audi Quattros are a great example of huge torque vectoring to great effect.

One motor front and rear (250hp/300Nm) can only make use of its torque on that axle

4 off 125hp/150Nm motors fitted one to each wheel can only provide power to that wheel.
So in the example max 150Nm to each wheel - as significantly worse position than being able to direct a full 600Nm to one wheel or for 2 motors 300Nm to any wheel

(the above ignores losses/efficiencies etc but illustrates the point)

imho Tesla are about on the money for drivetrain simplicity vs capability
 
  • Informative
Reactions: David99
more motors is not a panacea. It reduces max hp/torque available at each wheel.

for a fixed total hp and torque for a vehicle spec, use 500hp/600Nm torque for an example
One motor and the right diffs can send all its torque to any wheel. The latest Audi Quattros are a great example of huge torque vectoring to great effect.

One motor front and rear (250hp/300Nm) can only make use of its torque on that axle

4 off 125hp/150Nm motors fitted one to each wheel can only provide power to that wheel.
So in the example max 150Nm to each wheel - as significantly worse position than being able to direct a full 600Nm to one wheel or for 2 motors 300Nm to any wheel

(the above ignores losses/efficiencies etc but illustrates the point)

imho Tesla are about on the money for drivetrain simplicity vs capability

I think you may be under a misapprehension about how differentials work. The thing is, your basic open differential makes the torque on both wheels equal to whatever the one with less grip delivers - any torque beyond that disappears by that wheel slipping.

Torsen differentials use side gears to bias the torque, so that the slower turning wheel gets more, but it's still a ratio of the two.

The only thing that behaves the way you're describing is a locked differential - one that has the spider locked by a clutch forcing the two wheels to spin at the same speeds.

In any case except the locking differential, you'd have more power delivered to the ground by individual wheel motors than the differential. The locking differential might deliver more, but it comes with serious handling issues due to the locked wheel speeds...

Don't get me wrong - Quattro is a great thing - but individual wheel motors with enough torque to approach the tire limits on dry ground is the ultimate answer, and one that I think will gradually spread through the fleet from the top end.
Walter
 
  • Informative
Reactions: David99
more motors is not a panacea. It reduces max hp/torque available at each wheel.

for a fixed total hp and torque for a vehicle spec, use 500hp/600Nm torque for an example
One motor and the right diffs can send all its torque to any wheel. The latest Audi Quattros are a great example of huge torque vectoring to great effect.

One motor front and rear (250hp/300Nm) can only make use of its torque on that axle

4 off 125hp/150Nm motors fitted one to each wheel can only provide power to that wheel.
So in the example max 150Nm to each wheel - as significantly worse position than being able to direct a full 600Nm to one wheel or for 2 motors 300Nm to any wheel

(the above ignores losses/efficiencies etc but illustrates the point)

imho Tesla are about on the money for drivetrain simplicity vs capability

Like Saghost pointed out, even if you did have a 600Nm motor coupled to all four wheels, or a 300 Nm motor on each axis it could never put down more than 150 Nm per wheel except with a type of differential set-up that you will mostly only find in terrain cars (say for example a Toyota Landcruiser where it's possible to lock the differential to get through very difficult terrain or some such situation). A car with a locked differential can't really be driven safely on regular roads though, at higher speeds, that is if you like to be able to steer :)

So yes, in-wheel motors is the optimum solution. Now if these could be dimensioned so that they didn't need gearing at all (direct drive on the wheel 1:1) that would likely be even greater.
 
sorry, you are correct - I should have stated torque vectoring diffs (of whatever type), not open diffs.

"but individual wheel motors with enough torque to approach the tire limits on dry ground is the ultimate answer"

indeed but then even in EV world you are probably iro 250hp/wheel, not a cheap solution.
 
So yes, in-wheel motors is the optimum solution. Now if these could be dimensioned so that they didn't need gearing at all (direct drive on the wheel 1:1) that would likely be even greater.

Maybe. In wheel motors are great for packaging, and do simplify the powertrain. However, in trade for the 8 CV joints you eliminate and the space you gain in the middle of each axle, you pick up a bunch of unsprung weight (bad for handling,) and also now have a bunch of high voltage wires (and hoses if the motors are water cooled,) that are moving with every suspension change and presumably likely to fail sooner.

My guess is that the ideal for road cars will remain similar to the Rimac, SLS AMG, and C-X75 - one motor for each wheel, hard mounted and driving the wheel through a conventional half-shaft. (Off road is different - in wheel motors might have an advantage there because that will allow much more suspension articulation.)
Walter

- - - Updated - - -

sorry, you are correct - I should have stated torque vectoring diffs (of whatever type), not open diffs.

"but individual wheel motors with enough torque to approach the tire limits on dry ground is the ultimate answer"

indeed but then even in EV world you are probably iro 250hp/wheel, not a cheap solution.


Okay, but true torque vectoring differentials are heavy, complicated, expensive, and add failure points. I wouldn't be surprised if the quad motors were cheaper and lighter - they'll certainly be more reliable.

Also, note the P85D approach - you really only need big motors on the rear wheels, and would likely gear the front motors higher for efficient highway cruise.
 
