Speak EV - Electric Car Forums banner

221 - 240 of 280 Posts

·
Registered
Joined
·
540 Posts
It's very simple... I either have 4 motors operating at point A, or 1 motor operating at point B for the same performance. The 4 motor at point A option improves efficiency by about 10%.

PS... Most of the peak torque/power curve is limited by the battery not the motor.
132275

 

·
Registered
Joined
·
3,826 Posts
PS... The peak torque/power curve is limited by the battery not the motor.
No it's not, it's limited by the mechanical design of the motor and its ability to reject excess heat and survive the torque being created. There is no upper limit to the size of battery that can be linked to a motor, just a practical limit to current and voltage applied to avoid overstressing the motor.
On a vehicle as expensive as a Semi ($150k+) Tesla could afford to design and implement say 8 smaller motors if there were significant benefit in doing so, or even mix motor types for efficient cruising and improved maximum power (steep grade) performance. I also notice no change speed gearing has been suggested.
Playing your game, 2 motors at Point C would be better than 4 at A with greater efficiency and better turndown performance.

132276
 

·
Registered
Joined
·
3,826 Posts
That's for a particular battery, not the motor design.
 

·
Registered
Joined
·
540 Posts
The motor is capable of the peak torque at all speeds except for the range labeled "voltage limited"... the reason the peak torque drops with increasing speed is due to the battery current limit setting in the controller, until one reaches the "voltage limited" portion of the chart, at which point the current and torque drop regardless of the software defined battery and phase current limits, because the back emf voltage generated by the spinning magnets, which opposes the battery voltage, approaches the same voltage as the battery in this speed range.
 

·
Registered
Joined
·
3,895 Posts
Or 3 times over its limit. 🤦‍♂️

Perhaps there is an optimal size of motor and economies of scale in using an existing product? A motor in Semi will be running closer to its limits than a typical M3 - the Semi is rated at a continuous 60/65 (depending on who you believe) up a 5% grade so does not have the reserves of power that the same motors in an M3 have.
I think there's three clear reasons why the Tesla Semi uses four Model 3 motors. (or a derivative of it, we don't know if it's 100% identical in ways such as cooling since the semi has not launched yet, only the prototypes are using Model 3 motors so far..)

1) Commonality of parts. If you can take four off the shelf motors designed for the rear of the Model 3 - motors that are very efficient, powerful, compact and well proven (thus far anyway) and have been designed for "a million miles", (notice the oil filter on the side of the gearbox casing - how many other EV motor/gearboxes have a large external oil filter on them ? None that I've seen so far..) then why design a larger motor to be used on it's own or perhaps in a pair which are common to no other Tesla product ? A lot of time and money to do the R&D on yet another new motor, more skews to manufacture at the already overloaded production lines, for no particularly good reason. The Semi will never come anywhere near it's promised launch date if they were to design an all new motor for it and then have to ramp production up for it. (New production line and/or tooling etc)

By using the Model 3 motor they skip all that money and time intensive R&D and already have a production line churning them out today - as soon as the Semi is ready to go they can "divert" a percentage of Model 3 motors to the Semi as and when needed on an agile basis. It makes a lot of business sense.

2) The Semi has four driven wheels, and four motors.... think about that for a second. What is the obvious thing to do with that ? Electronic differentials of course. Why have a hulking great differential (or more than one) to drive four wheels from two motors or even one motor... Differentials are large, heavy, lose power, and allow for loss of power transmission if one wheel slips, without a limited slip differential, which in itself loses more power and increases tyre wear on corners.

One motor per driven wheel means elimination of all differentials, and that has a lot of advantages. Better miles/kWh efficiency, reduced cost, weight and volume consumed in the truck by the mechanics, and radically better traction and traction control, with electronic differential algorithms that are under computer control and field upgradable. One wheel hits a patch of ice when you're trying to get started up a hill ? No problem, it has no effect on the other 3 wheels which aren't on a patch of ice as there is no differential connecting them. The torque and wheel speed can be individually controlled for each wheel on low traction surfaces allowing the truck to pull away on slippery surfaces where traditional trucks will get stuck. Traction control can be provided entirely by adjusting motor torque without having to use differential brake control, (on the slipping wheel) which isn't as fast or effective as brake calipers can't respond as fast as electric motors. It's win win all around.

