New can crusher coil / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

New can crusher coil

previous 1 2 3 next

Move Thread

LAN_403

klugesmith

Thu Feb 10 2011, 02:21AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

Microwatt wrote ...

but you are missing a point here is that the machined coil will beat any wire wound coil in mechanical strength tests. your bittur coil windings will seperate and reduce the induction on the piece. with this ring enclosed inside a polycarbonate ring. I am sure I won;t end up with a C clamp!

Nice picture.
Where are the wires connecting the 1-turn "coil" to the power source?
How thick are they, and how does the current spread into the coil piece?
Are they far apart (adding more inductance than the coil itself)
or close together (so they repel each other with coin-crushing force levels)?

I think the lengthwise slot in the metal is the weak point: under internal pressure higher than that in firearm chambers, the slot will open up and the metal will indent the inside of the polycarbonate (which is vastly less stiff than steel or even copper). The theory of thick-walled cylinders under pressure comes into play.

Like you (that's a guess), I've given lots of thought to re-usable coin shrinker coils, but never even tried an expendable one wound from wire. Would suggest instead of a thick plastic outer shell, a very thin shell of hard insulating material backed by a steel shell. (Or maybe a carbon fiber composite -- not sure if it's stronger and/or stiffer and less compressible per unit volume in this "gun" configuration).

The thick coil itself I would make in many turns (to eliminate the longitudinal slot and consequent bending stress; also to facilitate the power connections and current/voltage matching). Rather than a Bitter coil of separate plates, or trying to edge-wind a copper strip into a spring, how about sawing a thin helical slot like a flexible shaft coupling?

The slot would need to be filled with hard insulating material able to hold off the per-turn voltage (perhaps 1000 V) and withstand the axial compressive force, since the solenoid wants to contract in length and expand in radius (both directed towards increasing its inductance).

I think a key requirement is hardness and strength of the coil material close to the workpiece (one quarter-size coin).
Perhaps a strengthened copper alloy like that in some resistive Bitter magnets -- in our application there's not such a premium on high conductivity.

I hope you get to try your idea and tell us about it. Not likely for me to shrink any coins in the foreseeable future.
Not likely that either of us has thought of something Bert Hickman hasn't considered. Bert?

Patrick

Thu Feb 10 2011, 02:44AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

You can forget carbon fiber-Graphite composite, they are light, stiff and strong, but have appauling failure when stuck or to explosive force.

As for your steel shell idea, I thought- and then rejected this idea due to impulse transformer like coulping wich would turn the shield in to a 1-turn short. I do not know of a non condutive material that can still cope with these incredible forces.

However, Inconel-X could be used for the work coil, ($$$) then at the cost of increased resistence perhaps it could be wire EDMed, need sleep now.

Microwatt

Thu Feb 10 2011, 04:32AM

Registered Member #3282 Joined: Wed Oct 06 2010, 05:01PM
Location:
Posts: 224

Klugesmith wrote ...

[Nice picture.
Where are the wires connecting the 1-turn "coil" to the power source?
How thick are they, and how does the current spread into the coil piece?
Are they far apart (adding more inductance than the coil itself)
or close together (so they repel each other with coin-crushing force levels)?

The ends of the coil will not attract or repel becuase there is no current flow there. see 1st picture for location of input. I would like to know if bert though of this idea.

klugesmith

Thu Feb 10 2011, 06:04AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

Microwatt wrote ...
The ends of the coil will not attract or repel becuase there is no current flow there.

Won't the shrinkage be quite unsymmetric, due to uneven magnetic field because there is no circumferential current near the slot?
Referring to compass points: if the "ends" (the slot) are in the south direction, then the gap wants to expand because of repulsion between northbound current on the whole east side, and southbound current on the whole west side.

Getting the current in and out is a challenge you can't ignore. Remember, quarter-shrinking engineers cite currents of 50 kA to 100 kA with 10-turn coils. For same effect with a 1-turn coil you need 0.5 to 1 million amperes; then interconnection resistance and inductance and Lorentz forces will be more significant by a factor of 100.

