Controlling coils for magnetic accelleration

[THX-1138]

Two related cases involving multiple coils around a cylindrical barrel. Two separate goals (but probably related math):

1. Coils are activated in sequence to accelerate a magnetic object down the bore. Think 'Gauss-pistol'.
2. Coils are activated (possibly with polarity reversal at/as needed) to smoothly accelerate a magnetic plunger down the bore -- and maybe cushion it when nearing the ends. When it reaches one end, the process is reversed to accelerate it back to the other end. Think 'rectilinear actuator'.

Physical feedback (e.g., optical interruption) about the object's position in the bore is my first preference, since fancy harmonic circuits aren't my forte.

I'm basically curious how the coils' fields should build and decay relative to the object in order to get the best effect (e.g., so one field's decay doesn't interfere with the next one, or 'suck back' on the object in the bore).

 

Digital circuits, microcontrollers, A/D thresholds, etc. -- these I can do. The magnetic field stuff, though, is right out of my experience.

 

I know this is kinda rambling and unclear, but any feedback, comments (constructive), or pointers would be extremely welcome.

 

Thanks!

[THX-1138]

Should I have asked in the Physics forum, maybe?

Edited June 4, 2015 by THX-1138

[Enthalpy]

Some elements:

 

I'd take permanent magnets rather than an iron plunger. What matters is the induction at the mobile part, and magnets do that better. Less worries with varying induction too.

 

You need the coils to interfere. They should even overlap partially. Though, you don't necessarily need nice sine drive currents; you can stop the current in one (the induced voltage will exist though) and put power in the other.

 

You need a rather fine sensor for the position. Switching the coils at wrong moments would waste quite a bit.

 

One design example would use flat cylindrical magnets and iron disks, all stacked, to achieve a radial field that alternates NSNS (needs some convincing mechanical means!). The corresponding coils would be solenoids on the stator. Consider 0.3T reasonable with Nd magnets, then v*B*L is an induced voltage and I*B*L a force (L over several turns, where the induction is strong). Many-many other designs are possible and sensible.