Re: Motor controllers, RFI & stuff (esoteric question).

["Jon \"Sheer\" Pullen" <sheer@brassrat.net>]

> I'm going to show another gap in my (our) experience here, we were having
a
> 'how about' session, and the subject of RFI came up, leading to this
> question: knowing how much RFI is generated and radiated by normal chopped
> DC motor controllers, what would be the effect of running a motor off DC
> provided from a controller that switched fast enough to provide "pure" DC
> from 0V through to pack voltage? The fact of feeding the motor with

'switched fast enough'?

By definiition if it is switching the power on and off, it is AC. MIght be
uberfast AC, but it's still AC. And, in fact, radiating in the megahertz
range instead of 100 khz might make a EV a considerably less welcome
neighbor. Also, with most silicon I'm familiar with, there is a short period
when switching it where it operates in linear mode, and therefore is less
efficient - so the faster you switch, the less efficient your controller is
and the more heat you have to get rid of. You have to balance this against
not wanting to radiate any audible noise from your inductor/motor.

There's another option, but our silicon isn't really up to it yet. Instead
of running the controller in saturated mode, it would be theoretically
possible to run it in linear mode - hence giving the 'pure DC' you wanted -
the problem is that these devices are much less efficient [and therefore run
much, much hotter] when run in linear mode - compare a thousand watt
amplifier's heat loss with a hundred thousand watt controller and you'll see
what I mean. Hence, you'd end up with a hotter, less efficient controller,
and very little if any efficiency gain in the motor - what's the point?

> Do the pulses from a conventional controller provide additional torque
over
> what you would have at steady DC? ie if a controller from 120V was at 20%

No.

> While it is technically possible to build a high power switching regulator
> (done all the time in welding equipment), it's not done in EVs, so for
what
> reason? Purely economics? or is it a case of "its always been done this
> way" and the advent of fast enough powerful enough devices has been so
> recent that no-one has done it yet?

In essence a DC controller _is_ a high power switching regulator - the
difference is that in the EV world, you use the motor as the inductor
because hey, it's already there, free inductor! ;-) Maybe one of the more
guruish on the list will weigh in with what the possible advantages of
having a outboard inductor are, but I can already tell you the
disadvantages - bigger controller, heavier, and my hunch is very little
efficiency gain.