Improving my railgun / Electromagnetic Projectile Accelerators / Forums

Forums

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

« Previous topic | Next topic »

Improving my railgun

1 2 next

Move Thread

LAN_403

McFluffin

Tue Mar 28 2006, 06:01AM

Registered Member #119 Joined: Fri Feb 10 2006, 06:26AM
Location: USA
Posts: 114

Hello all! Maybe this should be in the projects section, but here goes... I have been assembling parts to build a railgun for the last year or so, and finally shot off a small one on Thursday. However, it didn't work so well, so I am going to try to work on things.
I started off with four 3600uF capacitors that can be charged up to 350V. I used a piece of plastic from an old monitor LCD screen to tie together two pieces of um...copper/aluminum. I didn't have thick enough copper, so I took a piece of alluminum and put a copper jacket around it to improve conductivity and strength of the rails. Charging is done with a half bridge and a variac, which lets me charge it up to around 190V in a fairly short amount of time. I'm not sure how fast I could do it, but I am keeping the variac at 30W during the charging. The large range for it goes up to 1800W, but I had an incident where the fuse exploded(yes, it actually shattered), so I decided that I shouldn't try to charge things up to quickly. Triggering is via an SCR puck. I have three of them clamped in a nice array(common -, but different +s), and a box of 21 more if I need them. This was hte setup that I used to fire my coilgun, so I figured that it was at least worth trying before I started changing things around. I wasn't really sure if the 1 SCR was really sufficient, but it was a fairly small capacitor bank/voltage, so I thought I was fairly safe.
Connections are made between close devices with some fairly thin(maybe a couple of mm) copper. I have several large sheets of it and could probably stack it up if I need to make things heavier duty. Wires are made from some fairly thick cables that come from Toyota Priuses(my dad works for Toyota and they had some sort of recall, so I can get a lot of these. I also managed to get a broken inverter which had 7 IGBT bricks and 3 large 450V capacitors :) among other cool stuff). The wires are maybe .25 inch and are stranded copper. I have used these along with some aluminum buss that came with some of the capacitors to try to further reduce resistance across the capacitors.
I fired the device and nothing really happened. It made a loud pop and the aluminum sort of dissipated. Not real exciting. I then took some copper and connected the 3 SCRs together and wired them up. I also added a voltage doubler and added a second bank of 8 capacitors that are 3300uF @ 430V(Jason might recognize these :)). I am using less than half of my capacitors right now. The system when charged to 350V stores around 2.5kJ of energy. I had seen another amature railgun thread in which someone used steel wool to get bad results, but results nonetheless on their mini railgun, so I decided to try it out. I fired it off and it shot lots of sparks off out the front, but really none to the back, so I was fairly happy that at least things seemed to be moving in the right direction. The steel wool mostly just burned and moved a couple of cm, but not out of the barrel. I could smell burned plastic and the plastic sheet was a little melted. Additionally, the steel wool contact areas gouged the rails. I tried another test with it later, and it did the same thing, so I decided to stop trying to use it.
I then tried several aluminum projectiles. The first projectile was aluminum and didn't work well because it made poor contact. This made a nice flash, but didn't really do much for the railgun concept. These were all fabricated from aluminum foil folded over many times and put into a U shape. When I fired the first one, it moved much further than the steel wool had(no sparks out of the front though), but not out of the barrel. It just wasn't strong enough and tore apart before leaving. I made a third projectile that included a piece of aluminum in the center to try to make it more solid. This time, I blew out an SCR. I am guessing it had to do with not having enogh power to fire them all. I had been warned about this, so I sort of deserve it. I have 12 9600uF caps in a bank that I might use to try to prevent this in the future. I was hoping to make a massive SCR bank out of the box of SCRs, but don't have any plastic to use as an insulator to make the clamp. A fried was suppose to bring some over, but he forgot. I'll try bugging him again.
So, this has led me to several conclusions. The first is that I need to upgrade the SCR firing circuit. The second is that steel wool really shouldn't be used as a projectile as it is too damaging to the device. I tried making a projectile out of carbon, but found it fairly hard to machine. I might also try one of the plasma sabot projeciles using an aluminum backing. Third, I need to work with some augmented rails some more. I didn't really have enough wire to signifigantly augment them. Is there really an advantage of parallel augmented rails? I could try dumping quite a bit of power into a large electromagnet if anyone thought it might be interesting to try(I am afraid too much power into this small device might blow it up from using series augmented with any more power). Finally, I am curious if it would be worth it to run the system at 700 or 800V. I hear the higher the voltage, the better.
I have some pictures, but need to either put a floppy drive on my computer or get a chance to barrow my dads computer. I'll also see if I can get some dimensions. The device is maybe around 8 inches long and 3 inches wide. Perhaps .5 inch thick.

