TheEEStory.com

Lensman wrote:

grizz wrote:

Why would an electric motor provide braking if it is not connected to its load. All motors I have ever seen turn freely with no load attached.

If so, then why does the Tesla Roadster perform significant engine braking when you take your foot off the accelerator?

It's designed to do so. They could just as easily have decided to have regen set to zero. If they really wanted to make it more complex to get back into acceleration they could simply have floated the power electronics and coasted.

The presence and amount of regeneration is a design decision.

Lensman wrote:

I'd *like* to believe you're correct, but what you're saying doesn't match what I know about the real-world performance of EVs.

You are confusing design attributes of the system with the inherit characteristics of the motor.

Robert

chacha wrote:

I may be wrong in some respect, but I think the power of the [Tesla Roadster's] electric motor was not changed.

Okay, you may be right; I may have misinterpreted what I read about how the motor has been "enhanced". From a Motor Trend website article about the Tesla Roadster's "drivetrain 1.5":

Instead of achieving their original acceleration bogey via a two-speed tranny, they're simply beefing up the motor's power by enhancing the PEM (Power Electronics Module) and adding an advanced cooling system to the motor. Folks who are delivered early cars with the interim hardware will be called in (coincident with the production increase) for an update to the latest spec, free of charge. What isn't clear is whether this hardware swapping will include a new, cooling-enhanced motor as well, or instead see a client's existing motor somehow retrofitted.

It's not clear to me whether or not the motor was redesigned to produce more horsepower.

---

Robert's right all the way. It's really a performance concern. Lots of hard accel/decel gives higher losses (but great torque) at low motor speeds. This is simply a matter of how motors work. At the instant a stopped motor get's current, that current is highest until it builds a field, then that field has it's worst losses until CEMF is built by motion of that field in relation to other conductors. The idea behind dual speeds is to allow a higher torque at the wheels for a longer duty cycle by getting the RPM up at a lower speed. It's really only a problem for motor heating on repeated ramps up and down. Since this is a performance car, they had to consider this. An efficient EV of much lesser (and more reasonable) power will not require frequent hard torque at low speed unless the pilot is a teenager in which case, the electronics will protect the equipment.

---

Lensman wrote:

Y_Po wrote:

I will repeat, wires in transformer are insulated and AC motors have no advantage over DC motors, in fact they are heavier and less efficient at low speed

Perhaps you're right-- that a DC motor at low speed can be made slightly more efficient than an AC motor. The advantage of the properly engineered AC motor is that it can achieve high efficiency over a wide range of speeds-- which the DC motor can't. That's why a DC motor used to propel an EV needs a transmission, while a properly designed AC motor doesn't.

Now, as I've said, this applies only to inboard motors. For in-wheel motors, physical constraints require different engineering, and therefore different types of motors. It's my understanding that DC *can* compete, or may even be superior, for those applications.

But in-wheel motors are considerably more expensive, at least currently. Will we ever see an affordable "family sedan" driven by in-wheel motors? Only time will tell.

At present, for highway-capable EVs, AC rules.

Sigh you are both right or both wrong depending on your point of view.

Quick motor type overview

Series wound DC
+ Inexpensive, rugged. Very, very common in electric vehicles.
- Uses brushes
- Regen very difficult, electrical braking usually done via plugging.

PM DC
+ Common in smaller vehicles
+ Easy to speed control
+ Regen is easy
- Doesn't scale well
- Can be sensitive to temperature and shock, something that has improved with newer magnets.
+ Doesn't need to generate a field so usually more efficient
- Fixed field strength

Shunt wound DC, uses a separate low current winding for the field.
+ Control can be as simple as that for PM DC along with all its benefits
+ Scales better than PM
+ Currently replacing series wound motors in electric vehicles as the motor of choice.
+ Because field is generated it can be strengthened to increase torque or weakened to increase speed. A wider variety of speed torque charaqcteristics are available at run time

Induction AC
+ Very rugged
- Speed control needs to be a lot more sophisticated to match drive frequency with motor speed
+ Like shunt wound motors the two interacting fields of the motor are generated independently allowing wide selection of torque speed curves.

