Linear motor rules of thumb

[Wedwin]

Hello all!

I have been thinking about making a linear motor. Several permanent magnets on a rail and a trolley with six anchors that is turned on and off in sequence.

(sorry if the gif is hogging all resources)

I have tried to draw this with a few different measurements. But I can't decide which is the better. In the animation, there is 48mm between magnets, and anchors are one and a third length, that is 64mm. the anchors are spaced with 6mm difference.

My question is: is there a more optimal way of defining the length of the anchors? Should they be a bit closer to the spacing of the magnets?

To suppress a little of the "cogging" the anchors are switched off as it nears a magnet. And I still don't want the magnetic fields to be too far apart.

What throws me is that the anchors are energized for such a short period, and my knowledge is not sufficient to make a scientific decision.

 

[itsmoked]

I believe you can use regular polyphase motor design rules with excellent results by just unrolling the cylindrical motor body. Those equations then all work the same. Spacing, winds, gaps, most everything.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[sreid]

It's hard to tell from the animation. Do you have 6 phases? If so, it makes for a more complicated amplifier. Also,usually, there are three coils between the magnets;the coils are the hard part of the manufacturing.

 

[waross]

Your design seems to be more of a magnetic bumper jack than a linear motor. You have recognized this in your timing adjustment to reduce cogging.
The industrial Linear induction motors that I have worked with do not use permanent magnets.
In the 400 HP rail propulsion units they use a design as described by itsmoked.
There are three or six coils driven by a three phase supply. The motor coils and iron are very like the stator of a three phase motor rolled out flat. In the place of the rotor is the "linear induction motor rail" or LIM rail for short. This is fixed to the ground between the car support rails. For the motors in question the LIM rail is a strip of steel about 12" wide and infinite length. This is covered by an aluminum plate about 1/8" thick. The coils and core are suspended from the car above the LIM rail with a very small air gap. The cois magnetize the steel rail. This induces eddy currents in the aluminum. The field resulting from the eddy currents interacts with the field produced by the motor coils and produces thrust instead of rotation.
The motor coils are driven by a first cousin of a VFD.
Where a conventional motor has a synchronous speed, a LIM has a synchronous velocity.
In a conventional motor the synchronous speed is a function of frequency and the number of pole sets.
In a linear induction motor the synchronous velocity is a function of the frequency and the spacing between the coil centers.
A conventional motor will regenerate if driven above synchronous speed and will be retarded if the frequency is dropped below the frequency corresponding to the speed of the motor.
In a rapid transit propulsion LIM the frequency is ramped down to decelerate the train as it approaches the stops. The drive regenerates and returns the energy to the DC bus.
With regeneration the train ma be slowed almost to a stop. The train is going so slow that when the brakes do apply the train makes a smooth stop in 12 or 15 inches.
Wedwin, I suggest you do a little research into industrial LIMs. I believe that you will find them easier, smoother and cheaper than using anchor magnets. You can get a start by examining the windings and spacings in an induction motor.
A tip, the air gap is more important than the magnetic path through the iron.
respectfully

 

[Wedwin]

Sorry for the dalay.

Itsmoked, I don't know why, but I never came to think about that. Will check.

sreid, yes there are six phases. I chose to make the animation continous, and so it became a bit confusing to watch. I don't understand the part with amplifiers. Turning the windings on and off, should be simple enough with an H-bridge. And winding the coils couldn't possibly be a problem. I guess you speak of a more industrial design than I can achieve

waross, bumper jack is a cool name for the technique. Thanks for the pointers to LIM. Research will indeed commense.

But the goal at this stage is to make something move back and forth along a rail about a meter, maybe two.

I haven't planned on using anything but DC switched on and off, and as the motor you speak of would run on Variable frequency drive, it is a bit out of my range. I also aim to keep the DC at 12 or 24V. The controller will be a simple PIC-processor. Commutation will be controlled by Hall- or Magneto resistive sensors.

I think I can achieve precision enough to keep the air gap small.

 

[itsmoked]

Then make everything adjustable!! That way you can change the coil spacing at will.

Let us know how you get along.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[Wedwin]

It was some time ago that I asked the question in this thread:

http://www.eng-tips.com/viewthread.cfm?qid=143198&page=1

But it is what is the foundation of what has become this thread.

I might well report my findings. Just don't forget to breathe until then ...

Thanks a lot for the valid pointers and kind words.