Large Radius High Speed Stepper Motor

[[email protected]]

Hi,

I am trying to make a fan using a stepper motor where the radius of the
rotor is very large, say, 1 meter, and the drive is applied at the
periphery of the fan (the fan will have a rim at the outer
circumference).

I want super high rotational speed. The faster the better. Natuarlly
the atmospheric drag will increase with the rotational velocity, but I
was hoping that I could simply increase the speed by increasing the
power delivered.

I can drive the stators in any sequence at any rate with any amount of
current under software control (USB).

Is there a speed beyond which there is a fundamental limitation on how
fast I can pulse my stators? Can I expect to be able to get the fan to
turn faster simply by increasing power? What type of efficiency can I
expect (electrical power to frictional atmospheric forces?) I am
assuming that efficency losses will be in joule heating and radiation,
so it seems that I should be able to get high efficiency. And finally,
a big one, what is the optimum geometry of the magnetic assembly at the
perimeter? How many magnets should I use? Is more really better?
Should I use some sort of super cap array to deliver the pulses? What
types of (electro) magnets should I use? Will I have to worry about
vibration? Lubrication? Metal bearings for direct contact? d-phi-b-dt
effects?

I know I am asking a lot, and I don't need all the answers now, just
some hints to keep me from trying all the silly things.
-Chaud Lapin-

 

[Paul Burke]

Hi,

I am trying to make a fan using a stepper motor where the radius of the
rotor is very large, say, 1 meter, and the drive is applied at the
periphery of the fan (the fan will have a rim at the outer
circumference).

I want super high rotational speed. The faster the better. Natuarlly
the atmospheric drag will increase with the rotational velocity, but I
was hoping that I could simply increase the speed by increasing the
power delivered.

You increase the speed of a stepper by increasing the step frequency. It
will have a huge angular momentum, so don't try to accelerate quickly.
But why a stepper motor? Surely you don't need to control the
instantaneous position of a fan? I would have thought a linear induction
motor operating on an aluminium band round the rim would have been a lot
easier.

Paul Burke

 

[Anders F]

Hi,

Hi!
You are indeed looking for trouble!

I am trying to make a fan using a stepper motor where the radius of the
rotor is very large, say, 1 meter, and the drive is applied at the
periphery of the fan (the fan will have a rim at the outer
circumference).
I want super high rotational speed. The faster the better.
Natuarlly
the atmospheric drag will increase with the rotational velocity, but I
was hoping that I could simply increase the speed by increasing the
power delivered.

Of course it takes power to do work (move air)... So you're right on that
one. Fast rotation of a 2m diameter rotor: kWs - and a lot of them....

I can drive the stators in any sequence at any rate with any amount of
current under software control (USB).
Is there a speed beyond which there is a fundamental limitation on how
fast I can pulse my stators? Can I expect to be able to get the fan to
turn faster simply by increasing power? What type of efficiency can I
expect (electrical power to frictional atmospheric forces?) I am
assuming that efficency losses will be in joule heating and radiation,
so it seems that I should be able to get high efficiency.

You'll burn a lot in driver/coil losses.

And finally,
a big one, what is the optimum geometry of the magnetic assembly at the
perimeter? How many magnets should I use? Is more really better?
Should I use some sort of super cap array to deliver the pulses? What
types of (electro) magnets should I use? Will I have to worry about
vibration? Lubrication? Metal bearings for direct contact? d-phi-b-dt
effects?

What's the application? Sounds like a no go in my ears... I'd go for more
smaller (traditional) fans.

Cheers,
Anders

 

[[email protected]]

You increase the speed of a stepper by increasing the step frequency.
It

will have a huge angular momentum, so don't try to accelerate quickly.
But why a stepper motor? Surely you don't need to control the
instantaneous position of a fan? I would have thought a linear induction
motor operating on an aluminium band round the rim would have been a lot
easier.

Paul Burke

Hi Paul,

No, I don't need to be able to control the position accurately, but I
would like to be able to reverse the direction. It's ok that the
blades would have high angular momentum. As a matter of fact, I was
thinking that it might actually be better that there is some rotational
kinetic engergy for stability.

My overall concern, I guess, is to get maximum efficieny in transfer of
engery from an electrical source to the blades, either in the
rotational kinetic enery or through forced movement of air by the
blades.
I am going to check out the linear induction motor.

-Chaud Lapin-

 

[[email protected]]

What's the application? Sounds like a no go in my ears... I'd go for more
smaller (traditional) fans.

Cheers,
Anders

Hallo Anders,

The application (among other things) is a wind tunnel where the input
is free-atmosphere. Again, the radius of the fan needs to be 1 meter
at least.

I am mostly concerned with the efficiency of transfer of energy from an
electrical source. Though I have near-zero experience with stepper
motors (college work is over a decade old), it seemed intuitive that
placing the magnetic drive at the perimeter of a large radius and
pulsing the stators under software control at just the right moments
could yield some very interesting scenarios. I also get the feeling
that brief pulses delivered through low-reistance coils will minimize
core losses. I would use standard techniques like quadrature to
determine angular position and velocity. The key in all of this would
be the software, not so much to force the rotor to assume a certain
rotational pattern, but to make the rotor turns as fast as possible
using strategically activated pulses, more or lest opportunistically.
-Chaud Lapin-

 

[Boris Mohar]

Hallo Anders,

The application (among other things) is a wind tunnel where the input
is free-atmosphere. Again, the radius of the fan needs to be 1 meter
at least.

