Re: Permanent Magnet Motor Questions -> for E-Bike

[Lee Hart <leeahart@earthlink.net>]

PETER VANDERWAL wrote:
>     Yes what's your point?  I don't see anything in your statement to refute
> my original statements.

Perhaps we need to restate our assumptions to avoid arguing over
semantics and misunderstood assumptions. I am assuming we are talking
about using a typical modest-quality permanenent magnet motor; not
super-high efficiency, but not junk quality "toy" motors either.
Something like an automobile fan motor. We want to use them on a small
EV; something between a bicycle and a golf cart. Cost and simplicity are
important, so we want to restrict ourselves to inexpensive controller.
Since it is easy to get regen with a PM motor, we want to consider using
the motor as a generator without overly complicating the controller.

>     To reiterate:
>         1) Regen braking with a contactor controller will only be effective
> at relatively high speeds.

Are you assuming a) fixed gearing, b) fixed battery voltage, and c) no
series/parallel switching of motor(s) or batteries? If so, then
generated voltage is a function of speed, and you only get a reasonable
charging voltage at one speed. If everything is set up for motoring,
then I agree, you will only get regen at higher than normal speeds.

On the other hand, if you can switch your motors and/or batteries in
series for motoring and parallel for braking, you can get regen braking
down to half your motoring speed. If *both* batteries and motors can be
series/parallelled (two 12v batteries, switchable for 12v or 24v, and
two 12 motors switchable in series or parallel), then you have regen
braking down to 1/4 of max speed.

>         2) Dynamic braking can burn out the motor unless you have a constant
> current load and still won't brake all the way to a stop.

Of course I agree; any uncontrolled voltage or current can burn out
anything. This just means you need to include appropriate protective
devices, for *any* form of motoring or generating.

Dynamic braking means connecting the motor to a resistive load. You
don't recover the energy; you burn it up as heat, same as normal
friction brakes.

Braking force is proportional to current, so you can pick a load
resistor for any desired current at any desired rpm. At low speeds, the
calculated resistance may be less than the motor's own internal
resistance. However, I think you'll find that shorting a motor (minimum
resistance) gives you more braking than you know what to do with. It
won't bring you to a dead stop, but it will drag you quickly down to a
very low speed. 

The main problem is that the heat is being generated in the external
resistor and the motor. The types of motors we are talking about here
have negligible cooling, so dumping extra heat in the motor is bad.

>         3)  Plug braking can definitely burn out the motor unless you only
> use it a low speeds and it still won't brake all the way to a stop.

Again, any excessive unlimited voltage or current is bad. You just need
to limit it safely. Fuses, circuit breakers, resistor, and diodes are
not expensive or complex.

Plug braking means running the motor in reverse while it is spinning in
the forward direction. Since current is torque, reverse current is
reverse torque, i.e. braking. The simplest plug braking setup is a
double-pole double-throw switch that reverses the motor leads, and a
series resistor to limit the braking current. Such a setup supplies
smooth, even braking torque right down to zero rpm (braking torque is
1/2 as much at zero as it is at full motor speed).

But at zero rpm, it will keep right on generating reverse torque. That
means the vehicle stops, and starts accellerating in reverse. All it
takes is a diode to prevent motor voltage from reversing to prevent
this.

The other problem with plug braking is that you are consuming battery
power to stop as well as to motor. Thus it's rather wasteful of energy.

>         4)  A PWM 2 quadrant (or 4 quadrant) controller can brake all the
> way to a stop, but you have to be carefull because they buck the voltage up
> and current down and at low RPMs the motor current can get VERY high which
> could get dangerous to the motor.

Again, I agree. But by the time you get to a PWM 4-quadrant controller,
it's going to have some form of current limit to protect itself and the
motor.

>> 3. Are the brushes advanced to optimize its performance as a motor (and
>> thus penalize its performance as a generator)?
 
> Possibly, but this would be true when you are using the motor for regen too.
> Unless you expect us to put the brushes to a nuetral position and loose
> efficiency as a motor (where it spends most of it's time).

Brush advance doesn't have much effect on efficiency. The manufacturer
sets the brushes at neutral if the motor is going to be reversed (run in
either direction), or if it will be used as a motor and generator.

Advancing the brushes decreases brush arcing in the favored direction of
current flow, at the expense of worsening it when current flows in the
other direction. But arcing is not an issue with such small motors below
about 30v. 

But advancing the brushes changes the motor constant K (volts per rpm).
You'll have one value of K when current flows in a particular leadwire,
and a different value when it flows out. To see if your brushes are at
neutral or advanced, run the motor in each direction and see if the rpm
changes. If it runs at the same speed and same current, they are at
neutral. If different, they are advanced to favor one direction.

>> surplus 12v 12amp motors... shaft-to-shaft as a motor-generator...
 
>     So what was the voltage out of the generator when load down to it's
> rated voltage?   Possibly around 4V?

No-load is 12v 4amps in, 11v 0amps out. Full load for the motor is 12v
12amps in, 5v 8amps out. As you can see, they are fairly inefficient
(50-60%). Shorting the generator makes the motor pull over 30amps at 10v
(the most my test power supply could deliver).

> we are talking about using them as motors with regen capability.
> Ruining their performance as a motor to improve their generator
> performance isn't an option.

Sigh... If you don't like how it performs as a generator, optimize the
brushes for generator operation. This won't ruin its performance as a
motor. It will decrease motor performance, efficiency, and life
slightly, but *improve* generator performance, efficiency, and life.
It's a trade-off; optimizing one un-optimizes the other. Depending on
how you use it in each mode, the weakest link will ultimately kill the
motor. Optimizing motoring, and then using strong regen anyway will kill
it.
-- 
Lee A. Hart                     Ring the bells that you can ring
4209 France Ave. N.             Forget the perfect offering
Robbinsdale, MN 55422 USA       There is a crack in everything
(612) 533-3226                  That's how the light gets in!
<leeahart@earthlink.net>                Leonard Cohen