(1000 joule) 2004
The left photo above shows the coil made from 8 turns of 2 mm wire. Most can crushers only use a few turns. I have not tried this before as I believed that I did not have enough power to crush a can. Interestingly my 800 V supply with SCR switching barely makes a dent in the can although the delivered energy of 1500 joules is greater. A rapid rise time is important. The centre photo shows the flash when it fires. The right photo shows the momentary trigger switch (the grey and yellow plastic device above) with large metal contacts. I have added a 2 kV low intensity arc to show the contact points. Not really cool like a triggered spark gap. On one occasion the can shorted the coil and I could feel the blast wave from the spark. I wear hearing and eye protection and look away (while firing and taking the picture)!
The can before and after. Note the distorted windings where a short circuit has occurred and vaporized some of the wire. One wonders how many turns there were effectively.
Video of the can being torn in half (720 k but worth the download for the few frames with action. Sound is good too). Watch the voltmeter too.
There is noticeable crushing at 200 J, better at 300 J and the can tears apart at 1500 J. Also dependent on the turns used. Note that 3 turns at 2400 V 110 uF (300 J) had an effect but 3 turns at 5200 V 35 uF (300 J) had no effect. So experimentation is called for.
Enlarge the picture and you can see the "5% crushed lemons" and "low joule" labeling!
Picture above shows the smaller Red Bull cans crushed at various energies and the effect of crushing a full can. The full can is interesting and bears further analysis. My interpretation is that the thermal and mechanical effects are on such a rapid timeframe that there is no conducting away of heat but the fluid has inertia and incompressibility. This has the effect of preventing the aluminium from folding in to the centre. It remains in the area of highest field which is right in contact with the coils. Hence, can disruption into the two halves is enhanced and since there is no infolding the "cut" is cleaner. Note that the energies on the photo are incorrect and should be 200J, 300J, 1500J and 1500J.
The fluid in the can undergoes a major drop in pressure as the two halves of the can start to separate. Hence the observation above that the upper part of the can is folded in. It is sucked in by low pressure rather than by the induced field.
This process is symmetrical and the net result is that the fluid forms a central column that stretches and becomes fountain shaped as in the video.
Video (700 k) shows effect of crushing a full can. The can rapidly disappears leaving a long stream of fluid. Look at it frame by frame if you can.
Some frame grabs at 1/120 second from a video sequence on tearing apart a full can.
This shows the results of attempting to crush a frozen can. Although it can't crush the can does separate with a fine almost laser like crack with separation of the adjacent paintwork.
This graph shows the effect on capacitor life expectancy of voltage
reversal. If you have a capacitor rated for a given life at 80 %
reversal (top curve) then it will last 40 times as long if the reversal is
reduced to 10 %. Avoiding this can be done in several ways.
Left photo shows the voltage reversal on 10 turns of unloaded coil
with the can crusher caps charged to 30 V. Time base is 20 us/div and
5 V/div. This is without the SCR and shows 80 % or more voltage reversal
Right photo shows the SCR acting as a diode and reducing voltage
reversal from 80 % down to 15 % despite the high speeds involved. (good and
will potentially prolong the life of the caps by 40 times or so).
As I have 12 hockey puck SCR's that I found in the dirt in a junk yard
(amazingly) this is the biggest acreage of silicon that I have.
Unfortunately I can't find the SCR's data (3RW 9103-OCG 5660) but I have
tested the forward breakdown at 1800 V. I was hoping they could handle 1000 A
with perhaps 30,000 peak for a 50 Hz half cycle and probably more with a
A better arrangement is here:
Can crusher 3
(3kJ) Aug 2005
The left photo above shows the capacitor bank, gap contacts and crushing coil next to a can crushed recently. The centre photo shows a close up of the tungsten contacts and the movable switch arm attached by heavy braid. The right photo shows a through the middle shot with the connections to the capacitors.
The left photo above shows a smaller coil setup. The centre photo shows the coil mounted inside a can and the effects of it with expansion rather than crushing. The right photo shows a can which had 3 turns wound lengthways but the power was inadequate to crush it. At this stage I had strong suspicions that the caps were failing.
The top left photo above shows the partially finished unit with the 6 defib caps mounted into a microwave oven case with an NST for charging. The top centre photo shows the defibrillator paddle that I use as a remote switch. The top right photo shows the functioning unit as set up for public display. The bottom photo shows it completed with signs and instructions. The can crusher has a flip open front which has the active HV covered. The coil can be angled to shoot an aluminium ring forwards as well. The switching is mechanical with tungsten contacts by a solenoid, triggered by the switch on one defibrillator paddle. There is a voltmeter and a red strobe to indicated charging obscured by the paddle in the pic. The 6 caps shown had died however so they were removed and 4 new ones were added to the original 2 larger defib caps to give a total of 3 kJ. I strongly suspect that these have now died too.