Microwatt wrote ...
see 1st picture for location of input.

You mean the two small circles in plan view? Where, in 3 dimensions, do the wires run?

If they are parallel to the coil axis, both on the same "end", the wires will repel each other strongly and add to the force trying to expand the "coil" and open up the slot. If they exit at opposite ends, it will be hard to bring them both around to the power source without the external interconnection loop having more inductance than the work coil itself.

Either way, the current injected into the winding won't be distributed uniformly across its cross-section when close to the slot. Within a distance similar to the winding width (slot length), the current will be concentrated on the side closer to the entry or exit connection.
Suppose the entry and exit connections were flat bars connected to the full winding width, on each side of the slot. Then 1) you might get uniform current distribution along the axis -- but still nonuniform in the radial direction, crowded toward the outer diameter until far enough from the slot, leaving a zone of significant magnetic weakness at the workpiece. And 2) such flat bars, exiting close together in the radial direction, would be the lowest-inductance way to reach the power source. But that configuration would maximize the Lorentz repulsion between the feed bars, to more than coin-shrinking force levels since it's two primary currents, not a primary current and an induced current (in workpiece with smaller cross-sectional area so much higher resistance than the work coil).

Haven't even tried to consider the skin effect in an enormously thick winding. High-speed videos show shrinking complete after 30 or 40 microseconds, implying frequency on the order of 10 kHz.

Bottom line: With due respect, I don't see any benefit of single-turn coil except, as previously noted, possibly can crushing with thousands of uF and less than 1000 volts. Coils with more and thinner turns are easier to feed, azimuthally uniform in the work area, and let you define the axial distribution of current density because each turn carries the same current.

Microwatt

Thu Feb 10 2011, 09:27PM

Registered Member #3282 Joined: Wed Oct 06 2010, 05:01PM
Location:
Posts: 224

You could be partially correct here. Plot is of current density 100A input current.

Patrick

Thu Feb 10 2011, 10:57PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Microwatt wrote ...

You could be partially correct here. Plot is of current density 100A input current.

Partially right!? It would seem to me Klugesmith called it exactly right! the end holes, the red near the center, the green to the outside, all of it. Though, to float my own ego I called the opening up part.

I think you should pursue this differently, try building in stages, starting with this...

1297378676 2431 FT108222 Sam 0410cropped

What software did you use to render that?

klugesmith

Thu Feb 10 2011, 11:30PM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

That's a good picture from a good tool.

Here's another 2-D model of DC current density (in this case using Student Quickfield):
your 1-turn coil unrolled to a flat ribbon, with one connecting wire on the top and the other on the bottom.

Even in the green contour zone, the current flow is axial or backwards near the "ends".

Microwatt

Fri Feb 11 2011, 03:13AM

Registered Member #3282 Joined: Wed Oct 06 2010, 05:01PM
Location:
Posts: 224

I used a trial version of comsol. i might even be able to couple the forces from the slug and the mechanical analysis to find out how much force and energy it would take to compact a quarter into a ball of metal. maybe not becuase we are talking about nonlinear systems and this would take hours or months to set up.

Patrick

Fri Feb 11 2011, 04:10AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

try to biuild a working set-up that will suit your future needs, then start running evaluation tests on different coils of different arrangements, then report back to here. Then, we can advise you or be proved wrong.

start simple, then work towards the complicated.

klugesmith

Fri Feb 11 2011, 04:18AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

That's a very respectable challenge, and I wish you well. I don't see any significant nonlinearities to model, but there is time-varying geometry and time-varying material properties (such as resistivity of the workpiece, I bet due to heating much more than to strain).

A more modest milestone, which I have never quite done myself, is to model the electrical dynamics of a can-crusher with wire-wound coil, where the can is on an insulating mandrel so nothing actually moves. This is easy to instrument in the lab, for comparison. A 1000 J shot should be similar to a 10 J shot, with same timing and 10x the voltage, current, and field strength -- if it weren't for the fact that the higher-power shot heats the workpiece and raises its resistivity significantly.

previous 1 2 3 next

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