Third generation railgun project / Electromagnetic Projectile Accelerators / Forums

Forums

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

« Previous topic | Next topic »

Third generation railgun project

previous 1 2 3 4 next

Move Thread

LAN_403

Steve Conner

Mon May 19 2008, 09:45AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Holy sh

That is one of the more impressive railguns I've seen in the years I've been involved with 4hv. Looks like you need a lot more bracing of every conductor in the whole setup though, to keep your Lorentz forces under control. Remember that every part of the circuit feels just as much force as the projectile does.

I imagine the projectile more or less exploded on impact, and the material around the crater is the remains of it. The copper alloy is probably quite soft, and the projectile might have got hot enough to soften it even more. Or maybe like another poster mentioned, the projectile partly melts in the barrel, so the gun actually shoots a cloud of molten copper droplets with a hot slug of copper in the middle

Quantum Singularity

Mon May 19 2008, 12:02PM

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

I agree. Very nice. Too bad you've only gotten two shots. That's the problem I have with railguns and why I abandoned my project... just seems like they don't hold up well. You didn't say if you were able to record any data... velocity/ efficiency? About the copper coating around the impact, I'd say you got a couple things going on, both from the impact and like mentioned from the muzzle flash as well. But I'd wager a lot is from the impact, you didn't punch through, so the full amount of energy was dissipated in the impact. Being that the projectile was probably already pretty hot, I'd think that upon impact it probably partly liquefied and vaporized. How much of the copper did you recover? That should pretty much tell you.

badastronaut

Tue May 20 2008, 04:05AM

Registered Member #222 Joined: Mon Feb 20 2006, 05:49PM
Location:
Posts: 96

For the aluminum projectile, there was nothing left of it except aluminum oxide dust everywhere. The copper alloy projectile seems to be either vaporized or welded to the steel plate. There is no evidence that it survived intact after hitting the steel plate.

I do have oscilloscope data. Peak current is about 200kA, peak power is about 50MW, and pulsewidth is about 600us. I was unable to find the speed of the projectile because the muzzle flash blinded the high speed camera. I am estimating a speed of 900-1500m/s for the copper alloy projectile.

I have attached some high speed footage of the copper alloy projectile shot.
]x7.avi[/file]

Dr. Shark

Tue May 20 2008, 02:32PM

Registered Member #75 Joined: Thu Feb 09 2006, 09:30AM
Location: Montana, USA
Posts: 711

Badastro, I am bowing down in respect to you, that is some really impressive stuff you did. In your posts in the old 4HV you have shared a lot of theoretical understanding about railguns, so would you care to elaborate a little bit about what makes this gun so much better than the usual amateur attempts ("spark shooters")?

It does not seem like you took any measures to increase the inductance gradient in the gun, so probably most of the work went into getting the resistance in the circuit really, really small. I guess your capacitor bank is around 0.5mOhm, and from the way you described hammering the projectile into the barrel I take the resistance of the sliding contact to be of a similar magnitude.

Guessing LÂ´=5*10^-7 and R=2*10^-3 this is definitely one of the better guns out there, but where is your secret? Following your own maths here

At 100m/s you will still be burning about 40J in the resistance for each Joule transferred to the armature, so getting the projectile up to an efficient speed from rest seems like a massive feat to achieve.

Mathias

Tue May 20 2008, 08:09PM

Registered Member #1381 Joined: Fri Mar 07 2008, 05:24PM
Location: Hungary
Posts: 74

What i dont really understand is how come that there's even a bit of the armature is left still be intact after firing the gun ?
I mean , what i have read from other sources whom were using way less energy , if they were'nt using some kind of launching mechanism/procedure , (so that the armature would gain some starting velocity) , then they armature would just evaporate without begining to accelarate .
If i didn't misinterpreted something , then if you have to hammer the armature in it's place to start with then, it should be stuck untill most of it's surface had melted ,am i right? So im guessing alot of the energy should be wasted there

badastronaut

Tue May 20 2008, 09:52PM

Registered Member #222 Joined: Mon Feb 20 2006, 05:49PM
Location:
Posts: 96

Railguns don't need to have any sort of injection or starting velocity. I think that's a myth perpetuated by Powerlabs. Another myth is that the projectile will weld to the rails. This is simply not true under a certain condition.

The projectile is solid metal, and it takes a lot of energy to vaporize metal. From Joule heating alone, this would have to mean that the projectile is made of some highly resistive material, which it is not. If the projectile vaporizes from Joule heating, then the rails would similarly vaporize. If the rails and projectile are made of the same material and have the same cross sectional area, then they would heat up at similar rates assuming no motion and perfect contact.

