A yen for serious coin shrinking / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

A yen for serious coin shrinking

1 2 next

Move Thread

LAN_403

Tesladownunder

Sun Jun 04 2006, 06:55AM

Registered Member #10 Joined: Thu Feb 02 2006, 09:45AM
Location: Bunbury, Australia
Posts: 1424

I haven't used my big capacitor bank for coinshrinking yet. (215uF 12kV, 16kJ) I have tried with smaller setups using 1.5kJ but with only a few percent shrinkage of an aluminium disc. This has been documented on my site and in 4HV. Other related coinshrinking topics include the definitive site from Bert Hickman and Steve Ward's MO cap shrinker from the old forum.
A few months back I was rifling though the physics prof's loose overseas change and found 8 Japanese yen. These are small light aluminium like coins.
So now I had some yen for serious coinshrinking. Get it? Sorry....

Now first the setup.
Normal coinshrinkers use a coil and some dowel but no external support.
I wanted more of the force to be directed to the coin and to keep the work coil in proximity longer
The work coil was still left over from last years experiments and is 10 turns of 2mm wire. The key is that it was interoven with fibreglass cloth as the axial forces will slam the coil turns together and short them out. Plastic would get squashed but I was hoping the fibeglass might do better.
The whole thing was then wrapped in cloth from some garish print dress (not mine). It was then placed in a PVC endcap and short pipe, filled with water and frozen. That in turn went in another layer of endcap and pipe with the gap filled with water and frozen. The hope was that the cloth would add strength to the ice like glass fibre does to epoxy to give fibreglass.
I was all enclosed in a wooden box. I hid behind a blast shield and also had the setup monitoring current on the CRO with a Rogowski coil.

So here are the setup pics and the damage done following a 5kJ discharge.
The ice was shattered as were all the PVC components and a panel was broken off the box. It wasn't that loud using hearing protection and much less than an exploding wire.

Pics of results on the next post.

Peter

1149404137 10 FT0 Coinshrink4fibreglasscoil475

1149404138 10 FT0 Coinshrinkingyen5kjbox

ragnar

Sun Jun 04 2006, 07:01AM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

So where do you get your yen anyway?

:P

Steve Ward

Sun Jun 04 2006, 07:41AM

Registered Member #146 Joined: Sun Feb 12 2006, 04:21AM
Location: Austin Tx
Posts: 1055

Something ive wondered about but never just tested for myself... ice isnt conductive (electrically)?

I would think 5kJ could turn a little aluminum coin into a small ball at 5kJ... considering what ive done to quarters and half-dollars.

Looks like you need some better containment

Tesladownunder

Sun Jun 04 2006, 08:57AM

Registered Member #10 Joined: Thu Feb 02 2006, 09:45AM
Location: Bunbury, Australia
Posts: 1424

A small ball is right. The coin was a little unsupported on one side so I expected a bit of a lopsided result.
It is interesting trying to see the corresponding detail. Particularly the number 1 which looks like it is fracturing out.

Peter

1149411464 10 FT10774 Coinshrinkingyen5kjflipside

1149411464 10 FT10774 Coinshrinkingyen5kjnumberside

Madgyver

Sun Jun 04 2006, 09:53AM

Registered Member #177 Joined: Wed Feb 15 2006, 02:16PM
Location: Munich, Germany
Posts: 214

Nice. I always figurred, that they could be great in Disc shooters, as they are realy light.

Tesladownunder

Sun Jun 04 2006, 11:19AM

Registered Member #10 Joined: Thu Feb 02 2006, 09:45AM
Location: Bunbury, Australia
Posts: 1424

Here are the copper fragments. Interesting how one part of the coil is almost unscathed apart from doubling its diameter where the other half is fractured, compressed and in pieces.

The current reading is interesting (first picture) and my Rogowski coil indicates a heavily damped waveform with a peak of 'only' about 50 kA. (50kA/div, 50uS/div) I had expected more. There are no discontinuities to fit with the wire disintegrating.
The second picture is a test run at only a few joules prior to the main shot (?2A/div, 50uS) shows the waveform of the same combo but with much more prolonged ringing so at full power the energy is being dissipated much faster although it difficult to be sure since the amplitude on screen is so much different.

Peter

1149419951 10 FT10774 Coinshrinkingyencoilpost

1149419951 10 FT10774 Coinshrinkingyen5kjcro

1149419951 10 FT10774 Coinshrinkingyen10jcro

Quantum Singularity

Tue Jun 06 2006, 04:12PM

Registered Member #158 Joined: Sun Feb 12 2006, 09:53PM
Location: Central Ohio
Posts: 282

I was wondering if any of you shrinkers have tried fiber reinforced packaging tape? I think several layers of that would be similar to fiber reinforced laminate and probably alot easier to use and cheaper? I am actually in the process of constructing a quarter shrinker and with the couple hours every other week I have available to work on projects, will probably take another couple months. I have most all the parts, just need to assemble.

Tesladownunder

Tue Jun 06 2006, 05:58PM

Registered Member #10 Joined: Thu Feb 02 2006, 09:45AM
Location: Bunbury, Australia
Posts: 1424

I am not sure that fibre reinforcing is the way to go yet and this was my only serious shot at coin shrinking. I have looked through Bert Hickmans gallery of coins and the energies at which they were fired. It seems to be efficient for 5kJ.
I have asked Bert for his opinion.
I expect the ice around it is more important but the coil expanded to twice its diameter anyway. The fibreglass was shredded anyway

Next time I will use my box girder section as combination blast shield and restraint for the ice. Getting harder to fit it into the freezer though..

Peter

Bert

Tue Jun 06 2006, 07:17PM

Registered Member #118 Joined: Fri Feb 10 2006, 05:35AM
Location: Woodridge, Illinois, USA
Posts: 72

Something very strange occurred with this coin. The marked non symmetrical shrinkage was probably due to the coin being constrained only on one side. Since I normally strive for uniform shrinkage, I've never tried shrinking with only one dowel rod in place. Interesting effect!

I assume that 2mm wire is about the same as #12 AWG. From the work coil photo, the length of the coil looked to be ~3.2 cm, and the coil ID was perhaps about 2.5 cm. (BTW, I always use the coil ID, not wire center-to-center) to estimate the "effective" inductance, since proximity effect will bunch the current into a thin stripe along the side of the wire that faces the interior of the coil). The estimated open core (no coin) Wheeler inductance of your work coil was perhaps ~1.6 uH or so. The low power scope shot indicates that you got four complete cycles over a period of ~475 usec, for a resonant frequency of about 8.4 kHz. Assuming a 215 uF bank capacitance, your overall initial loop inductance (including "loaded" work coil, caps, switch, and bus bars) was about 1.67 uH - in the ballpark.

Assuming 5 kJ of bank energy (~6800 volts) and no change in work coil inductance, the predicted current peak would be about 77 kA. However, we know that the coil stretched and expanded as the current increased. In order for the measured current peak to only be ~50 kA, the "effective" work coil inductance must have increased to ~3.9 uH, implying that the work coil expanded to an average diameter of about 1.7" by the first current peak. Further, based on the shape of the intact portion of the coil, the portion that remained intact actually expanded to at least 2X the original diameter, while the portions nearest the coin expanded even more and fragmented. Fragmentation causes arcing that then bridges the gap between still-connected portions of the remaining coil. However, the coil was also simultaneously being axially compressed. This most likely led to some turn-to-turn shorting as well. Since most of the system energy (as you approach the first current peak) resides in the magnetic field surrounding turns of the work coil, any arcing between turns would not be directly reflected in your current waveform, other than showing up as markedly increased system losses.

Looking again at the high power waveform, there is still significant residual system inductance during the high power shot, since you don't see a sudden jump in resonant frequency (which might imply that the entire coil was bridged by an arc). The high power current waveform DOES show considerable increase in lossiness though, the hallmark of one or more arcs becoming the major lossy element in the system. Also, the decay envelope looks to be approximately linear - again a very strong indicator of arc(s) being the major lossy element(s) in the system. And, perhaps by encasing the coil in ice, you further "constricted" arcs that formed during the initial wire breaks, causing them to be more "resistive" than unconstrained arcs. In any event, I have little doubt that arcing is the source of your excess system losses.

Based on your results, it appears to me that your work coil began fragmenting on the way towards the first current peak, and that system energy was then lost at a very rapid rate through arcing. It might be possible to more easily detect when fragmentation begins by monitoring the voltage developed across the work coil. I'd recommend repeating the experiment, but this time using a simple close wound coil of heavy build polyamide-imide magnet wire, centering the coin with identical dowels on each side, and simultaneously monitoring coil voltage and current.

And obviously, you also need to develop a more robust containment method before repeating the experiment.... :^)

Good luck, and remember to wear your helmut and flak jacket...

Bert

Bjørn

Tue Jun 06 2006, 10:18PM

Registered Member #27 Joined: Fri Feb 03 2006, 02:20AM
Location: Hyperborea
Posts: 2058

What about using a photodiode to detect when the first arcs appear?

1 2 next

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