PM AC AKA Synchronous AC AKA BLDC AKA ECPM Motor. Note that these are definitely AC motors, not DC motors
+ Similar to PM DC motor in performance characteristic but requires slighty more complex control.
- Magnets

Switched Reluctance
+ Very Very rugged
- Even more complex to control

Currently, shunt wound motors are replacing series wound in EVs, with AC motors replacing shunt wound motors. European EVs appear to have largely skipped the shunt wound motors, moving to AC earlier than North America.

BLDC hasn't seen much use in EV except in automobile because the present controllers and especially motors have too high a premium. Also by the time you have the power electronics for a PM AC you have all the power electronics to support induction AC, you just need the sophisticated control to use the cheaper, more flexible, more rugged motor.

There's not a lot to choose between the motors on efficiency grounds although the PM motors will be more efficient.

Robert

Robert wrote:

You are confusing design attributes of the system with the inherit characteristics of the motor.

Thanks for that, Robert. Some Googling shows you are correct:

The Tesla Roadster has regenerative braking linked to the accelerator. The regen braking is designed to emulate a manual transmission car with an ICE. When you lift off the accelerator the car slows down. The amount of deceleration is proportional to the amount you lift off the accelerator and regen is stronger in 1st gear than in 2nd gear (again to emulate a standard transmission car).

I had read some complaints from owners or prospective owners of Roadsters who wanted to be able to control the degree of engine braking. (Altho apparently that's just *simulated* engine braking.) At least one comment I read said that Tesla would need to provide a small amount of current to the motor to allow "coasting", to counter the effect of normal engine braking.

I guess whoever wrote that was wrong, then. Well, this is why I participate in this forum: To learn things! So thanks again, Robert.

And apparently the Tesla Roadster 1.5 drivetrain doesn't have a clutch, either.

---

Innovator wrote:

First, there are a variety of DC and AC motor types. Two main catagories of DC are brushed and brushless. Brushless DC motors are actually AC motors with built-in electronics so they can be powered from a DC supply.

Generally only small BLDC motors have the electronics built in. Something that's possible but rare for other motor types.

Innovator wrote:

Brushed DC motors are mostly obsolete and only used by ammature conversions.

Brushed motors are common (I'm not certain they are still dominant) in commercial EVs, just not in automobiles.

Innovator wrote:

It is possible to tow an EV since the motor just spins with the wheels.

Yes

Innovator wrote:

The controller must be designed to accomodate the high voltage produced in this situation,

The voltages in this situation are no higher than the voltages used to drive the motor to reach the speed it is being toed at. For induction AC and shunt or series wound DC it won't even be that high since no field is being actively generated and any regen voltage will be the result of residual magnetism and therefore. Although in AC it is possible to use residual magnetism to bootstrap a generator.

Innovator wrote:

but is no big deal to add MOVs or transorbs which are passive and the controller does not have to be powered up.

First MOVs and transorbs simply won't absorb the energy for such a task. Second the problem is not excess voltage but that regen in this case powers up the controller's DC bus through parasitics. The controller has to be designed to be passive in the face of this or ...

Innovator wrote:

Another solution is a relay disconnect.

that'll work too.

Robert

Lensman wrote:

At least one comment I read said that Tesla would need to provide a small amount of current to the motor to allow "coasting", to counter the effect of normal engine braking.

That would be true if the mechanical drive had a lot of friction. Otherwise the strength of regen braking when the throttle is released in a tuning parameter in the controller. It's not so much adding current to the motor as reducing the current taken from the motor. Whether the engineers let the unwashed masses at it is an open question, but it would be possible to let customers vary this over a range.
Robert

Y_Po wrote:

I will repeat, wires in transformer are insulated and AC motors have no advantage over DC motors, in fact they are heavier and less efficient at low speed

That's a lot less true than it once was. With the advances in control techniques AC motors can pretty well match DC motors right down to near zero speed. At zero is still difficult but a few drive manufacturers claim success, notably ABB with their DTC drives.