3) A somewhat minor point, but having 4 independent motors each with their own independent driveshaft to their own wheel gives a lot of redundancy. On a single motor plus many differentials design if the motor fails you're stranded - just like an ICE truck. One motor per wheel with each motor having it's own independent driveshaft to its own wheel - if any motor fails you simply lose 1/4 of your total power and 1/4 of your traction. Your top speed and speed up and down inclines will be reduced, so you'll need to drive more carefully, (and the software in the truck can force you to drive slower and more carefully by going into a semi-limp mode...) but you'll still be able to drive to the nearest service depo or even the destination. I expect the motors used in the Semi to be very reliable but having the truck still able to continue it's journey if a motor fails is a major selling point to reassure customers that this thing will be reliable and won't let them down.

I think 4x Model 3 motors in the Semi makes total sense from both a business and engineering perspective.
 

·
Registered
Joined
·
540 Posts
One motor per driven wheel means elimination of all differentials, and that has a lot of advantages. Better miles/kWh efficiency, reduced cost, weight and volume consumed in the truck by the mechanics, and radically better traction and traction control
I agree with your current assessment that multiple motors improves efficiency.

What does that do to your theory that dual motor must be more efficient than a single motor ?

It's not in this case because Tesla added a second motor to increase performance, not efficiency.
 

·
Registered
Joined
·
3,895 Posts
I agree with your current assessment that multiple motors improves efficiency.
It can if it eliminates heavy, power wasting limited slip differentials as in the case of the Semi. Driving 4 wheels from one motor requries 3 differentials, driving 4 wheels from 4 motors requires zero differentials. Just a step down gear and a half shaft per wheel, which you need regardless of whether you have a differential or not.

However in the case of a Model 3 going from one motor to two means you're actually going from one differential to two. So losses are increased. You're also going from 2 half shafts to 4 half shafts. More losses.

In the Model 3 case the number of driven wheels is changing, in the Semi case it isn't - tractor trailer units always have 4 driven wheels whether they have 4 electric motors or one diesel engine driving the wheels.

Nice try though. :)
 

·
Registered
Joined
·
540 Posts
Also the 2nd motor in a model 3 is an induction motor which suffers additional rotor excitation losses compared to a 2nd identical permanent magnet motor.
 

·
Registered
Joined
·
3,895 Posts
The motor is capable of the peak torque at all speeds except for the range labeled "voltage limited"... the reason the peak torque drops with increasing speed is due to the battery current limit setting in the controller, until one reaches the "voltage limited" portion of the chart, at which point the current and torque drop regardless of the software defined battery and phase current limits, because the back emf voltage generated by the spinning magnets, which opposes the battery voltage, approaches the same voltage as the battery in this speed range.
Rubbish. The motor can only produce peak torque up until the speed where the motor reaches peak power. This is typically around 25-40 mph on most BEV's depending on motor and gear ratio. From that point on (until the "voltage limited" point you mention - which is beyond the maximum allowed road speed of the car so not particularly relevant to real world conditions) it is in the constant power range which means torque falls with increasing speed.

All electric motors have a maximum thermal dissipation capability which set a limit on the maximum power it can operate at, a limit enforced by the controller. The battery pack and the electronics each have their own maximum continous power limits too. On a given car it's possible that the battery pack sets the peak power limit (in the case of a small battery) or that the motor sets the peak power limit. (In the case of a large battery)

But rest assured that there is a wide peak power region on all real BEV's regardless of whether the battery or motor are hitting their limit first, and it's certainly not true that a motor can produce peak torque all the way up to the voltage limited region for any length of time as it will be exceeding its power dissipation and cooling limits.
 

·
Registered
Joined
·
540 Posts
Rubbish. The motor can only produce peak torque up until the speed where the motor reaches peak power. This is typically around 25-40 mph on most BEV's depending on motor and gear ratio. From that point on (until the "voltage limited" point you mention - which is beyond the maximum allowed road speed of the car so not particularly relevant to real world conditions) it is in the constant power range which means torque falls with increasing speed.

All electric motors have a maximum thermal dissipation capability which set a limit on the maximum power it can operate at, a limit enforced by the controller. The battery pack and the electronics each have their own maximum continous power limits too. On a given car it's possible that the battery pack sets the peak power limit (in the case of a small battery) or that the motor sets the peak power limit. (In the case of a large battery)

But rest assured that there is a wide peak power region on all real BEV's regardless of whether the battery or motor are hitting their limit first, and it's certainly not true that a motor can produce peak torque all the way up to the voltage limited region for any length of time as it will be exceeding its power dissipation and cooling limits.
On my skateboard, If I set the battery current / motor current limits in the speed controller to 80a/80a, then I get peak torque until I reach the voltage limited speed (30-35mph - green and yellow lines, top left chart).