The photo shows the cancrusher setup included in my public demo that I set up for the Physics Dept during the University of Western Australia open day on Sept 20, 2005. It was a popular display with both the can crusher and the Tesla coil generating large noises to bring the curious from afar. I did lots of demos crushing both light and standard strength beer cans with a good crowd response. Although it crushed the cans well it was probably only due to the original large defib caps at 1 kJ that were still functioning. The whole idea of using defib caps is flawed without correcting the voltage reversal.
So, time to move on but not before an autopsy of one of the caps.
The left photo above shows the opened can of a modern type small 35
uf 5.1 kV defib cap. The right photo shows the cap being
unwound after one end was sawn off. What happened next was
interesting... As I unwound more and more (about 10 feet) I started to
get little shocks and these became stronger the further I went. These
caps had just been tested to 5kv then discharged fully. Typically they can
recover a couple of hundred volts to give a visible spark when shorted.
This is one of my most interesting photos despite how dull it looks.
It is a 13 second time lapse with 2 or 3 tears of the capacitor dielectric
showing rows of tiny blue sparks (roughly in line with the two wood screws
seen here). My blue-green gardening gloves have a white cuff that can just
be made out in the motion blur. This shows the dielectric is charged but not
allowing the charge to flow elsewhere. Alternatively it is
triboelectric generation (think van der Graaff generator).
This can crusher has now gone to the Physics Dept at the Uni of Western Australia. For details of the upgrade see the public display section
Can crusher 4
Spring force is from two bedsprings. Copper braid (2 inch) connects to a
copper plate as one contact (I will replace with brass later).
The left photo shows the light blue Maxwell capacitor specs. 110 uF, 12 kV, 8 kJ, 75 kg. General Atomics have given the specs as 80 % rated voltage reversal at 100 kA. 75 A RMS CW current. Design life 30,000 shots or 1600 h DC. Inductance 40 nH. The right photo shows the 2 gray Aerovox capacitor specs. 50 uF, 20 kV, 10 kJ each. They are smaller and despite the higher energy are rated at only 20 % reversal. They were apparently used for 'Star Wars' laser work and saw only 2 hours service in the 1980's. These are all proper energy discharge capacitors with low inductance low profile, high current terminals. They are all second hand so remaining life is not known. I built the special trolley for it to accommodate various experiments as 195 kg is just too much for me to lump around.
The total energy able to be stored of 28 kJ can be compared with the kinetic energy of an AK-47 bullet of 2 kJ in flight. An exceptional level of care is required to avoid unintentional discharge in close proximity in view of the extreme sound / flash / EMP levels. Also metal fragments of exploding coils are of high velocity and very capable of injury.
With these I hope to be able to do a variety of interesting things such as extreme can crushing, coin shrinking, exploding wires, and projectiles as well as a couple of interesting pulsed Tesla coil experiments.
Left photo shows a 1 kJ shot at dusk to get the
exposure long enough so I could coordinate a pull on the string and take the
right photo shows the gap wear is fairly mild but I have only had a
few shots at 4 kJ max at this stage. Still so sign of failure or problems so
should handle higher powers yet. Peak current was 70 kA so far. Actually
there is surprisingly little black copper oxide charring. I attribute this
to the hard positive contact with little bounce. Note the nick in the
braid from an exploding coil (or the aluminium tube as it disappeared down
the wormhole portal). I am still using the same switch and contacts 2
years later in October 2007 so it has been durable and reliable.
Left photo shows the a can crushed and torn apart at 4 kJ. The right photo shows the coil distortion from a similar power shot.
Since the peak can current is around 2 times that of the coil current (see below), then a 70 kA coil current suggests a peak can current of 140 kA.
Left photo shows a frozen can fired at 2 kJ. Note the folding and wrinkling and apparent area of missing can. The centre and right photos show similar shots.
Left photo shows a can in a longitudinal coil about to be fired at 4 kJ. The right photo shows the result.
What will happen here with a split can (1/2 inch gap) and 1 kJ shot?
Left photo shows a can with a split about to be fired at 1 kJ. One might expect that eddy currents won't appear and that nothing will happen. However, the centre and right photos shows the result almost indistinguishable from a 1 kJ shot with a complete can with top and underside views. Why? Because there is still a large return path for the eddy currents where the field is lower around the top and bottom rim of the can.
Left photo shows an interrupted strip about to be fired at 1 kJ. The right photo shows the result. There is no crushing In this split can strip. It is more narrow than the windings, hence the field is uniform and there is no return path for eddy currents. This is despite the similarity to the interrupted can above. Hence here it really does matter that the strip is broken.
Left photo shows a flat sheet about to be fired at 1 kJ. The right photo shows the result which is minimal distortion as the eddy currents would be formed at right angles to the sheet.
