Re: (ET) Speed controller

["David Roden (Akron OH USA)" <roden ald net>]

I've looked into the 4QD controllers and I think they are probably a good 
candidate for the ET.  I'm waiting to learn how yours works before taking 
the plunge.

Here are some of the things I've been thinking about.  Since I haven't 
tried any of these, take them as random ramblings.  If you know more 
about this stuff than I do (which wouldn't take much), please don't 
hesitate to comment or correct.

I can see a couple of ways to handle the field.  One is, as you suggest, 
to connect it to full battery voltage through the main contactor and 
leave it.  However I don't think I'd want to leave it energized when the 
motor is not running.  So I'd want to to find a way to shut it down when 
the potbox is released, or better yet when the motor actually stops (so I 
can have regen braking).

Another way would be to simply parallel it to the armature.  I wonder if, 
with the field paralleled to the armature, I'd still get regen.  Maybe 
the residual magnetism would keep the current flowing and therefore 
maintain the field strength.

I don't know which would work better; I'd experiment with both.

With either of these schemes, I would lose the higher GE speeds achieved 
by field weakening.  To get the higher speeds back, I could (1) retain 
the GE's speed lever (linking the potbox to it), card 3, and FW relay; 
(2) add a second PWM controller to control the field; or (3) increase the 
pack voltage to 48 volts or more.

Curtis used to say their controllers could be used with shunt motors if 
the field were paralleled with the armature.  The reason I'm reluctant to 
try this is that I've heard recently that newer Curtis controllers act 
strangely if the motor inductance is too low -- it reportedly fools the 
current limit so the maximum current isn't available, and can also cause 
jerky starts.  I believe most shunt and sep-ex motors (as in the ET) have 
lower inductance than the series motors usually used in golf cars. 

To answer the question of "what controls the controller," any 5000 ohm 
potentiometer can do so.  Obviously you want a high quality, sturdy, well 
made pot rather than the cheap Chinese knockoffs sold at Radio Shack.  
Most controller makers will sell you a good quality pot enclosed in a 
rugged metal box, set up for use with an accelerator cable or linkage.  
Curtis has a complete potbox assembly that looks like a sewing machine 
pedal.

Some advice for those who want to try rolling their own.  Designing a 
controller for the currents involved in running an ET is not trivial.  
Unless you are an electronic engineer with considerable experience in 
electric vehicle control design, you are likely to smoke a lot of silicon 
before you have a reliable design.  In the end, you'll most likely spend 
more on sacrificial silicon than you would have on a good factory built 
controller.

If you really want to do it yourself, email Mark Hanson 
(mhanson valcom com).  He's an electrical engineer who used to work for 
GE's EV division before they essentially closed it down a year or so ago. 
Mark designed a controller for his own ET running at 48 volts (using 8-
volt golf car batteries), and he'll sell you the plans for a few bucks.  
That's a lot cheaper than exploding large quantities of high power 
silicon.  But I think you'll still find that you're money ahead to buy a 
mass-produced controller -- it's just a matter of finding the right one.

If you're looking for cheap, you might look into Wilde Evolutions' 
rebuilt Curtis controllers (http://www.Wilde-EVolutions.com/PriceList.htm 
-- page down to see the rebuilt controllers branded as "Wilde 
Evolutions").  If you go this route, please let us know how it works out.


David Roden - Akron, Ohio, USA
1991 Solectria Force 144vac
1991 Ford Escort Green/EV 128vdc
1979 General Engines ElectroPed 24vdc
1974 Honda Civic EV 96vdc
1970 GE Elec-trak E15 36vdc