If I set my battery current / motor current limits in the controller to 20a/80a, then torque drops between 5-10mph as the torque is battery current limited (blue line, top left chart and black line, bottom left chart).
132290






 

·
Registered
Joined
·
3,895 Posts
When did we start talking about skateboards ?

Check the actual torque/power vs speed curve of any BEV. (Easy to derive from the motor curves with a single speed gearbox...)

More or less constant torque from zero to some speed (about 25-40mph) then constant power and falling torque up to near the top speed with typically a small amount (maybe 10%) fall off in power shortly before top speed where the motor is starting to become less efficient. (Depends on type of motor)

Looks like this although most BEV motors actually go to 8000rpm or more so the small droop in the red line at the right hand end is off the end of the graph:

132301


(Image pinched from here)

Whether the constant power region is dictated by the battery power output limits or motor power dissipation limits varies from model to model but it's always there. The controller has a set maximum power that it will allow to be drawn from the battery to protect both battery and motor, resulting in the curves above.
 

·
Registered
Joined
·
540 Posts
The motor in your chart is capable of 400 Nm @ 6000rpm but the battery isn’t. In any case, with a better battery that can output more amps— 2 motors from the chart at 200 Nm @ 6000rpm will be more efficient than 1 motor at 400Nm @ 6000rpm

PS... Most of the peak torque/power curve is limited by the battery not the motor.
 

·
Registered
Joined
·
3,895 Posts
The motor in your chart is capable of 400 Nm @ 6000rpm but the battery isn’t. In any case, with a better battery that can output more amps— 2 motors from the chart at 200 Nm @ 6000rpm will be more efficient than 1 motor at 400Nm @ 6000rpm
Sorry, but on what basis are you saying the motor is capable of 400 Nm @6000 rpm ? There's nothing in the graph or the article I borrowed the graph from for you to make that determination. You're just guessing based on no information.

As I've already said whether the motor or the battery is the limiting factor in the peak power region depends on the battery and motor combination. If the battery is relatively big and actively cooled and the motor modest it can very easily be the motor that hits its thermal limits first as the motor has a much smaller thermal mass and dissipates its power in a much more confined space - compare the physical size and mass of a battery vs a motor.

You're trying to extrapolate theory from your skateboard that doesn't apply to cars.

Electric skateboards aren't operating at power levels of hundreds of kW, they don't have pumped active liquid cooling systems with radiators, they don't have 1-3 hours of continous run time at 70mph for heat to build up like a car does, they aren't expected to last an 8 year warranty with no failures, or last 10-15 years in total.

All these factors cause the designer to choose a maximum safe power level for both the battery and the motor, and this is programmed into the controller to give the constant/peak power region.

All EV's have this region, and none of them produce peak torque past about 40% of their maximum road speed because of it. To attempt to do so would exceed the continuous power rating of the battery, motor or most likely both.

Electric motors have both maximum torque limits and maximum power limits based on dissipation. In the size and form they're used in EV's no electric motor is rated for the power that would be needed to produce peak torque all the way to maximum rpm.

Show me even one example of this on a commercial EV.
 

·
Registered
Joined
·
3,826 Posts
I think 4x Model 3 motors in the Semi makes total sense from both a business and engineering perspective.
Agreed, there's a synergy between a motor per driven wheel and the existing product range.
My argument is against @metastable and his assertion that more motors is always better.
 

·
Registered
Joined
·
540 Posts
Sorry, but on what basis are you saying the motor is capable of 400 Nm @6000 rpm ?
Specifically you can tell where the power line flattens that it’s hitting the battery current limit setting in the electronic speed controller. On the other hand, the same motor current (phase current) at 6000rpm or 0rpm gives same torque and same resistive losses in the motor (iron losses a bit higher at 6000rpm). The motor is capable of 400Nm as shown at 0rpm. The leveling off of power and dropping of torque is a result of the hitting battery current limit setting in the controller at a particular rpm & torque so if one draws more amps from the battery (but same phase amps as 0rpm) at 6000rpm you can get 400Nm. Specifically it takes at the very least 251kW of power from the battery for 400Nm @ 6000rpm which exceeds the indicated 140kW battery power limit by more than 100kW. Motor constant KT (torque per phase amp) doesn’t change with rpm. While capable of 400Nm at 6000rpm, it brings the motor outside its efficiency “comfort zone” so 2 motors at 200Nm @ 6000rpm will be more efficient than 1 motor at 400Nm @ 6000rpm. The motor won’t produce significantly more heat doing 400Nm at 6000rpm as it would doing 400Nm at 0rpm (any minor extra heating can be attributed to bearing and iron losses).





