Author Topic: In Response to Dave's SloMo Video Here's The Box'o'Bangs (warning, lots of pics)  (Read 3112 times)

0 Members and 1 Guest are viewing this topic.

Offline lemming

  • Contributor
  • Posts: 12
  • Country: au
  • A Lemming is as Lemming Does
    • HSBNE
One of the Caps

Each one of the 16 capacitors in this bank are like this one.
2500V 40uF. 16 Caps gives us a total of 640uF or 2180J
We were told these come from large scale industrial photo-flashes, but being an ebay score have no real idea. In this shot we'd wired them up with some spare hookup wire for testing... we ended up blowing the discharge lead apart...

Beginning to solder up the bank properly.

Having realized just how much current we were dealing with, we decided to solder up the bank properly. Originally some of the caps did have little wire clips on the post and can, but these were so dead as to be useless.

More Soldering

To put it in slightly more simple terms. In this configuration the bank is capable of outputting about 4000A at 2500V or 10'000'000 Watts for a split second. We have footage at 1200 FPS with the discharge being over in less than 1 frame.

More Soldering

Closeup of one of the Cross Bars

We didn't actually have enough heat to get a solder joint between these beastly cables to stick, and wanted to reinforce the joint anyway. Solution was taking a strand from another piece of the cable and lacing it around each joint, this has worked well and has so far lasted thousands of shots.

Completed with Insulation panel and chicken stick

This is the first shot of the completed bank, if you look closely you can see a laser cut sheet of plastic used to insulate the positive rails from the negative rails.

In situ for the first firing.

First fire with the upgraded bank.

This is a bottle cap instantly spot welded to a piece of sheet steel. You can see this happen (in 400FPS Glory) below:

Test discharge on a piece of blank PCB

We've found this is a reasonably good indicator of the strength of a bank.

Series of test shots.

Going from left to right, couple of test shots from he first thinly wired bank. Test shot after rewiring with the thick cable. Test shot after replacing the diode bridge with a better one.

These are the parts of our charging setup.

From left to right. 2 Kilo Ohm 200 Watt dump resistor, Dump contactor, 68Ohm charging resistor and the 9:1 step up transformer. Please note this is the only pic of it I have, it's been made far FAR safer than is apparent in this pic. This is how it was delivered to me.

The Variac

Because if we just rectified the power out of the transformer we'd be severely overvolting the caps, we use this variac to dial down the input voltage. We're normally running at about 200V on the input.

Diode Bridge

Inside the project box you can see there is a Bridge Rectifier. This was custom built out of 12 1000V rated diodes and has so far lasted through probably a thousand charge/discharge cycles. The box is oil filled for insulation and heat absorption

Repairing a solder joint.

We discovered one of the solder joints on the caps had broken, it took 3 soldering irons and a hot air gun to get enough heat into it to get it to flow again so we could rework it.

A very, VERY rough circuit diagram done for a friend.

Upper left is input from the variac, lower left is the discharge contactor.

Fuses... A.k.a. Guinea pigs :-D

We make these little beasts go pop in spectacular fashion.

The remains of a fuse.

This is what happens after you discharge the bank through them...

A pencil about to be disintegrated.

Here's another pencil being popped in an earlier version of the blast chamber.

Axial Electrolytic after being popped.

T8 9" tube before discharge.

T8 9" tube after discharge!

CCFL tube from an IBM laptop.

This was the backlight out of an IBM Laptop LCD Display. This has to have been one of the most unexpected results we have seen.

A very bent, but still intact CCFL tube from an IBM laptop!

We were incredibly surprised to see the result of this one, we thought for SURE the tube was just going to vaporize... but no apparently it handled the full discharge (and probably some more from the charging circuit). Going frame by frame on the video (which you can see below ) it lasted ~37.5ms before it became soft enough for the envelope to fail and then fall into the shape seen here.

Closer shot before detatch.

What's even MORE amazing about this, is the wires survived! those tiny little hair thin wires (like 32AWG or 0.1mm) survived the HUGE amount of current going through them. We estimate it to be about 40A

Tube detatched

In this shot you can see the tube itself and how bent up it is, this AMAZES me that it became pliable enough to do this without shattering or failing.

Oh yes, we launched cans as well

This is a little coil made out of some spare wire and some moulding silicone. You can see one of the cans just leaving here.

Our Math says the can leaves at about 85 Km/h

Hole Punched Can

This is what happens to a can if you place it on a conductive surface connected to the negative side of the bank, and then touch the positive probe to the can. The can also flies a fair distance.

A very popped electrolytic.

This one gave it's all in the line of duty and popped in rather spectacular fashion as evidenced here by the complete peeling open of the petals.

The Next Stage

We're working on getting the bank to be self contained and a lot safer to use for demo's etc on this trolley.

All 4 sides of the trolley will be enclosed with perspex, and with interlocks on the doors. Open a door and the bank is immediately dumped across some BIG resistors.

Also planning to have the blast chamber on the top interlocked so you cannot charge the bank without the covers in place.

Along with that we're also adding various other safety systems such as a power interlock (no power, bank is connected to discharge resistors) and a wireless firing trigger.

The firing trigger will trigger the new air switch we are building, of which you can see a rendering below.

The Air Switch.

Blue is the air piston, a 100 PSI 50mm stroke with spring return. Grey is the retaining nut for said air piston. Yellow is the brass contacts. One moving, one stationary with a threaded section to allow us to adjust the gap. Yes, Gap, because if these two touch they would likely instantly weld together! White is 20mm thick HDPE out of which the whole thing will be built.

Start of the build.

This is the current state of the two contacts, they're 1" diameter solid brass bar, the stationary contact on the left has had an m22x1.5 thread cut on it, just need to spend some more lathe time threading the moving contact and then mill a flat for the 1" braided copper grounding strap that will connect it to the bank.

Comments and criticism are welcome, but please be aware while this is incredibly dangerous and there's some dodgy looking wiring etc here. We do have safety procedures in place for using this, and are working on improving those even further.

In over 2000 firings of this bank, there has not been a single injury, and I plan to keep this record intact for now and forever.

We also have another bank in the works thanks to a member of this forum, it should be 0.12F (yes 120'000uF) at 350VDC. It should give us about 3 times the potential energy of this bank...
« Last Edit: March 16, 2016, 01:43:54 AM by lemming »
The better it felt when you said it, the more likely it is to get you in trouble.

IRC is just multi-player Notepad.

Offline HAL-42b

  • Frequent Contributor
  • **
  • Posts: 423
That might come in handy for putting a few multimeters trough their paces.

Offline German_EE

  • Super Contributor
  • ***
  • Posts: 1851
  • Country: de
For soldering large wire like this I have 'Thor'. It's a 500W soldering iron that has a working temperature of 300 C and I have yet to find something that it can't solder. 25mm copper pipe is easy and a copper plated chassis is no problem so your little wires should be easy.
Should you find yourself in a chronically leaking boat, energy devoted to changing vessels is likely to be more productive than energy devoted to patching leaks.

Warren Buffett

Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo