Motor Controller Design: A Photo Journal!

AC, DC, amps, volts and kilowatt. It's all discussed in here
zeva
Senior Member
Posts: 422
Joined: Sat, 15 Dec 2007, 05:09
Real Name: Ian Hooper
Location: Australia
MSN: sigmunky@hotmail.com
Contact:

Motor Controller Design: A Photo Journal!

Post by zeva » Wed, 03 Dec 2008, 09:18

A minor update with pics of the waveforms, for interest:

http://zeva.com.au/tech/controller_design/#Dec3Update

My initial impression is that switching losses could indeed be a lot faster, but I'm looking forward to hearing what the EVTech guys have to say about the traces.
Last edited by zeva on Tue, 02 Dec 2008, 22:23, edited 1 time in total.
Ian Hooper
--
"Never doubt that the work of a small group of thoughtful, committed citizens can change the world." - Margaret Mead
http://www.zeva.com.au

zeva
Senior Member
Posts: 422
Joined: Sat, 15 Dec 2007, 05:09
Real Name: Ian Hooper
Location: Australia
MSN: sigmunky@hotmail.com
Contact:

Motor Controller Design: A Photo Journal!

Post by zeva » Tue, 09 Dec 2008, 09:16

Just finished testing the controller with 3x faster switching; seems to work fine. I'll put it in the MX5 again tomorrow for some "field testing", hopefully there'll be a significant improvement in efficiency!

http://zeva.com.au/tech/controller_design/#Dec8Update
Ian Hooper
--
"Never doubt that the work of a small group of thoughtful, committed citizens can change the world." - Margaret Mead
http://www.zeva.com.au

flux
Noobie
Posts: 1
Joined: Sun, 14 Dec 2008, 04:22
Real Name: Simon Bakker
Location: Gladstone QLD

Motor Controller Design: A Photo Journal!

Post by flux » Tue, 16 Dec 2008, 22:57

Looks very impressive, would it be possible to use IGBTs instead of the MOSFETs or would you have issues with the triggering of the IGBTs. The reason I ask is that I have a couple of 400amp 1200v IGBTs and could get an 800amp unit for a reasonable price. I'm wanting to build a 120v motor controller with about the same peak current as the one you have built.
Flemings Right-hand rule for the direction of induced voltage, Flemings Left-hand rule for the direction of force on the conductor

bga
Senior Member
Posts: 492
Joined: Mon, 01 Sep 2008, 19:27
Real Name: Bruce Armstrong
Location: Perth WA

Motor Controller Design: A Photo Journal!

Post by bga » Wed, 17 Dec 2008, 01:17

Hi Flux,

Yes, an IGBT module would work. The double IGBT units have good diodes in them for flyback. It'd be tidy, 3 terminals and bolts onto a heatsink.

The only problems may be:
Up to 800Amp IGBT modules get pricey, not prohibitive, though. At 1000Amps and above they are getting exotic. I would have reservations about paralleling large IGBT modules, although it should work.
The gate drive requirements are similar to MOSFETs.

For a lot of duty, IGBTs and MOSFETs perform similarly, although there are differences at the ends of operation.

MOSFETs aren't available in the giant modules that IGBTs are. MOSFETs parallel easily and large currents can be handled this way.
Many of the capable MOSFET devices are cheap and available off the shelf locally.

MOSFETs are restricted to about 200V and below for EV applications. The high voltage devices have too high an 'on' resistance to be practical.

At low-ish (typical running) currents, MOSFETS have lower losses than IGBTs. An IGBT has a 'fixed' forward voltage drop of about 2 volts, as compared to approximately 0.5V for a well sized MOSFET driver. This affects the cooling requirement on the controller significantly.

MOSFETS generally have lower switching losses then IGBTs.

IGBT based controllers can easily have maximum system voltages of up to about 800 volts without using power station-sized devices.

In summary, for systems below about 200V, MOSFETS are a better solution and above, IGBTs are a better solution. There is room for considerable crossover, however.

Hope this clarifies it.

