Varicoil coilgun design thoughts / Electromagnetic Projectile Accelerators / Forums

Forums

4hv.org :: Forums :: Electromagnetic Projectile Accelerators

« Previous topic | Next topic »

Varicoil coilgun design thoughts

Move Thread

LAN_403

ThreeGigs

Fri May 02 2008, 03:08AM

Registered Member #1224 Joined: Fri Jan 11 2008, 03:11PM
Location:
Posts: 2

Longtime reader but first time poster here. I've been rattling an idea around my head about a different approach to a coilgun design, and I'd like to bounce my thoughts off of the community here.

First though, I'd like to make sure I'm understanding a particular principle correctly. It's actually quite difficult to find any sort of generalized performance data or 'rules of thumb' for reluctance coilguns. From what I can gather, there's an optimum length ratio between a projectile and a single coil, based solely on peak force applied to the projectile at a given power level. I'm thinking in terms of a solenoid here, not necessarily something designed to shoot anything.

My best guesstimate is that a coil should be either the same length as a projectile, or twice as long as the projectile, and that the peak force will occur just bfore the projectile fully enters the coil windings. In theory, applying a current pulse with about 2/3 of the projectile in the coil would result in the greatest acceleration. In reality, current reversal, suckback and pulse length all throw monkey wrenches into the equation. But, if one were to apply say 400 volts at 1 amp constant power into a coil of say 10 cm length, with a projectile of 5 cm length, then the greatest 'pull' on the projectile will occur with only about 1 mm of the projectile outside the coil. However there's a range of overlap between coil and projectile that yields numbers close to peak, and I *think* that range starts at about 70% of the projectile being in the coil, and continues until 30% of the projectile is past the midpoint in a coil twice the projectile length. I *think* that that's where the peak efficiency of the system happens.

Now, if you could move the coil, such that as the projectile accelerates the coil would always be just ahead of the end of the projectile, you'd get the maximum delta v over a given distance. Naturally if you can accelerate the coil, you could just accelerate the projectile and do away with all the magnetism.

And this is where my idea comes in.

Imagine a 100 cm long coil of a single layer using 1mm thick wire (including insulation, just to make numbers simpler). You'd have 1000 turns of wire.
Now, if you connected turn 1 to one terminal of a power supply, and connected the other terminal to a tap on turn 100, you'd have a 10 cm long coil, which is a subsection of that longer coil. You could connect another power supply to turns 50 and 150, another to 80 and 180, etc. Depending on which power supply you connected, you could 'move' the active portion of the coil.

So, my thinking goes like this:

Make a coil as above, about 1 meter long.
On one side, you connect IGBTs to every 30th turn of wire, stopping at 90 cm (900 turns). You feed all those IGBTs from a common busbar.
Same thing on the other side, an IGBT tapped in every 30th turn, starting at turn 100 and going through to the end. The other side of those IGBTs connects to another common busbar.

Connect both busbars to a DC power source.

What you've got now is something like a multi-tap transformer, only it's a multi-tap coil. You can energize an arbitrary portion of the coil in units of 3 cm. So you could have a 12cm long coil from 1 to 12 cm. And a millisecond later a 12 cm coil between 3cm and 15 cm.

If you have 30 IGBTs feeding and 30 IGBTs grounding, with the ability to pick any two IGBTs to feed current through, you'd effectively have a 30 stage coilgun. And if the timing is right, there won't be any suckback because once the projectile goes past the middle of the energized section, you'd simply de-energize the first subsection of the coil (one tap's worth), and energize another tap after the end.

Benefits (unless my understanding is flawed):
Better efficiency.
Higher acceleration in a given distance as compared to multi-coil guns, since the virtual coils are effectively overlapping.
Ability to change coil length on the fly to suit different projectiles.

Thoughts? Criticicism? Potential improvements?

Berni

Fri May 02 2008, 08:17AM

Registered Member #1132 Joined: Mon Nov 19 2007, 06:15PM
Location: Slovenia
Posts: 84

This is basicly a multistage coilgun, but with a diferent aproach

Isted of useing one big coil and makeing taps rather make a few independant coils and then put some kind of trigers betwen them(optical gates,contacts or even spakgaps)So then evry one of these trigers is conected to the coil next to it so when the projectile hits the triger that means its in the perfect position and the coil fires.This way you only need one switch per coil and its proabobly better to use a SCR since they handle the huge pulse currents better.

And yes this dose inprove efficiency over a single stage design.

Further inprovment can be made by giving the projectile a small starting speed by punching it out with a high speed selonoid

ThreeGigs

Fri May 02 2008, 11:37PM

Registered Member #1224 Joined: Fri Jan 11 2008, 03:11PM
Location:
Posts: 2

so when the projectile hits the triger that means its in the perfect position and the coil fires

See, that's my thinking. 30 10cm long coils would be 3 meters (10 feet) long, if end to end.
But with my idea, you effectively wind up with multiple 'overlapping' coils. With a multi-stage gun using 10cm long coils, you'd get one 'perfect' position every 10cm. My design would allow you to move the 10cm energized zone in increments of 3cm, thus more 'perfect positions' per meter, yielding higher acceleration for a given length.

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob