Mr. Sharkey via EV wrote:
What we seem to have deduced is that the controller and motor are a
matched pair. It seems without doubt that the controller manages the
voltage delivered to both the armature and the fields, otherwise the
"reverse switch" wouldn't be workable.
Yes, that sounds likely.
I'd image that the sequence would be something like:
1) accelerator released: controller ready, no field or armature
2) accelerator depressed: full field, armature current limited, probably
ramping up.
3) accelerator depressed further, full armature current, full field.
This would occur at what I'm calling "idle" speed.
4) at some point, pressing the accelerator further results in the
beginning of field weakening while maintaining full armature.
5) release accelerator partially: full armature current, restore field
current to a greater level. Regenerative braking proportional to pedal
position.
6) completely release accelerator: Well, this is a bit of conjecture.
Dropping the armature at any point during deceleration would result in
the regenerative braking going away all at once, which might be
dangerous if the driver wasn't expecting it, so I'd say that the
controller maintains the armature current until the vehicle has slowed
to the point where regen is weak or nil, probably the "idle" speed, then
it ramps down or cuts the armature current.
That's a good guess. Though, a sepex controller is capable of very
sophisticated motion control. You don't see sepex in fork lifts very
often; but you do find them in more sophisticated EVs, and applications
like cranes and elevators, where the operator wants precise control of
position, torque, and speed.
My old sepex controller was very crude; but it worked! :-) The armature
had 4 steps; off, 36v with series resistor, 36v direct, 72v direct. The
field had a big rheostat in series to the pack (36v or 72v).
When the accelerator pedal was released, the pack was switched to 36v.
The field rheostat was 0 ohms, so full 36v field. As you pressed the
pedal, cam switches gave the armature 36v with resistor, then 36v
without resistor. Above that, the rheostat began increasing resistance
to weaken the field and speed up the motor. The field got to minimum
resistance near full throttle. At full throttle, a cam switched the pack
from 36v to 72v to get "full warp speed".
A characteristic of this setup is that it tried to be a constant-speed
drive. If I held the pedal in one position, the car tended to go at the
same speed, drawing a lot more power uphill, and doing regen down hill.
I didn't want to suddenly push the pedal to a new position, because the
motor would "fight like hell" to get to the new speed as quick as it
could. The only thing preventing me from breaking drive shafts or
getting my teeth planted in the steering wheel was that it was a
relatively small motor (70 lbs; rated 30v 500a) and a heavy vehicle (a
1974 Datsun pickup with a dozen golf cart batteries).
If, like Lee suggests, it might be a low voltage field, then the
controller might have a buck/boost function for the field, which would
complicate our armchair diagnosis.
My guess is that they wound the field for some fraction of pack voltage,
so they didn't need a buck/boost controller. They could get (say) 4x
field by applying 120v to a 30v field winding.
Your controller just has to be aware of how *long* it can over-voltage
the field before heating becomes a problem. The field has a lot of mass;
it can stand large over-voltages for many minutes, and there is usually
a blower that runs all the time for cooling it.
I do remember when I was researching the Siemens 1GV series motors last
year during my lithium conversion, I ran across some documentation that
seemed to show some series field windings along with the shunt/sepex
field. The compound field arrangement might be the key to having
stall/low RPM torque available so the motor doesn't need to idle.
Yes. The most sophisticated applications for big DC motors are normally
compound (multiple series and shunt field windings). You can get just
about any imaginable characteristic just by careful choice of which
windings are powered. There are also interpoles, which add even more
possibilities. But that's a whole 'nother topic.
Thinking about it, it's entirely possible that the SCT developers went
with the full-armature/idling motor both because they didn't want to
have to build controllers that could handle the armature current, but
also because that thought that a car that "idles" would be more
intuitive for drivers used to ICE vehicles.
Well, when people convert ICE's with automatic transmissions, they often
*do* need to keep the motor idling, just to keep they transmission
pumped up and working. And people have come to expect cars to "creep".
Earlier this month the local utility contacted me about entering my car
in the town's annual Spring parade. I had to decline because idling at
1,800 RPM in first gear results in a ground speed of 9 MPH.
The simple fix is to somehow drop the voltage to the armature. You can
make a shunt or sepex motor idle at 10 RPM! Full field voltage, and
something like 2v on the armature. :-)
--
If you're not stubborn, you'll give up too soon. If you're not flexible,
you'll pound your head against the wall and miss a different solution.
(Jeff Bezos)
--
Lee Hart, 814 8th Ave N, Sartell MN 56377, www.sunrise-ev.com
_______________________________________________
UNSUBSCRIBE: http://www.evdl.org/help/index.html#usub
http://lists.evdl.org/listinfo.cgi/ev-evdl.org
Please discuss EV drag racing at NEDRA (http://groups.yahoo.com/group/NEDRA)