Robert

Robert wrote:

Currently, shunt wound motors are replacing series wound in EVs, with AC motors replacing shunt wound motors. European EVs appear to have largely skipped the shunt wound motors, moving to AC earlier than North America.

Robert, every time we start debating AC versus DC for EV motors, you start talking about all the different motor types. Well, maybe there are a wide variety of motor types, but on the forum for California EV owners which I belong to, nobody gets into all these fiddly details. The *important* consideration is: Do you need to supply AC or DC power to the motor?

For *practical* purposes, an "AC motor" is one which needs AC power provided to it, and a "DC motor" is one which needs DC power provided to it. What happens to the current inside the motor itself is pretty irrelevant when it comes to choosing components to build a car out of, or choosing which *type* of components to design for a specific EV. I understand that for an electrical engineer like you, what goes on inside the motor is very important. But for most of us, it's the power, efficiency, and performance of the motor that matters. Exactly what happens to the current inside the motor is not. So from my viewpoint, you're just confusing the issue.

---

Lensman wrote:

Robert wrote:

Currently, shunt wound motors are replacing series wound in EVs, with AC motors replacing shunt wound motors. European EVs appear to have largely skipped the shunt wound motors, moving to AC earlier than North America.

Robert, every time we start debating AC versus DC for EV motors, you start talking about all the different motor types.

Probably because there is repeated ignorance of motor behaviour portrayed in these discussions.

Lensman wrote:

Well, maybe there are a wide variety of motor types, but on the forum for California EV owners which I belong to, nobody gets into all these fiddly details. The *important* consideration is: Do you need to supply AC or DC power to the motor?

Maintainence, power curve and inherit controller complexity. Not what I consider a fiddly detail. I did leave out the more subtle sensored vs sensorless control,

Lensman wrote:

For *practical* purposes, an "AC motor" is one which needs AC power provided to it, and a "DC motor" is one which needs DC power provided to it.

Which, if you look carefully at what I wrote is how I characterised them. I agree with you on this point.

Lensman wrote:

What happens to the current inside the motor itself is pretty irrelevant when it comes to choosing components to build a car out of, or choosing which *type* of components to design for a specific EV. I understand that for an electrical engineer like you, what goes on inside the motor is very important. But for most of us, it's the power, efficiency, and performance of the motor that matters. Exactly what happens to the current inside the motor is not. So from my viewpoint, you're just confusing the issue.

You should note that I was talking about power, efficiency, cost and performance in those short outlines.

Robert

Lensman wrote:

But for most of us, it's the power, efficiency, and performance of the motor that matters.

Ah, but that is not an AC vs DC question rather that's a which particular type of AC or DC question. It is not possible to say either AC or DC is superior, even for a particular application.

Robert

Robert wrote:

It is not possible to say either AC or DC is superior, even for a particular application.

All the present-day auto makers designing BEVs who are using inboard motors are using motors powered by AC. True or false? Presuming that's true, I'd call that pretty solid evidence that AC is superior.

Contrariwise, so far as I know, those few who are using in-wheel motors are all using DC power.

As I said, Robert: You appear to be trying to confuse the issue.

---

Y_Po wrote:

Wow, Lensman you tell me that?
You forgot your "wires in transformer are not insulated" incident?

Y_Po, I think it's endlessly amusing that you keep reminding me of the mistake I made there, as though you're "scoring points". Making mistakes is how we learn things. Since you virtually never admit to any of the mistakes you make, you virtually never learn anything.

So that's a point for me... not for you.