BGA

PS Ian, I hope the gate driver fix has yielded a good result.


Last edited by bga on Tue, 16 Dec 2008, 14:48, edited 1 time in total.

User avatar
Hatchet
Groupie
Posts: 44
Joined: Mon, 16 Jun 2008, 02:49
Real Name: Brett Bishop
Location: Brisbane
Contact:

Motor Controller Design: A Photo Journal!

Post by Hatchet » Thu, 05 Feb 2009, 07:28

Ian, what is the update on the Zeva controller? Are you still developing it or are you busy with other stuff (I know you're a busy guy)? I'm interested as I'm due to purchase a controller shortly .... but I can hang off for a wee bit.

Brett

zeva
Senior Member
Posts: 422
Joined: Sat, 15 Dec 2007, 05:09
Real Name: Ian Hooper
Location: Australia
MSN: sigmunky@hotmail.com
Contact:

Motor Controller Design: A Photo Journal!

Post by zeva » Thu, 05 Feb 2009, 21:13

Hi Brett et al,

The webpage for the project is pretty up-to-date, i.e the latest prototype is still working OK but still not efficient enough for use in my MX5, which I have been using as the benchmark.

I haven't done much on this project in the last month or so as I often find I have to "take a step back" to rethink things! I'm starting to believe the MX5 may be an unreasonably tough benchmark - controlling an 11" motor direct drive involves much higher average currents than a typical 8-9" motor through a gearbox.

So the current prototype will soon be going into a different (more conventional) conversion we're undertaking here, it'll be interesting to see how much of a difference it makes. If it works well in those conditions, we could start building them on the understanding that they're not suitable for direct drive or large motors!
Ian Hooper
--
"Never doubt that the work of a small group of thoughtful, committed citizens can change the world." - Margaret Mead
http://www.zeva.com.au

User avatar
Hatchet
Groupie
Posts: 44
Joined: Mon, 16 Jun 2008, 02:49
Real Name: Brett Bishop
Location: Brisbane
Contact:

Motor Controller Design: A Photo Journal!

Post by Hatchet » Fri, 06 Feb 2009, 00:59

Thanks Ian, I've bookmarked that page. Having a 9" motor and a gearbox I'll wait it out. Lots of luck on the prototype.

Brett

HeadsUp
Groupie
Posts: 265
Joined: Sat, 28 Feb 2009, 09:13
Real Name: Mark W
Location: Sydney

Motor Controller Design: A Photo Journal!

Post by HeadsUp » Thu, 05 Mar 2009, 08:36

Hey Zeva

The hell with open source

get it perfect , scrub off the identification labels on all the components and then sell them bud.

cheers

User avatar
Taffy
Groupie
Posts: 238
Joined: Mon, 23 Jun 2008, 15:12
Real Name: Taffy Flynn
Location: Melbourne
Contact:

Motor Controller Design: A Photo Journal!

Post by Taffy » Sat, 09 May 2009, 15:44

Hey Ian,

i am curious as to how the testing went?

Cheers

zeva
Senior Member
Posts: 422
Joined: Sat, 15 Dec 2007, 05:09
Real Name: Ian Hooper
Location: Australia
MSN: sigmunky@hotmail.com
Contact:

Motor Controller Design: A Photo Journal!

Post by zeva » Sat, 09 May 2009, 19:15

Hey Taffy,

The latest prototype has been powering this EV for the last month or so with apparently no issues, so that's very encouraging.

Since the v0.4 build I had to add more cooling via a couple of computer type fans blowing down into the heatsink, which still suggests to me that it could do with being more efficient.. So next on the To Do list is having a crack at implementing synchronous rectification, which could better than halve losses/heating.. but is a reasonably significant modification.
Ian Hooper
--
"Never doubt that the work of a small group of thoughtful, committed citizens can change the world." - Margaret Mead
http://www.zeva.com.au

User avatar
Taffy
Groupie
Posts: 238
Joined: Mon, 23 Jun 2008, 15:12
Real Name: Taffy Flynn
Location: Melbourne
Contact:

Motor Controller Design: A Photo Journal!