mike0t4ever

Tue Mar 28 2006, 06:59PM

Registered Member #321 Joined: Wed Mar 15 2006, 03:33AM
Location:
Posts: 14

just a quick question
does your projectile start at rest in the rails or is it injested at some initial speed? (injection can be very beneficial)

i'm also curious as to your bank connection layout (the scr's in particular) a circuit diagram would be nice (hint hint)

augmentation can help but only if you set it up properly (you also need to know when to power the rails and when to power the augmenting coils, ideally you want to fire the coil before the rails to let the current and magnetic field build up, but not too soon)

permanent magnets can be an alternative to that (especially with a stationary projectile)

interesting work though keep it up (just be safe)

McFluffin

Wed Mar 29 2006, 02:55AM

Registered Member #119 Joined: Fri Feb 10 2006, 06:26AM
Location: USA
Posts: 114

I drew up a rough schematic that can be found at:

The diodes are 100A surge, 600V rated diodes. There are perhaps 20 on the capacitors, but only 3 across the rails. I have 100 of them, so it would probably be in my best interest to add some more. They are in bunches of three crimped with connectors so that they can be mounted on each capacitor and are mounted on most capacitors. This just seemed a convienent way to mount them and doesn't really have to do with trying to protect individual capacitors as I assume that they should be able to clamp the backlash nearly equally well from anywhere on the capacitor bank. However, I now realise that it is probably much better to put them directly across the rails rather then on the capacitors. Not included in the schematic is also a diode coming out the the voltage doubler. I had some EMF kill the diode in the voltage doubler, so I might add an inductor to try to stop this.
Currently the projectile only rests on the rails. I have some pneumatics equipment, so I might try to inject is pneumaticaly in the future. However, for simple tests its sufficient to just let it start at rest, however inefficient.
As far as safety, I'd definitly like to shield the cap bank in case of failure, but don't have any material to do so. I might go down to my local junkyard and try to pick up some polycarbonate or other clear material. I added a bleeder resistor, which should help in any long term charge effects with the capacitors.
How much do magnets really help? I've heard a lot of debate over this. I have a pile of magnets(some neolydium) that I could try using. One of the main concerns that I have is how I should actually place them. I'd imagine if you don't place them right, you might make things worse.
Thanks for your help!
EDIT: Just fired a piece of alluminum quite successfully. It was launched out and did no damage to the device. Hit a target a few feet off and bounced back two feet or so.

dan

Wed Mar 29 2006, 06:28AM

Registered Member #223 Joined: Mon Feb 20 2006, 06:42PM
Location: Ottawa, Ontario, Canada
Posts: 125

Why are you using SCR's? Just use projectile switching with pnumatic injection. You also don't have to worry about back-emf too much since a rail gun has minimal inductance anyway. Any back-emf that is generated can easily be absorbed by the capacitors.

McFluffin

Thu Mar 30 2006, 12:23AM

Registered Member #119 Joined: Fri Feb 10 2006, 06:26AM
Location: USA
Posts: 114

I don't have enough building material currently to make something for pneumatic injection. So, using the SCRs and just letting it sit will have to do for now.
If you think the EMF shouldn't be a problem, why is that diode blowing up? I checked things out after the last fire, and it died again :( I had three together that time. I tried adding an inductor(a shorted transformer that I had, maybe 4 inch cube iron core) and it only charged up to 190V which is like I wasn't using the doubler at all. However, it went fairly slow which meant that I didn't have to manually turn the voltage up. Also, isn't the back EMF in the other direction which would cause reverse charging of the capacitors? I have been warned to avoid that.

dan

Thu Mar 30 2006, 01:07AM

Registered Member #223 Joined: Mon Feb 20 2006, 06:42PM
Location: Ottawa, Ontario, Canada
Posts: 125

