On 10/5/2012 9:08 AM, Electron wrote:
> At 12.17 2012.09.03, you wrote:
>> Joe, decide on power or torque limiting (these are 2 different things), =
then
>> implement it. It is a closed loop so you just do Pwmcontrol =3D (limitin=
g)?
>> Demand: Demand - LimitingSignal; Cycle by cycle current limiting is anot=
her
>> thing, and is actually a good idea, but that requires that you see the
>> individual current pulses. It may be, that individual pulse current limi=
ting is
>> all you really want, and that is best done in hardware alone imho.
>
> By the way, if I'm not mistaken, DC brushed motors are "constant power ov=
er the
> RPM range" (at an approximation), or in other words "exponentially decayi=
ng torque".
>
> Well, what kind of electric motors are instead "constant torque over the =
RPM range"?
> (of course at an approximation even here, as such an engine if it existed=
 would have
> infinite power at infinite RPM if let run freely ;) ).
>
> Thanks,
> Mario
>
Well, I don't need to get all that technical in my application.  I=20
thought I found a reference saying that I'd get max power (and / or=20
torque?) at about 1/2 the no load max rpms.   But, with a simple goal of=20
simply not blowing fuses while getting as much power as I an out of the=20
motor, I just put in a high side current sensor monitored by the PIC=20
that PWMs the motor.  It samples every 5 mS and if over limit, kills the=20
PWM entirely for the next 5 mS.  The current sensor output is subject to=20
some offset dependent on motor current, due to the sensor being in the=20
battery compartment and the monitoring PIC is in the motor control=20
compartment.
I put about 12 miles on the eBike today, and it definitely had less=20
power up hills toward the end than it did at the start.  The battery=20
voltage was about 25 at start and about 23 at end.  Under load toward=20
the end, it was under 20V if I can believe the 5 LED gauge I set up but=20
did not calibrate.  And I'm not sure if the throttle circuit is still=20
within the limits established empirically.  I need a way to get at the=20
data, I think it's a great excuse to hook up a Roving Networks bluetooth=20
module that's been waiting for a project.

Joe W

--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.