"Mass driver" launcher project / Electromagnetic Projectile Accelerators / Forums

Forums

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

« Previous topic | Next topic »

"Mass driver" launcher project

1 2 3 next

Move Thread

LAN_403

Andyman

Sun Jan 06 2008, 02:50AM

Registered Member #1083 Joined: Mon Oct 29 2007, 06:16PM
Location: Upland, California
Posts: 256

Hello! I'm new to these forums, so if I sound like a newb, forgive me.
Anyways, I'm about to begin construction of the magnetic accelerator cannon found on amazing1.com (I got the Electronic Gadgets for the Evil Genius book for Christmas). I'm not sure what would be the best choice for a capacitor bank. The construction of the project is designed for the use of amazing1's "96uF/4.5Kv capacitor bank or similar". Total possible energy: 972 joules. Total cost: $600!
I was wondering if I would be able to use instead (4) 3900uF/400V capacitors from theelectrostore.com in their place. Total possible energy: 1250 joules. Total cost: Just over $50
I don't have a whole lot of money, and I especially don't have $600. Mathematically speaking, the second option is more powerful, but is there an advantage of having higher voltage rather than higher capacity or something? I'd love to know if I can save myself a few hundred dollars, even if it means sacrificing a LITTLE bit of efficiency.
Thanks!

BTW I'm using an old 5.25" hard drive disc platter for my projectile. I can't figure out how to widen the inside diameter to 2" without going out and buying a lathe, though.

Barry

Sun Jan 06 2008, 07:10PM

Registered Member #90 Joined: Thu Feb 09 2006, 02:44PM
Location: Seattle, Washington
Posts: 301

The most important aspect of a capacitor bank is the stored energy (joules).

Don't worry about the particular capacitance or voltage rating; figure out the stored energy and work from there. A great way to compare parts for purchase is to compute the "price per joule".

The last time I compared parts from surplus shops, I found pricing was all over the map by this measure. Everywhere from US $1 down to $.10 per joule; the cheapest ones were the tiny little 300v photoflash caps but who wants to wire up a few hundred of those.

You won't lose efficiency by deviating from any given design. Indeed, anything you ever build will need tuning anyway. If efficiency is your goal then plan on trying lots of adjustments and different coils and so on. If your goal is maximum speed and impressive results, then keep piling on all the joules of stored energy you can afford.

Electromagnetic launchers will work with almost any combination of capacitance/voltage. For practical reasons the extreme ends of these combinations can be problematic. At really high voltage (several kV) it gets tricky to build a suitable power supply, and arc-over protection takes some careful design. At really low voltage (a few dozen volts) you need huge welding wire for all the connections, and it gets tricky to make a good high-current low-resistance switch. Everything in between is fair game.

To widen a hole in aluminum, try a Dremel tool with a drum sanding attachment and nibble it to death. Dremel tools are my weapon of choice for almost anything.

Cheers, Barry
Barry's Coilgun Design Site

Andyman

Tue Jan 08 2008, 01:45AM

Registered Member #1083 Joined: Mon Oct 29 2007, 06:16PM
Location: Upland, California
Posts: 256

Thanks for the info! I found that my dad's router has a tungsten carbide bit and that works even better than the dremel.
For a switch, I have a thick spring that I managed to wrench a galvanized pipe cap into. The back of the spring is clamped to a board. The other contact is a section of 1/8" steel bar. I plan on separating them with a piece of wood with a string on it, and when I pull it out, the pipe cap hits the bar and closes the connection. I attached my craptastic paint image to show what I mean.

Would it be ok if I used 12 awg wire with a jacket on it instead of magnet wire for the coil? Or would the jacket cause too much spacing between the wires and result in decreased performance?

FastMHz

Wed Jan 09 2008, 02:52AM

Registered Member #179 Joined: Thu Feb 16 2006, 02:08AM
Location: Hagerstown, Maryland - Close to Prime Outlets
Posts: 287

The efficiency will be reduced quite a bit by going with standard insulated wire....magnet wire is *the* way to go for coil winding.

Barry

Wed Jan 09 2008, 06:40AM

Registered Member #90 Joined: Thu Feb 09 2006, 02:44PM
Location: Seattle, Washington
Posts: 301

12awg insulated wire? I say go for it. Build with what you've got, and refine it later if needed.