Y_Po wrote:

AC motors have no advantage over DC motors

For certain applications they certainly do! Now, are you capable of learning from your error here?

---

Lensman wrote:

Y_Po wrote:

Wow, Lensman you tell me that?
You forgot your "wires in transformer are not insulated" incident?

Y_Po, I think it's endlessly amusing that you keep reminding me of the mistake I made there, as though you're "scoring points". Making mistakes is how we learn things. Since you virtually never admit to any of the mistakes you make, you virtually never learn anything.

So that's a point for me... not for you.

Y_Po wrote:

AC motors have no advantage over DC motors

For certain applications they certainly do! Now, are you capable of learning from your error here?

Claiming that wires in transformer are uninsulated is not just a 'mistake'. It demonstrates a fundamental ignorance of the technology you pontificate upon.

Not long ago, you demonstrated an equal level of innocence of what parts of a CRT monitor one should avoid touching if you wish to continue living.

There are other equally egregious errors you have displayed, which clearly indicates that your grasp of even basic physics is extremely tenuous.

I suggest you stop posting and indulge in a little study instead. When you have done so, I'm sure you will have a very valuable contribution to make.

Please believe me when I tell you I mean this post kindly.

Tec wrote:

Claiming that wires in transformer are uninsulated is not just a 'mistake'. It demonstrates a fundamental ignorance of the technology you pontificate upon.

Tell us again, Tec, about how diesel motors are more efficient than electric motors. Or how diesel fumes aren't toxic. Or how toast fumes are.

Just like Y_Po, you refuse to learn from your mistakes.

Who's the "class clown" on this forum, Tec? Hint: It ain't me.

---

Robert wrote:

Innovator wrote:

The controller must be designed to accomodate the high voltage produced in this situation,

The voltages in this situation are no higher than the voltages used to drive the motor to reach the speed it is being toed at. For induction AC and shunt or series wound DC it won't even be that high since no field is being actively generated and any regen voltage will be the result of residual magnetism and therefore. Although in AC it is possible to use residual magnetism to bootstrap a generator.

Innovator wrote:

but is no big deal to add MOVs or transorbs which are passive and the controller does not have to be powered up.

First MOVs and transorbs simply won't absorb the energy for such a task. Second the problem is not excess voltage but that regen in this case powers up the controller's DC bus through parasitics. The controller has to be designed to be passive in the face of this or ...

Robert

Transorbs and MOVs are just to limit high voltage spikes and prevent damage to the electronics. They do not provide any appreciable load to the motor. You are quite right when you say that the voltage generated by the regen is not expected to go very high, but it is partly a function of the load, and with no load there is the possibility of high transients.

---

All in-wheel motors that I know of PMLs Hi-Pa series, tm4, Mitsubishi iMEV are all AC permanent magnet motors.

---

As I said, Robert: You appear to be trying to confuse the issue.

Disagree, he's pointing out that an "AC vs. DC" argument is an oversimplification.

---

Robert wrote:

Brushed motors are common (I'm not certain they are still dominant) in commercial EVs, just not in automobiles.
Robert

Interesting distinction between EVs and automobiles.

I agree if you are refering to EVs as low speed NEVs, fork lifts, and golf carts. This is true mostly because they are old designs and there is no compelling reason to re-design.

Brushed motors like the Warp series are still in demand because of the simple control, but as more people realsize that there are low cost AC controllers on the market, the demand for these motors will decline.

---

Lensman wrote:

Robert wrote:

It is not possible to say either AC or DC is superior, even for a particular application.

All the present-day auto makers designing BEVs who are using inboard motors are using motors powered by AC. True or false? Presuming that's true, I'd call that pretty solid evidence that AC is superior.

Contrariwise, so far as I know, those few who are using in-wheel motors are all using DC power.

As I said, Robert: You appear to be trying to confuse the issue.

see http://en.wikipedia.org/wiki/Brushless_DC_elect... for a reasonable primer on DC and AC. I don't think Robert it trying to confuse the issue, just sharpen the discussion being held by the non-technical.