Post by Taffy » Sat, 09 May 2009, 19:45

Good to hear that its working well, i am not even going to pretend to understand what synchronous rectification is.

Are you getting closer to the sale version?

zeva
Senior Member
Posts: 422
Joined: Sat, 15 Dec 2007, 05:09
Real Name: Ian Hooper
Location: Australia
MSN: sigmunky@hotmail.com
Contact:

Motor Controller Design: A Photo Journal!

Post by zeva » Sun, 10 May 2009, 19:17

I'm a bit of a chronic perfectionist when designing, so things are never 'finished'! But it's definitely encouraging to get some field testing done with no unpleasant surprises. I guess that means it's running well enough for "public consumption".. but then, it would seem a pity to ramp up production of a design that's only 99% efficient when synch. rect. could make it 99.5% efficient!? (1% of a car's energy is still a lot of heat to get rid of.)

Guess I'd better get busy..
Ian Hooper
--
"Never doubt that the work of a small group of thoughtful, committed citizens can change the world." - Margaret Mead
http://www.zeva.com.au

User avatar
Hatchet
Groupie
Posts: 44
Joined: Mon, 16 Jun 2008, 02:49
Real Name: Brett Bishop
Location: Brisbane
Contact:

Motor Controller Design: A Photo Journal!

Post by Hatchet » Mon, 11 May 2009, 00:00

Well done Ian, I've been hangin and waiting for your controller to go public. Give me a hoy when you want to sell one.

Squiggles
Senior Member
Posts: 742
Joined: Wed, 22 Apr 2009, 03:19
Real Name: Neil
Location: Newcastle NSW

Motor Controller Design: A Photo Journal!

Post by Squiggles » Mon, 11 May 2009, 05:13

Ian,
Ever thought of having a "gear change" input to the controller?
Idea being that you could have a gear lever with integrated lockout switch (like an auto). When you go to change gear the controller inhibits drive during gear change.

Actually in the simplest case of a two speed transmission a future controller could probably adjust engine speed to suit the alternate gear.

By the way I will be watching developments closely as well, love the idea of an Aussie controller. Being an electronics engineer tends to make me a little curious also! Mind you power electronics was not my thing, once you go above 12V you start letting the smoke out of all sorts of devices and no matter how hard you try you just can't get it back in.

zeva
Senior Member
Posts: 422
Joined: Sat, 15 Dec 2007, 05:09
Real Name: Ian Hooper
Location: Australia
MSN: sigmunky@hotmail.com
Contact:

Motor Controller Design: A Photo Journal!

Post by zeva » Tue, 12 May 2009, 02:12

Squiggles wrote: Ever thought of having a "gear change" input to the controller?
Idea being that you could have a gear lever with integrated lockout switch (like an auto). When you go to change gear the controller inhibits drive during gear change.
You could do this with most existing controllers using the power input pin. I usually recommend wiring the pot box's microswitch up to the power pin for redundant safety (i.e if the pot fails on, you can still power down the controller by taking your foot off the pedal) but you could also wire it in series with a switch on your gearstick to kill power when shifting.
Squiggles wrote: Actually in the simplest case of a two speed transmission a future controller could probably adjust engine speed to suit the alternate gear.
I have heard the idea suggested, some kind of sensor(s) in the gearbox to automatically match revs for the gear you're selecting - especially useful in conversions which got rid of the clutch, to make shifting easier/faster with only the synchros. Out of interest, I hear the new Nissan 370Z (petrol burner) does this with the manual version; matching engine revs to gears as you select them.

Of course personally I think in time electric drive will make gearboxes a thing of the past! With a good versatile electric motor, you just don't need more than one ratio. The GM EV1 had a single ratio, Tesla has changed to one, and a lot of converters are opting for "direct drive" now.
Ian Hooper
--
"Never doubt that the work of a small group of thoughtful, committed citizens can change the world." - Margaret Mead
http://www.zeva.com.au

Post Reply