Left photo shows a can strip with a fine split about to be fired at 1 kJ. The centre photo shows the setup before firing. The right photo shows the flash of both the main contact as well as arcing across the gap in the strip. Note the shower of sparks from the strip arcing in the bottom right. You can tell this by ray-tracing back to the source.
Left photo shows the crushed strip from the shot above at 1 kJ. Remember that this won't crush if there was no connection so proof that it did arc across . The right photo shows the strip reformed into a circle to show the vaporized ends.
The left photo above shows a double winding of 3 + 3 turns in the same direction about to be fired at 4 kJ. The centre photo shows the can which has been split in three. The right photo shows the remains of the coil which is broken is several places.
Now the CRO shot (20 kA/div and 50 us/div) above is interesting. The current
curve on the CRO shows the current ring down from a peak 45 kA with 65 %
reversal, broadly as expected. Just after the peak (about T + 40 us)there is
a spike of current then the curve looks funny after that and the period of
the first voltage reversal reduces to about 35 us. The next is 55 us and the
last around 55 us as well.
Below is an experiment to test whether fine wire (.024 inch) will maintain crushing at 4 kJ due to the arc despite the wire being vaporised. This is in comparison to the heavy wire before in which the can was divided into three parts. To keep the physical integrity of the coil with a comparable mechanical strength to the heavy wire, I have put it in thin plastic tubing. Of course the mechanical strength is actually very different between the two and this does limit the experiment.
The left photo above shows the setup with the heavy lead in wire joining to the thin wire in the plastic tubing. The centre photo shows the coil mounted on the can. Note that the coil is actually thin wire inside a plastic tube. The right photo shows a can after 4 kJ with minimal crushing and the coil has been vaporised. No copper remains in the tubing which has been split along its length and blackened from the copper oxide. There was no remaining charge in the capacitor.
The CRO readout above of current on the same scale as before (20 kA/div vertical and 50 us horizontal). This shows a small current pulse of perhaps 9 kA with no reversal. Fine, but after almost 200 us there is another pulse of almost 30 kA with no reversal. Explaining this second contact is more difficult. Switch bounce is possible in retrospect but I will have to check how the safety box was positioned as this shifted as one of the panels separated and my have interfered with the switch. Alternatively, the wires may have recontacted but this seems unlikely. I may need to repeat the test when I get more tubing.
Here is a well spaced spiral wind of a can that is full and FROZEN. Unfortunately I missed the current reading on the CRO but it was out of range and over 80 kA. Looks like I will have to reset my Rogowski for 200 kA FSD. Unfortunately, I only finger tightened the bolts so there was some charring around the contacts and a bit of destruction of the brass thread.
The left photo above shows the setup with the heavy rectangular 3
turns. The right photo shows the can after with a rough
spiral cut. The work coil wasn't even dented and the can was not
disrupted. I was still able to slide it out of the plastic covering which
was not punctured. Lots of small aluminium particles. Basically, it had
nowhere to go. The shot was "relatively" quiet as well. I probably lost
some energy with the loose bolts though. 5 kJ is my biggest shot yet. I
suspect it would have been over 100 kA though. Beyond the limit of the
Maxwell alone but not the three caps together. No problems so I can push
The left photo above shows a 3 kJ can crush. The right photo
shows a few moments later. It was gettting a bit dark by the time I got set up.
Listen to the can half hitting the ground a few secs later.
Crushing steel cans is a lot harder due to the strength and the opposing
effects of magnetic attraction and the eddy currents which will be lower due
to the higher resistance of steel. Although I have already shown that
resistance of the can is not the major limitation.
The pic above shows shots at 2 kJ and 4 kJ. With the 2 kJ shot, I opened the
smoking wooden box to the rather pungent smell of burning tuna. I had washed
the can but some liquid had got under the label. There is perhaps only 10 %
Next is tearing the can into strips with an internal coil.
The left photo above shows the setup with the coil inside the can to explode rather than implode the can. The wooden block in the centre is not needed as the net forces are outward. The right photo shows the can torn into strips which were blown away on the free end and curled around the attached end like a large dead insect.
The left photo above shows the setup with two cans in one coil. The centre photo shows the two cans crushed together. The right photo shows toucans.
This workspace is for various references until I get them organised
F. Bitter, Scientific American 213, 65 (Jul 1965).
http://sprott.physics.wisc.edu/demobook/chapter5.htm Physics demo U of Wisconsin
http://cas.umkc.edu/physics/sps/projects/cancrusher/cancrusher.html U of Missouri Kansas City
http://www.geocities.com/yurtle_t/experiments/pulse_discharge.htm 10 kJ coin shrinker Yurtle Turtle
http://members.tripod.com/extreme_skier/cancrusher/ 2 kJ Tristan Stewart (Mad coiler) Pics of can torn apart.
http://www.physics.umd.edu/lecdem/outreach/phun.htm Travelling demos "Physics is Phun" tears can in half. U of Maryland
http://www.physics.umd.edu/lecdem/services/demos/demosk2/k2-62.htm Slo-mo video University of Maryland, College Park, MD Shoots ends 30 ft
http://hibp.ecse.rpi.edu/Can_Crusher/home.html Rensselaer Polytechnic Institute (RPI), 110 8th St., Troy, NY 12180.