^Torque is directly proportional to phase amps so when battery amps level off and phase amps drop, the motor is “Battery Amps Limited”
 

·
Registered
Joined
·
540 Posts
My argument is against @metastable and his assertion that more motors is always better.
Where did I say the words “more motors is always better?” You’re simply making a straw man argument.

That said, 2 motors can output twice as much power at peak efficiency as 1 identical motor, 4 motors can output 4x as much power at peak efficiency as 1 identical motor, etc.
 

·
Registered
Joined
·
3,895 Posts
Specifically you can tell where the power line flattens that it’s hitting the battery current limit setting in the electronic speed controller.
Correct.

But why does the speed controller limit the current (actually the power, since a constant current won't give a flat power curve as the cells discharge) at that particular point ? To ensure that neither the motor or battery (or controller itself for that matter) have their power dissipation ratings exceeded. The maximum power is part of the design envelope for the motor and indeed the whole high voltage system.

The motor has a maximum power which it can safely dissipate with the cooling available before overheating and possible damage. (which can be cumulative over time) Can you run it at a higher power if you wanted to ? Of course, for a short time, and if you don't mind it being damaged and not lasting the warranty period if its anything more than for a short time. (Seconds, perhaps minutes depending on the degree of overload)

On the other hand, the same motor current (phase current) at 6000rpm or 0rpm gives same torque and same resistive losses in the motor (iron losses a bit higher at 6000rpm). The motor is capable of 400Nm as shown at 0rpm. The leveling off of power and dropping of torque is a result of the hitting battery current limit setting in the controller at a particular rpm & torque so if one draws more amps from the battery (but same phase amps as 0rpm) at 6000rpm you can get 400Nm. Specifically it takes at the very least 251kW of power from the battery for 400Nm @ 6000rpm which exceeds the indicated 140kW battery power limit by more than 100kW. Motor constant KT (torque per phase amp) doesn’t change with rpm. While capable of 400Nm at 6000rpm, it brings the motor outside its efficiency “comfort zone” so 2 motors at 200Nm @ 6000rpm will be more efficient than 1 motor at 400Nm @ 6000rpm. The motor won’t produce significantly more heat doing 400Nm at 6000rpm as it would doing 400Nm at 0rpm (any minor extra heating can be attributed to bearing and iron losses).
In all of this you seem to be labouring under the assumption that batteries have power limits (which they do) and motors don't. The motor can't just take as much power as the battery will give it without overheating. I'm not sure what you're not understanding about this.

All electronic devices that dissipate power have power dissipation and resulting temperature rise limits and as a result safe operating areas of operation. I guarantee to you that any motor used in any production EV cannot put out full low speed torque at near maximum RPM without going way outside its thermal safe operating area. Forget about whether the battery can do it or not.

To design a motor that can produce full torque at full RPM essentially just means you are purposely sacrificing the low speed torque and this would not result in a satisfactory car. You are literally leaving low speed performance on the table, since acceleration at any given speed is proportional to torque not power. More low end torque means better acceleration of the line at low speeds.

When staying within the power dissipation envelope of the motor it naturally follows that the slower you go the more torque you can put out with the same power dissipation. This is exactly what you want for a single speed gearbox. At some point you reach the maximum torque limit of the motor (which may be a mechanical strength limit) and from that speed down the controller puts the motor into a constant torque range.

Please give up on the fantasy that EV motors can and should produce full torque all the way to maximum RPM. The desire for maximum acceleration at any given speed (optimising low speed torque) and the fundamental relationship that Power = torque x rpm precludes this from happening.

The graph I presented is very indicative of what practically every EV does, and I don't think it's the case that all EV's are battery power limited as you are trying to suggest.

Take a Kia e-Niro for example. The peak power of the motor is 149kW, and the battery pack is 64kWh. So at maximum, foot to the floor acceleration that's a C discharge rate of 2.3. Piece of cake for any Lithium Ion chemistry, especially one that has liquid cooling like the e-Nrio does. The battery is not the limiting factor here, it's not really even breaking a sweat. Even the 16kWh battery in my Ion is discharging at over 3C under full acceleration, and with no active cooling.

High performance Tesla's reach something like 5.5C discharge rate under full acceleration so in their case there may be some battery power limiting going on, but on EV's with decently large batteries and more modest acceleration it's going to be the motor power dissipation limits that are hit before battery limits.
 
221 - 240 of 280 Posts
Top