Lensman wrote:

grizz wrote:

Why would an electric motor provide braking if it is not connected to its load. All motors I have ever seen turn freely with no load attached.

If so, then why does the Tesla Roadster perform significant engine braking when you take your foot off the accelerator?

I'd *like* to believe you're correct, but what you're saying doesn't match what I know about the real-world performance of EVs.

My understanding of it is that a lot of people complained about driving EVs because they do not do this by default. The tesla runs lightly in reverse when you take you foot of the accelerator to simulate an ICE. It sounds stupid, but it is supposed to be quite disconcerting to drive an EV without this feature.

---

PaulF wrote:

Lensman wrote:

grizz wrote:

Why would an electric motor provide braking if it is not connected to its load. All motors I have ever seen turn freely with no load attached.

If so, then why does the Tesla Roadster perform significant engine braking when you take your foot off the accelerator?

I'd *like* to believe you're correct, but what you're saying doesn't match what I know about the real-world performance of EVs.

My understanding of it is that a lot of people complained about driving EVs because they do not do this by default. The tesla runs lightly in reverse when you take you foot of the accelerator to simulate an ICE. It sounds stupid, but it is supposed to be quite disconcerting to drive an EV without this feature.

An electric motor does not provide braking when there is no load. The load is controlled by the controller just like the power applied is controlled during acceleration. The load is the battery, and the controller during re-gen controls how much re-charging takes place.

It would be possible to have the accelerator pedal only provide acceleration and have the brake pedal control re-gen. I am not sure why this is not done, but suspect it is a safety issue, as the brake pedal must operate the mechanical brakes in the approved manner.

The alternative configuration is to have the accelerator pedal control both acceleration and regen. Most EV drivers say they like this configuration as they do not have to move their foot to the brake pedal to slow down, except when going very slow or require rapid deceleration.

The Tesla does not "run lightly in reverse". It does however have a "creep" function that simulates how an ICE with an automatic transmission creeps forward when you take your foot off the brake in drive.

---

Lensman wrote:

Robert wrote:

It is not possible to say either AC or DC is superior, even for a particular application.

All the present-day auto makers designing BEVs who are using inboard motors are using motors powered by AC. True or false? Presuming that's true, I'd call that pretty solid evidence that AC is superior.

Auto makers? I don't know, wouldn't surprise me. Certainly the ones I've looked at have been BLDC or Induction AC. That's as much or more a function of the voltage they've chosen to operate at as any inherit superiority though.

Lets walk through the characteristics

Efficiency
BLDC and PMDC both have a slight efficiency advantage sue to not having to generate a field. So no advantage to AC or DC.

Torque curve flexibility
Both induction AC and shunt wound DC can tailor these on the fly. No advantage to AC or DC.

Voltage
At high voltages (more than a few hundred volts) brushes are likely to be an issue. Advantage at high voltage goes to AC.

Maintenance/ruggedness
AC motors do not have brushes to replace (although current can end up taking a path through the bearings in some circumstances leading to reduced life). Induction motors have a simple construction. Advantage AC induction.

Potential long term cost
PM motors (AC and DC) are limited by magnet cost. Long term the simpler AC induction rotor construction should result in a cheaper motor. Advantage AC Induction

Current Cost
Series wound cheapest, followed by shunt wound, PMDC, AC induction, BLDC in that order from my pricing.

Regen
Other than series wound all other types regen easily. No advantage AC or DC

Power Section
Series wound will be treated as base. Call it 1P + contactors. PMDC is 2P, shunt wound 1.05P, BLDC and Induction AC 1 1/2P. Advantage shunt wound DC

Control complexity
Roughly in order series wound, PMDC, shunt wound, BLDC, Induction AC. Advantage DC

One small additional note. I put switched reluctance in the list because it is potentially cheaper and more rugged than induction AC. But that is a long term potential. Rather like fusion it so far has always been the technology of the future.