The worst that can happen is the coil will be a little larger diameter for the same number of turns, compared to enameled copper magnet wire. So you can't pack as many turns under the platter, not a big deal. ~~Barry~~

Andyman

Thu Jan 10 2008, 05:33AM

Registered Member #1083 Joined: Mon Oct 29 2007, 06:16PM
Location: Upland, California
Posts: 256

Alright, I'm going to turn this thread into a project thread. I'll post my progress as I go, and I want you "pros" to give me any advice you have, or tell me whether or not something will work.
So, here come the pics.
The spring contactor switch.

I'm not sure how efficient of a method this is for switching the high current. I think my smaller switch that welded every time worked better. Is there a better way than this besides an scr? I want the circuit as simple as possible (I'm lazy when it comes to soldering and using heat sinks).

My projectile.

It was fashioned from an old 5.25" hard drive platter. The center hole is a few thousandths over 2 inches. My dad had an OLD HP3000 a few years ago, and when he got rid of it, he gave me one of the broken HDs to take apart. Good thing I kept the platters all of these years

The coil housing/base.

As you can see, the center area is recessed, which is where the coil will be nestled. The center hole fits loosely over the flyway (in the top corner). I plan on adding semi-heavy springs in between the plywood disks to reduce recoil. Will this decrease projectile speed? Or is it worth it to prevent the whole assembly from destroying itself by the recoil?

Please, let me know if anything can be improved.

...

Thu Jan 10 2008, 06:20AM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

There are several things wrong with your design...
1. steel is a poor conductor for these things, brass/copper will perform much better
2. you should make the high current connections directly to the electrode, not to the far end of the spring...
3. Bigger wire would also probably be good...

1 and 2 could probably be remedied by replacing that steel cap with a brass one with a nice fat wire soldered to it. The bottom plate should also be replaced with a piece of copper/brass.

As to gettig more caps, there is a pretty decent deal on ebay right now, this dude has 1200 of 2400uf/450v caps ad is selling them for about $.01/J + shippig

Andyman

Fri Jan 18 2008, 12:16AM

Registered Member #1083 Joined: Mon Oct 29 2007, 06:16PM
Location: Upland, California
Posts: 256

Alright, my capacitors arrived and I upgraded my switch thanks to Peter's suggestion (can I call you Pete?). So here they are.
First off, a close up of the newly improved switch:

Now that I have used bolts instead of bent nails, the spring is VERY secure and doesn't appear to have ANY bounce. Of course, I need an oscilloscope to be positive.(no pun intended) The contact is very solid and is absorbed somewhat by the curved copper electrode. After firing a few shots, there is only a tiny pinprick spot at the point of contact. Virtually zero visible spark. (but that might just be because I blink every time it fires

)
I know copper is about 50% more conductive, but the aluminum terminals were MUCH cheaper. The protection diodes will go in between the two terminals. (20 in parallel of 1N4007 1000v 1A rectifiers i got on ebay for super cheap).

The whole cap bank/switch setup:

Each capacitor is 3900uF 400v. Placed in series parallel config for 3900uf 800v 1248 J. I know it's the same amount of energy if I places them all in parallel, but this way it saves me space. The white wire is the negative and is 10AWG stranded wire. The brown is the positive and is (3) 20awg stranded wires twisted together. I would use copper braid, but I don't have any and it's expensive.
Small buss bars are just sections of 1/2" copper pipe pounded flat. The big one in the middle is 7/8" copper pipe split down the middle and opened out flat.
Two questions:
Is there a difference if I split the center plate in half and have two seperate conductors?
If I fire just a few half-charge test discharges without the protection diode, will that significantly damage the capacitors?

...

Fri Jan 18 2008, 12:39AM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

I go by peter

You should be fine with aluminum contacts, you just need copper for the electrodes since aluminum has the Al2O3 oxide coating which makes it a poor choice for a contact.

It will not make much of a difference if you split the center contact, although you should add bleeder/balancing resistors across the caps. Just put a resistor across each of the caps, sized to bleed off the charge in a reasonable amount of time (1 minute is usually what I aim for things like this).

You can fire the caps without the diodes, just watch the reverse voltage and make sure it does not get too high. I would try to keep it under 50v for a bank like that (although electrolytics really don't like reverse voltages). But I had a bank of similar caps that took 20v a cap of reverse charge without showing any signs of trouble.

Andyman

Fri Jan 18 2008, 02:36AM

Registered Member #1083 Joined: Mon Oct 29 2007, 06:16PM
Location: Upland, California
Posts: 256

Can you see the reverse voltage on the meter? Or does it occur too fast?
Also, is it better to have a few, or even one, bigger diode(s) or a lot of smaller ones?

1 2 3 next

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