Subject:
|
Re: Balancing robots
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Tue, 27 Jan 2004 03:40:48 GMT
|
Original-From:
|
Jim Choate <ravage@einstein.ssz/avoidspam/.com>
|
Viewed:
|
1150 times
|
| |
|
|
Hi Peter,
On Mon, 26 Jan 2004, PeterBalch wrote:
> Motors really are tricky things.
>
> Oh dear, I seem to have stirred up a hornet's nest.
>
> I too have built magnetometers and they are far easier to understand and
> model. The problem with motors is that they go round - and they have a
> commutator.
To each their own. That they go round is called 'polar coordinates' if you
have problems translating it to cartesian. The result is the same either
way. The commutor is a (low) resistance. Now we could fall back on a
lumped-parameter model if you would like, that complicates the entire
model because we now have to model the inter-winding capacitance and
thermal effects. For this sort of application that is over-kill.
> As they go round, the torque produced varies depending on which coils are
> currently connected via the commutator and what angle they are at relative
> to the stator magnets. The magnetic field is nothing like the nice
> straight lines shown in a physics textbook (so "sin(theta)" is worth next
> to nothing). The back EMF depends on the which coils are currently
> connected via the commutator and what angle they are at.
I disagree with several points here, but see above.
> I got down at a range of textbooks from my bookshelf and "back emf" for
> motors is defined as "the induced EMF due to the generator principle" (from
> a book of 1907; a 1992 book gives much the same definition).
>
> The same books talk about the EMF due to the collapse (or build-up) of the
> magnetic field in a coil. They refer to it as the "induced EMF".
I've got a pile of books with clearer definitions. Start with Tocci (late
'70's). Try looking at some basic electronics books teaching basic 'network
theory' within the context of driving a RL(C) circuit with a square wave.
See what they call the response that happens on an L when the square wave
passes through its negative edge.
If you're going to make reference to books (guess you don't agree with
Agassi I see) you could at least provide the title. Science has -no- place
for a plea to authority, -especially- an anonymous one ;)
If you can't explain it in your own words it makes it hard to accept that
you actually understand what you're talking about, no? Do you know how
many times in history somebody has waved a book about and then been proven
wrong? It's not a reliable method to understanding nature, science really
isn't a political activity (or at least it shouldn't be). My personal
estimate is that given any book, at least 90% of its 'absolute' assertions
are probably wrong.
Agassi said to study nature not books, what he means is to -start- with
the books, not end with them. Don't trust them, prove them through
experience. As Asimov said, science isn't "Eureka" it's "That's funny...".
Strictly speaking 'back-EMF' is the resultant currents effect -any time-
an L is relaxed. We talk of 'generator current' and 'back-EMF' in the context
of a motor because it's usefull to differentiate that current from the driving
current. Now, if we move to a strictly 'component' model (ie we talk of
the behavior of the L and the forcing function only) we have two aspects;
forcing function and back-EMF (it is -not- a generator current because
there is no external forcing function [ie magnet coupled to shaft
rotation]). A back-EMF exists in -all- L's, whether they are in a motor or
not. It is -nothing- more than the recover of the energy stored in the B
field that wasn't used. Conservation of energy requires that energy goes
someplace. The L turns it into a current, which because all L's have R
in the real world means an induced voltage. That current is an 'induced
current' as well since it's the B that is causing the current. [1]
The fact you don't seem to understand the component behavior of L with
respect to 'back-EMF' or 'induced current' may be a part of why you find
motors so difficult. I can suggest several books if you'd like.
> Measuring the "back EMF" to estimate the speed of rotation of a PWM
> controlled motor ought to work both when the motor is "on" and when it's
> "off". My questions were: how reliable is it as a way of estimating speed
> and what's the best way to measure it.
How? Assertions require proof/demonstration.
The only way to measure the generator current when the motor is on is to
look at the current change with repect to what one would expect without
that effect present, and that requires a very stable voltage source.
Regular batteries or wall warts ain't gonna cut the mustard there, sorry.
What the generator current does is increase the impedence of the motor,
effectively making it require more energy than it 'should' (I -love- the
2nd Law!).
With it off the generator current you get won't be the same as when the
motor is on because you don't have the extra energy of the driving current
providing the torque to make the generator current. Not to mention it is
very small and very noisy (ie look at it on a scope some time). It also is
very sensitive to measurement device impedence (you can't measure
something without altering it, ever). The next suggestion that I'm sure at
least one of you will make is drive the motor mechanically and measure the
resultant, again the 2nd Law will bite you here as well. You'll have
mechanical loss in the linkage and that will slow the driving motor.
Because the generator current is so small this -will- be important.
Let's look at a actual robot moving across a smooth surface (let's
minimize that loss) and imagine that we cease driving one of the motors.
What is the result? The robot slows to that side, that causes a lateral
force on -both- motors that throws your whole little circus off as the
mechanical losses of those non-straight line forces come into play. This
will happen just as fast as you switch the motors on and off so there
isn't a way to out guess it by being faster than it is. The lateral forces
on the shaft bearing will become important, heat. Heat is a rendomizing
effect and increases the Z. This will reduce your current and to get the
same performance you'll need to increase your driving force.
One final comment, what one 'should' be able to do is often different than
what one can do in practice. It is always a good idea to keep that in
mind. I have a little mantra that I run on a loop in the back of my mind,
thanks to Heinz Pagel (RIP);
1st Law: You can't get ahead.
2nd Law: You can't break even.
3rd Law: You can't quit the game.
Ladies and Gentlemen, I'm done with this dicussion. You are welcome to
whatever opinion of the world you have or want.
Ta ta.
[1] You do realize that a magnetic field is nothing but a model? What we
actually have is photons bouncing from hither to yon and nothing more. An
electron picks up some energy and releases a photon (ie expanding magnetic
field), if something effects that electron or the photon effects some
other electron then we have a coupling. If the electron is not driven
continously then at some time t each individual electron will recover it's
emitted photon (ie collapsing magnetic field). To make reference to a
comment Einstein made about math and reality, don't confuse the name you
give something with that thing. And yes, if you want you can take the
proton/electron/neutron/photon model and turn it into quarks and photons
but it does nothing toward understanding the gross behavior of the
phenomenon. Sometimes the western reductionist approach isn't the right
one (e.g. emergent behavior and complexity theory).
[2] You might also want to look into Josephson Junctions. Refering back to
my example of the PVC pipe and rings a JJ is nothing more than a ring with
a -very- small gap in it. So small in fact that if some electrons are
driven with enough energy they can 'tunnel' accross the junction and hence
we have a current when in a strictly classical sense shouldn't exist. You
might also find SQUIDs of some interest as well.
-- --
Open Forge, LLC 24/365 Onsite Support for PCs, Networks, & Game Consoles
512-695-4126 (Austin, Tx.) help@open-forge.com irc.open-forge.com
Hangar 18 Open Source Distributed Computing Using Plan 9 & Linux
512-451-7087 http://open-forge.org/hangar18 irc.open-forge.org
James Choate 512-451-7087 ravage@ssz.com jchoate@open-forge.com
|
|
Message is in Reply To:
 | | Balancing robots
|
| Hi Gordon and Jim Motors really are tricky things. Oh dear, I seem to have stirred up a hornet's nest. I too have built magnetometers and they are far easier to understand and model. The problem with motors is that they go round - and they have a (...) (21 years ago, 26-Jan-04, to lugnet.robotics)
|
6 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