Lensman wrote:

Contrariwise, so far as I know, those few who are using in-wheel motors are all using DC power.

I thought they were mostly BLDC.

Robert

Innovator wrote:

Robert wrote:

Brushed motors are common (I'm not certain they are still dominant) in commercial EVs, just not in automobiles.
Robert

Interesting distinction between EVs and automobiles.

I agree if you are refering to EVs as low speed NEVs, fork lifts, and golf carts.

Also scissors and booms, AGVs, GSE, Mining locomotives, Die handlers and specialty equipment(1). The real electric vehicle market as opposed to the niche market the electric automobiles currently are :).

Innovator wrote:

This is true mostly because they are old designs and there is no compelling reason to re-design.

Also as you note simpler control and cost. There is also a depth of experience in service and maintenance. There will probably be a gradual move towards AC but I expect the smaller and lower volume vehicles especially to be slow to change.

Robert

1 - Well arguably the last two are specialty vehicles.

Innovator wrote:

Transorbs and MOVs are just to limit high voltage spikes and prevent damage to the electronics. They do not provide any appreciable load to the motor. You are quite right when you say that the voltage generated by the regen is not expected to go very high, but it is partly a function of the load, and with no load there is the possibility of high transients.

It doesn't matter what the load is, a motor under regen will not develop a higher voltage than to took to run it at the speed it is running, in fact the larger the load the lower the voltage.

Short term spikes do occur when driving (and during actively controlled regen) but those are a result of parasitic inductance in the power layout. That's very hard use for a MOV although an appropriately sized transorb might handle it. There are other, arguably better solutions but a transorb could be used for that.

Robert

Robert wrote:

Y_Po wrote:

I will repeat, wires in transformer are insulated and AC motors have no advantage over DC motors, in fact they are heavier and less efficient at low speed

That's a lot less true than it once was. With the advances in control techniques AC motors can pretty well match DC motors right down to near zero speed. At zero is still difficult but a few drive manufacturers claim success, notably ABB with their DTC drives.

Robert

What you call AC is actually DC.
The only AC which can be called AC is induction type AC.

---

Y_Po wrote:

Robert wrote:

Y_Po wrote:

I will repeat, wires in transformer are insulated and AC motors have no advantage over DC motors, in fact they are heavier and less efficient at low speed

That's a lot less true than it once was. With the advances in control techniques AC motors can pretty well match DC motors right down to near zero speed. At zero is still difficult but a few drive manufacturers claim success, notably ABB with their DTC drives.

Robert

What you call AC is actually DC.
The only AC which can be called AC is induction type AC.

I was referring to induction AC in all of that.

Robert

Robert, thanks. As an EE with no electric motor experience you have given me a weeks worth of terms to google and brush up on.

(brush... ha ha)

---

Robert wrote:

Y_Po wrote:

Robert wrote:

Y_Po wrote:

I will repeat, wires in transformer are insulated and AC motors have no advantage over DC motors, in fact they are heavier and less efficient at low speed

That's a lot less true than it once was. With the advances in control techniques AC motors can pretty well match DC motors right down to near zero speed. At zero is still difficult but a few drive manufacturers claim success, notably ABB with their DTC drives.

Robert

What you call AC is actually DC.
The only AC which can be called AC is induction type AC.

I was referring to induction AC in all of that.

Robert

OK, then you are wrong, AC are heavier and there is no way they can do low speed well.

And again, AC means induction type. The rest what some call AC are DC.

---

farmEEr wrote:

Robert, thanks. As an EE with no electric motor experience you have given me a weeks worth of terms to google and brush up on.

If you are interested in AC (particularly induction) the following additional terms may be helpful

Sensorless
Flux Vector
Field Oriented Control
Direct Torque Control
Parke Clark Transform

With any luck those are all spelled correctly :)

Robert