Re: [PSUBS-MAILIST] Re: Motor controller...

[John Farrington <jrf@prismnet.com>]

Hi Ken, nice talking with you this evening.

I did a sketch of a modification to your schematic.  This is a start at
modifying your original schematic in an attempt to add reverse capability
via joystick, and getting rid of the relay.

It's available here:
http://www.prismnet.com/~jrf/circuit.jpg

It is a possible implementation of the H-bridge that we talked about.
Would you mind glancing at it to see if I've set it up right?

The 'signal' lines would be the two lines where the PWM signal would
come in to decide forward/reverse thrust of the motor.

I'm not sure of a good way to decide which of the two lines to send the
signal down.  Perhaps some sort of comparator linked to the 1K pot?
(Not sure, really kinda out of my league here)  It would also be nice
to have a bit of a 'deadband' in the joystick so that the joystick
would have to be moved say 10% in either direction before it started
sending the PWM signal.

And I guess that there'd have to be some modification to the over-current
protection circuitry because now we can have current traveling in either
direction through TX1.

Also, I've drawn in just 1 IXFK90N20 at each corner of the "H".  These
are rated at 90 amps.

In your original schematic, you've got 4 IXFH58N20 FETs, which are rated
at 58 amps each, giving you a capacity of 232 amps.  Would just a single
one have done the job for you (but run hotter), or is it a good idea to
have the extra capacity?

(I'm trying to figure out if we can get by with 4 of the IXFK90N20s, 
rather than 8 of them.  They have an Rds of .02, versus an Rds of .04
for the IXFH58N20s, so the power loss would be the same .02*2=.04
(twice as many serially in the circuit than in your orginal schematic -
as you said)).

For everyone else, this may sound like Greek, and it was to me too, prior
to my conversation with Ken tonight.  He knows his stuff!


-John