Maybe I was wrong.. At the currents that are involved with a rail gun are so great that even the smallest stray inductance causes back-emf. Paralleling diodes wouldn't help that much since semies doesnâ€™t really like to share the current evenly. Maybe you need to get yourself a hockey puck diode rated for 1000A continuous and several 10KA's peak to clamp that nasty back-emf. (not sure if this is over kill) I'm not sure if this clamping diode across the rails would help much since this stray inductance is literally everywhere in the circuit but the diode would be clamping the majority of it. Electrolytics are bad for rail guns since they are generally very slow and they don't like to be reversed charged. However they are fairly rugged and should be able to take this abuse. I have shorted some of my electrolytic capacitor banks at full charge with a screw driver and they where completely undamaged. Their capacitance shifted by less than +/-0.1uf.(not sure about ESL or ESR though but probably not enough to care about.)

Having the projectile at rest also isn't really good either because it is probably going to partially weld it self on to the rails slowing it down. You really need some sort of injection either mechanical or pneumatic. Just don't pull a Sam Barros and let the injector do all the work. With the caps just vaporizing the projectile and making pretty sparks come out the muzzle but not actually contributing a whole lot to muzzle velocity.

pulslaser

Thu Mar 30 2006, 03:54PM

Registered Member #156 Joined: Sun Feb 12 2006, 07:04PM
Location:
Posts: 23

I think you shouldn't put to much work in injection, especially you should not use selfswitching. In this case you will get an arc in front of your projectil, where shurely should not be. In my last railgun I discharged 12 kJ through the projectil in rest and didn't got any arcing or welding but rather high speed (330 m/s). In my former experiments with injection and selfswitching there was always arcing and heavy erosion of the rails and low speed (< 100m/s).

dan

Thu Mar 30 2006, 08:12PM

Registered Member #223 Joined: Mon Feb 20 2006, 06:42PM
Location: Ottawa, Ontario, Canada
Posts: 125

Wow this is very interesting most amateur rail guns that I have seen not using some sort of injection fail miserably (projectile welds itself to the rails) or don't perform well at all.

It was my understanding that in order to reduce the chance of welding the projectile it had to be injected fast enough. Also in order to reduce the chance of welding the rails and projectile should be perfectly machined to avoid 'high points' which is what causes the majority of the arcing. But that some arcing (and thus rail erosion) was inevitable.

I have noticed that there is very little amateur research with rail guns do to their complex design and precision machining involved so it is interesting to see one work at all. Pulslaser, What kind of switch were you using? At these energy levels Solid State would be very pricey so the only option would be a triggered spark gap or something.

Rail guns seem to be horribly inefficient. This is why there really isnâ€™t any more serious reasearch being done on them anymore. Yes some do get muzzle velocities in the 10km/s range but this is firing a very small tungsten projectile(less than a gram) with several Mj of input energy. So it doesn't surprise me that you only get the projectile out of the muzzle with 1.2kj at only a few hundred volts. I would suggest rearranging your cap bank for more V since this would allow higher discharge currents. I would also get rid of anything solid state.

But the whole name of the game is exparimentation.. Try different things injection vs. stationary, different projectiles, etc. Maybe even using injection and then a few ms later when the projectile is in between the rails trigger the switching device to dump the capacitors charge into the rails.

pulslaser

Thu Mar 30 2006, 10:26PM

Registered Member #156 Joined: Sun Feb 12 2006, 07:04PM
Location:
Posts: 23

I got best results with isolating projectils loaded with copper brushes. The flexible brushes don't need excatly alligned rails with ultraflat surface and at least in my experiments they work without any arcing and even can be used for several shots if they are made from reinforced epoxid. Switching was done with a ignitron tube, for efficient spark gap switching te vltage was too low and I didn't had scr big enough. More images and pictures a found on my pages (

sorry in german)

Quantum Singularity

Fri Mar 31 2006, 03:22AM

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

Ignitron tube... I may have to look into this because if I ever get back into my railgun project I will be using 12kJ cap too. Is an ignitron that can handle that kind of power rare and expensive? And what V levels do you consider a spark gap to be ok? My big 2 caps are 10kV along with a few smaller 10kV caps too. I figure spark gaps are probably fine at these levels?

1 2 next

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