(powerpoint - little info)
http://www.powerlabs.org/pssecc.htm Sam Barros. Electrolytics give poor performance for 3 kJ
http://www.altair.org/crusher.html Altair 100 kV but only to 400 J
http://www.amasci.com/amateur/capexpt.html Bill Beaty's cap bank
http://members.tm.net/lapointe/Main.html Bobs HV 1 kJ 40 kV
http://www.redremote.co.uk/electricstuff/destructotron.html Mikes Electric stuff Excellent article
http://www.teslamania.com/ Bert Hickmans Excellent stuff
Measurement of high current pulses (or my
multimeter doesn't have a 100 000 amp scale) 2005
To measure these currents without connecting wires to the high voltage, I have used a small inefficient transformer which does not have a magnetic core called a Rogowski coil. There is no magnetic core to saturate so the pulse response is good. It is a transformer so cannot measure DC, however. The primary of this transformer is only one wire which is the main current carrying cable.
It will respond to the rate of change of magnetic field or current so has to be "integrated" using, in my case, a passive setup with a resistor of 2.2 k ohms in series followed by a parallel capacitor of 0.01 uF.
The left photo above shows the oscilloscope with the actual current waveform upper trace (the Rogowski coil output with passive integration) and the lower one shows the raw Rogowski coil output which shows rate of current change (di/dt). The middle photo is the winding of the Rogowski coil onto flexible coaxial cable. The coax centre is the return wire. The right photo shows the Rogowski coil wrapped around the main cable.
So far so good but it needs to be calibrated to work out what current in the main cable gives what output of the Rogowski coil. This is best done at the frequency concerned.
The calibration procedure setup in summary. A messy process with lots
of cables, meters, high voltage and high frequency power.
The oscilloscope shot of an actual firing which shows current at 850 kA/div i.e. peak of 15 kA (15,000 A) and 20 us/div time base on channel B. I still haven't found the top half of the can yet after it went ricocheting around my shed.
Now for more flexibility I have made an 'active' Rogowski integrator using a fast integrated circuit op amp and a few bells and whistles.
Left photo shows me working on calibrating an active integrator for my Rogowski coil using a TL072 op amp to replace the previous passive integrator. Middle photo shows the frequency vs output voltage for the active integrator. Right photo shows the completed integrator with a peak and hold detector. The Rogowski coil is the continuation of the white cable around the orange PVC support. The large black cable carries the high current under test. The thinner black cable is 10 turns through the Rogowski coil for testing at 20 A 50 Hz. I have two inputs 100 A/V (full scale 1000 A peak) and 5 kA/V (full scale 50 kA peak). Calibration was at 50 Hz but should be OK to 25 kHz from the graph above. It uses about +/- 16 V supply which is poorly regulated but current draw is only 3 mA. Output waveform is read on the CRO.
I have had a further round of calibration at 50 Hz and have been getting
current readings of 40 kA peak. High frequency output at 18 kHz is
about 10% below that expected from the 50 Hz calibration.
Left photo shows a 2 inch can section braced with plastic to prevent crushing and the Rogowski coil was placed around the section. Right photo shows the Rogowski coil around the 3 turns of the crushing coil.
Left photo shows the can current and the
Right photo shows
the coil current of the 3 turns together. Scale is
5000 A/div vertical and 50 us/div horizontal.
Note that this is a bit artificial in that the can does not crush. There may also be inaccuracies with the Rogowski being in close contact with the crushing coil. With a perfectly constructed Rogowski coil this should not matter however.
Another method to measure coil current is to measure the voltage generated in the can. Simply measuring the voltage across the cans with contacts made at opposite points will enable the current to be calculated. This is not simply Ohm's Law however as the pulse means that the inductance will also contribute.
The CRO shows a 60 V peak across the diameter when driven with the same
voltage and energy above. The vertical scale is 20 V/div and the
horizontal 50 us/div.
Measurement of high voltage pulses
I initially ran my CRO and Rogowski from a 12 V 150 W inverter to
isolate it from the MOT supply which is at a potential to earth. Later I
made a set of dual contacts with a long insulated handle as a DPDT switch to
isolate them. Voltage divider was 27 M ohm (330 K x 82) and 270 k ohm for 100:1
Frank Bitter Sci Am 1965
This page was last updated January 30, 2011