Walter, those are great points and maybe keeping the motors out of the wheel is better, for now. But just imagine that with in-wheel motors you could even do away with the axles completely! Electric power could likely be supplied across the body --> wheel through the center bore. The "wiring" to supply to electrical power could be an integrated part of the suspension (which is made from metal). Unsprung weight is a bit of an over exaggerated problem, isn't it? I'm thinking graphite wheels and light, but powerful, motors.
 
Walter, those are great points and maybe keeping the motors out of the wheel is better, for now. But just imagine that with in-wheel motors you could even do away with the axles completely! Electric power could likely be supplied across the body --> wheel through the center bore. The "wiring" to supply to electrical power could be an integrated part of the suspension (which is made from metal). Unsprung weight is a bit of an over exaggerated problem, isn't it? I'm thinking graphite wheels and light, but powerful, motors.

I'm not a suspension expert by any means, but what I've read is that a pound on the lower side of the suspension has the same effect on handling and bumps as three pounds on the upper side.

Once the motors get light/powerful enough, that might be traded against other traits for a better overall result - and with sufficiently reliable motors that have the torque to burn rubber on all four wheels, you could in principle eliminate the brake system and its weight (need some sort of a solution for energy beyond what the battery can take; locomotive style "dynamic braking" resistance/radiators is a possibility - with a little creativity, the HVAC system can probably dump power overboard at need by heating coolant as if for the cabin and sending the hot coolant to the radiators up front instead though it can't handle the surge power levels, so you'd need either an ultracapacitor stack or to hold some battery space for panic stops and dump it if needed.)
Walter
 
So how will they make up for the loss of that 9.73 torque multiplier? 9.73 times the current? 9.73 times the motor diameter? Direct drive may work for wind turbines and propellers where start up torque isn't important but I'm not so sure about vehicle motors.

Wikipedia, carbon nanotubes:
In theory, metallic nanotubes can carry an electric current density of 4 × 10^9 A/cm^2, which is more than 1,000 times greater than those of metals such as copper,


Electric cars will push multi-gear boxes into museums. In Tesla D models different gear ratios are not cause for better efficiency. Efficiency of an electric motor is not good at small fraction of its max power. So adding smaller front motor and using only it, when low power is needed, will increase efficiency.

- - - Updated - - -

To be clear those multi motor cars, such as the Rimac, are still using gear reductions, not direct drive. Even some of the wheel motors are using in wheel gear reduction.

Toroidion has 4 motors direct drive. Total 1 MW. Unfortunately there is not much data available.
 
Wikipedia, carbon nanotubes:
In theory, metallic nanotubes can carry an electric current density of 4 × 10^9 A/cm^2, which is more than 1,000 times greater than those of metals such as copper,


Electric cars will push multi-gear boxes into museums. In Tesla D models different gear ratios are not cause for better efficiency. Efficiency of an electric motor is not good at small fraction of its max power. So adding smaller front motor and using only it, when low power is needed, will increase efficiency.

- - - Updated - - -



Toroidion has 4 motors direct drive. Total 1 MW. Unfortunately there is not much data available.

This is because the car doesn't exist. The CEO said one month ago at the Monaco (?) car show: if someone threw a lot of money at us then maybe we could be here next year showing you what the production model looked like.
 
So yes, in-wheel motors is the optimum solution. Now if these could be dimensioned so that they didn't need gearing at all (direct drive on the wheel 1:1) that would likely be even greater.

In-wheel motor is a bad idea: It will increase unsprung mass.
If you happen to drive into a hole on a road you will damage wheel and motor.

If motor can be made very small with superconductors, nanotubes or something else, then perhaps in-wheel motor would be useful.

I agree: 2 motors are better than 1 and 4 motors better than 2.
 
This is because the car doesn't exist. The CEO said one month ago at the Monaco (?) car show: if someone threw a lot of money at us then maybe we could be here next year showing you what the production model looked like.

One Toroidion does exist. I expect some performance numbers in this summer. Production is far away. Perhaps never.

- - - Updated - - -

Theory doesn't move vehicles.



Theory is necessary first step. Practical method to manufacture good quality carbon nanotubes is missing.
 
Maybe. In wheel motors are great for packaging, and do simplify the powertrain. However, in trade for the 8 CV joints you eliminate and the space you gain in the middle of each axle, you pick up a bunch of unsprung weight (bad for handling,) and also now have a bunch of high voltage wires (and hoses if the motors are water cooled,) that are moving with every suspension change and presumably likely to fail sooner.
For reference, the train industry switched to wheel motors a looooong time ago. However, *they are sprung weight*. The connection is made through a quill drive.

I don't know if this could be applied to automobiles straightfowardly. It seems like the obvoious way to do it, but there may simply not be enough space. Railroad cars have a multi-stage suspension; the traction motors are mounted on the bogie frame. There's one suspension between that and the wheels, and another between that and the car body. Automobiles generally only have a single-stage suspension.

Thinking about it, it's probably straightforward on the unsteered rear wheels, but a big pain on the steered front wheels, which probably would require two layers of suspension with the motor at the intermediate level.
 
I don't know if I'd call a motor mounted to an axle driving a wheel through a gear reduction a "wheel" motor. Certainly not the "in wheel motor" that was being discussed.

Well, the point is that there's one motor per wheel, so the differential is electronic, not mechanical. I thought that was the essence of "one motor per wheel", which is the topic title.

(In older train designs there are full axles, but many newer ones *do* use half-axles, with separate motors on left and right wheels, to avoid wheelslip.)