Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
Disclaimer: The Parliamentary Library does not warrant or accept liability for the accuracy or usefulness of the transcripts. These are copied directly from the broadcaster's website.
Getting the noise out of electric motors -

View in ParlViewView other Segments

Transcript

I saw the first ever electric motor two months ago. I saw it at the Royal Institution in London and
it was designed by Michael Faraday.

Greg Hynes: Right, well, and you're going to wonder how much has happened since he designed that
motor.

Robyn Williams: Well, I'm simply concerned with the fact that given 200 years roughly, nearly 200
years of electric motors, I would have thought that working out how and why they make noise and how
to reduce it would have been pretty well solved. But you're telling me there's lots to do.

Greg Hynes: It depends on why the noise is being created. It's possible to design and create an
electric motor well so it is quiet, but say you want to make lots of them and make them very
cheaply, then they're probably going to make noise because they're not so perfectly made. And so
our goal, if we can come up with the control systems that can compensate for imperfections in
electric motors, then we can make them more cheaply and deal with the imperfections later.

Robyn Williams: That's Greg Hynes in his lab in Darwin. He's keen on electric cars as well as quiet
motors, and he's making good progress in getting the noisy ones under control. But it's a
surprisingly tricky process.

Greg Hynes: The main goal is that we'd run the motor once and measured the torque coming out of it,
and this torque is like a force in a rotary direction.

Robyn Williams: A twist.

Greg Hynes: Yes, a twist. And the noise is actually high-frequency variations in that torque. So we
run it once and we measure the torque that's coming out, and we split that torque ripple up into
the components that was creating it. The main things we found that were creating this torque was
either an inaccuracy in the current measurement or there's a cogging torque, which is the tendency
of the magnets to want to connect to the steel poles of magnet rather than the copper windings. So
those, if we can measure the torque and then split it up into those separate components, we can
then compensate for those.

Robyn Williams: And it's worth economically doing so, is it?

Greg Hynes: We're not sure at the moment.

Robyn Williams: In other words, you said 'mass production', if you take the trouble to design them
so that they don't have those two deficits, presumably that accuracy costs money in design.

Greg Hynes: Certainly, yes, but it also depends on what sort of applications you're going to use
it. Say if you were to drive a vehicle, we're all quite familiar with the sound an engine makes and
expect it to be quite noisy. And so in that case it's not really worth trying to get the last bit
of noise out of it, whereas if you were to put it into maybe an air conditioning duct or something
where people traditionally do expect to have a nice quiet motor, then it will be worth putting that
effort in. So what we're looking at is to use an electric motor, such as the ones that were
originally developed here for solar cars which were optimised for different things, in particular
high efficiency and high torque, and see if they can be used in other applications that do require
this quiet.

Robyn Williams: I would imagine that if you take the noise out...the noise is somehow an indication
of inefficiency, and presumably that would have its own costs as well. So if you make it quieter
and more efficient surely that's a saving as well.

Greg Hynes: Possibly, but the gains to be made for that inefficiency, the analysis we've done at
least shows that that's very minimal, you're getting very small increases in efficiency. So I
wouldn't like to suggest that getting rid of the noise will make it that much more efficient,
that's not why we're after it.

Robyn Williams: Give me some examples of the sorts of things that you've done and what kind of
noise reduction can you hope to get?

Greg Hynes: The motor we were working on here for my research was originally for an electric bike,
and for that sort of application noise isn't too much of an issue, it's a lot quieter than many
things. So we were able to get...normally we were getting about 9% just to do the standard control
system, and we were able to get that down to about 1%. So, sort of a nine-fold reduction.

Robyn Williams: That's fantastic.

Greg Hynes: That was two things. That was both some of it, we got down to 3% just using traditional
methods for compensating, but those traditional methods needed very good information about the
parameters I was discussing, about the cogging torque and the accuracy of the current. So we were
able to get the further three-fold reduction by using our methods to get those accurate
measurements.

Robyn Williams: And what will it be used for?

Greg Hynes: Either you can look at it from the noise perspective where you want things to be quiet.
So the examples I was giving were the air conditioning duct or something like that where
traditionally it's a very quiet application. The other possibility is where you want a smooth
torque application. So you're not worried about the acoustic noise, the noise you can hear, but
you're more worried about actually applying this twist in a smooth way. So if you were to use it in
robotic drives or something where you want to be able to have a very smooth application, if you're
using it in a milling application where you want a surface finish, your mill is defined by how
smoothly you can move the milling cutter across the face.

There are other applications. Other examples are if you go to steering by wire in a vehicle, you
probably don't want your steering wheel just to feel like a joystick, that you could spin it with
no reaction. So you want it to feel like as you turn it more there's more resistance against it.
But the thing that's giving you this resistance is going to be an electric motor, and you want that
to feel nice and smooth as you turn, you don't want to feel like your steering wheel has gravel in
it or something.

Robyn Williams: By the way, on the gossip front, I've just been to Cambridge and I met the
professor of aeronautics who is a woman and surprisingly young actually, and she told me that their
work on the silent aircraft is going extremely well and they have various parts that they're
testing as we speak. So the silent aircraft, which will have a fantastic role to play in the
future, is going very nicely. Are you at all in touch with them?

Greg Hynes: No, I'm not unfortunately, I don't know...

Robyn Williams: But what about your own electric car? I hear rumours that you're very keen.

Greg Hynes: Yes, I am building my own electric car at the moment. I'm converting a Mazda 121 bubble
car into an electric vehicle.

Robyn Williams: What's the main aim; hobby or what?

Greg Hynes: Just a hobby, there's nothing particularly groundbreaking in any of the components or
anything that's going into the electric vehicle, but it's just good fun to be able to see the real
practical applications. I spent months down here in the lab playing with maths, effectively, trying
to make these electric motors work better, and it's nice to have some practical applications of
these things.

Robyn Williams: And I see that Australia's first national electric vehicle festival is on in
Canberra this long weekend. Greg Hynes in mechanical engineering, University of Darwin the Top End.

Guests

Greg Hynes

Lecturer Mechanical Engineering Charles Darwin University Darwin Northern Territory

http://www.cdu.edu.au/

Presenter

Robyn Williams

Producer

David Fisher

Radio National often provides links to external websites to complement program information. While
producers have taken care with all selections, we can neither endorse nor take final responsibility
for the content of those sites.