I am mostly concerned with the efficiency of transfer of energy from an
electrical source. Though I have near-zero experience with stepper
motors (college work is over a decade old), it seemed intuitive that
placing the magnetic drive at the perimeter of a large radius and
pulsing the stators under software control at just the right moments
could yield some very interesting scenarios. I also get the feeling
that brief pulses delivered through low-reistance coils will minimize
core losses. I would use standard techniques like quadrature to
determine angular position and velocity. The key in all of this would
be the software, not so much to force the rotor to assume a certain
rotational pattern, but to make the rotor turns as fast as possible
using strategically activated pulses, more or lest opportunistically.
-Chaud Lapin-

Do some math on the angular momentum. Since you want the rotating part of
the motor to be on the circumference, the largest concentration of mass will
also be there which is the worst possible place. The requirement for the
tensile strength of the blades will doom the project.



Regards,

Boris Mohar

Got Knock? - see:
Viatrack Printed Circuit Designs http://www3.sympatico.ca/borism/

 

[Pat Ford]

Do some math on the angular momentum. Since you want the rotating part of
the motor to be on the circumference, the largest concentration of mass will
also be there which is the worst possible place. The requirement for the
tensile strength of the blades will doom the project.



Regards,

Boris Mohar

Got Knock? - see:
Viatrack Printed Circuit Designs http://www3.sympatico.ca/borism/

Having some experience with a BIG fan wind tunnel Boris is right. I've
worked on a large fan ( 6Mw motor, folker craft wing segments for the
blades) You dont want the mass on the outside of something that big.

Pat

 

[Ken Smith]

Hi,

I am trying to make a fan using a stepper motor where the radius of the
rotor is very large, say, 1 meter, and the drive is applied at the
periphery of the fan (the fan will have a rim at the outer
circumference).

I want super high rotational speed. The faster the better. Natuarlly
the atmospheric drag will increase with the rotational velocity, but I
was hoping that I could simply increase the speed by increasing the
power delivered.

Remember that power goes as roughly the cube of speed. Doubling the speed
takes about 8 times the power.

Is there a speed beyond which there is a fundamental limitation on how
fast I can pulse my stators?

The stator is inductive. The rotor shunts that inductance with some
resistance. All energy you put into the inductance must be removed when
you stop the current. You will be sloshing around large amounts of
energy. The faster you go, the higher the losses.

Should I use some sort of super cap array to deliver the pulses?

No, supercaps have too much internal resistance. You are better off
trying to ensure that the current draw is near constant for the whole
assembly.

What
types of (electro) magnets should I use?

Very expensive LN2 cooled or LHe cooled ones.

Will I have to worry about
vibration?

I wouldn't call it worry. Something closer to !!!!PANIC!!!! would be the
right level.

Lubrication?

If you are going fast, air bearings are the way to go.

 

[Rich Grise]

Hallo Anders,

The application (among other things) is a wind tunnel where the input
is free-atmosphere. Again, the radius of the fan needs to be 1 meter
at least.

I am mostly concerned with the efficiency of transfer of energy from an
electrical source. Though I have near-zero experience with stepper
motors (college work is over a decade old), it seemed intuitive that
placing the magnetic drive at the perimeter of a large radius and
pulsing the stators under software control at just the right moments
could yield some very interesting scenarios. I also get the feeling
that brief pulses delivered through low-reistance coils will minimize
core losses. I would use standard techniques like quadrature to
determine angular position and velocity. The key in all of this would
be the software, not so much to force the rotor to assume a certain
rotational pattern, but to make the rotor turns as fast as possible
using strategically activated pulses, more or lest opportunistically.

You're reinventing the "fan"?

To heck with electronics - if you want lots and lots of air, use a RAT:
Rotary Air Turbine.

Good Luck!
Rich

 

[Rich Grise]

Hi Paul,

No, I don't need to be able to control the position accurately, but I
would like to be able to reverse the direction. It's ok that the
blades would have high angular momentum. As a matter of fact, I was
thinking that it might actually be better that there is some rotational
kinetic engergy for stability.

My overall concern, I guess, is to get maximum efficieny in transfer of
engery from an electrical source to the blades, either in the
rotational kinetic enery or through forced movement of air by the
blades.
I am going to check out the linear induction motor.

Hasn't this been done? Look up "wind tunnel". Maybe even "turbine bleed
air". I haven't checked these things, but it sounds like all you want is a
controlled wind tunnel.

Designing and building a stepper motor from scratch is just stupid.

Good Luck!
Rich

 

[[email protected]]

You increase the speed of a stepper by increasing the step frequency.
It

will have a huge angular momentum, so don't try to accelerate quickly.
But why a stepper motor? Surely you don't need to control the
instantaneous position of a fan? I would have thought a linear induction
motor operating on an aluminium band round the rim would have been a lot
easier.

Paul Burke

Well, I must say, I feel like a dummy. LIM is exactly what I had in
mind.

Not like I thought the idea was new or anything, but all the various
"extra" features that I thought I could add like rapidily discharching
stored engergy using supercaps has already been done. That are several
nice, big, fat juicy papers put out by U.S. Department of Energy on
this subject.


-Chaud Lapin-