It is possible to calculate how much the rails will heat up. In this case, they would heat up by about 80 degrees C, which was a factor in choosing the dimensions of the rails.

Remember that not all of the initial stored capacitor bank energy is dissipated in the projectile as heat; only a proportionally small amount of it is. Therefore, it doesn't vaporize.

If there is enough contact force and contact area, then the rail-projectile interface will not melt. It did in fact melt for the first shot with the aluminum projectile. I haven't had a chance to look at the rails for the copper alloy projectile. If the interface melts, I would think that it would enable the projectile to be accelerated more easily because of reduced friction. A lot of energy is being wasted there, but the proportion is small compared to the total energy stored.

As for projectile welding, the only way it could have significant welding is if the current stopped flowing after that interface melted before the projectile gets a chance to accelerate. If that is the case, then the capacitor bank is too small to be used for a railgun with or without injection. A properly sized capacitor bank will accelerate the projectile before the metal has a chance to cool down and weld.

I don't actually know what the inductance gradient is for certain, but I think it is in the range of 3e-7 to 5e-7. The resistance of the entire circuit is 1.3milliohms. That resistance was found by matching data to simulated parameters. The capacitor bank resistance is somewhere around 0.3 milliohms. I think a very large amount of energy is being wasted blowing the rails apart, which makes the effective resistance bigger than 1 milliohm since it presents an extra voltage drop across the circuit. The best estimate of speed I have for the copper alloy projectile is 1200 m/s.

The secret to railguns is being able to deliver a large action integral using as little energy as possible. Action is defined as the time integral of current squared. It is non-linear and not conserved. You can directly increase the action integral by either increasing stored capacitor energy or decreasing resistance or both.

This essentially means that you need to have as small of a resistance as possible. This is the reason why I keep telling people to put their capacitors in parallel for railguns. Any other configuration is wasteful. This the problem with the powerlabs railgun in that he has too many capacitors in series. Increasing inductance gradient is a good thing but it is very difficult to do compared to how easy it is to increase the action integral.

For an even simpler explanation, take a look at the diagram below. Assume that we can model the transfer of electrical energy into mechanical energy with a resister Req. If all of the other resistances are lumped into R, the problem becomes how do transfer a larger proportion of energy into Req relative to R? The answer is to increase the ratio of Req to R. Even better would be to make R zero.

Also, resistance affects efficiency non-linearly. The smaller it is, the bigger effect it has. The Powerlabs railgun supposedly used ultra low resistance oxygen free copper rails. The reason why this doesn't work is because the resistance of his capacitor bank was far larger than the change in resistance of his rails. He might have shaved off 0.1 milliohms by using the special rails, but his capacitor bank's resistance is probably around 10 or more milliohms, hence poor performance.

Ask more questions please.

Andyman

Tue May 20 2008, 09:56PM

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

Wow. Just wow. That is absolutely INCREDIBLE!

That looks more like the impact of a .40 S&W round!!

To answer your question, the material is the splatter of the vaporized copper. The same thing happens with lead bullets at steel plates (I go shooting). When it hits, it turns into powder, and the force of the "explosion" embeds the particles into the surface of the steel. This attachment is probably only a few millionths of an inch deep.
That's my guess.

For the switch, you should put all of the sacrificial SCRs in parallel for current/voltage sharing.

badastronaut

Tue May 20 2008, 09:58PM

Registered Member #222 Joined: Mon Feb 20 2006, 05:49PM
Location:
Posts: 96

Well, the SCR's are too small by about an order of a magnitude or more, and I only have one left.... :P

What kind of gun and ammo does it take to do similar damage to a 1/4" thick steel plate? What does it take to punch through it?

rp181

Wed May 21 2008, 12:29AM

Registered Member #1062 Joined: Tue Oct 16 2007, 02:01AM
Location:
Posts: 1529

Do you think 1 set of rails or multi stage like ea6's would preform better? same energy, multi stage the power would be split between the rails (seperate banks).

Sam barros used a current limiting inductor, do you have such? yours is only ~foot long, is this because you didnt limit the current for a short pulse?

badastronaut

Wed May 21 2008, 01:00AM

Registered Member #222 Joined: Mon Feb 20 2006, 05:49PM
Location:
Posts: 96

I can't prove why multistage would help except to reduce the effect of really long rails. Multistage is much more complicated, and I can't justify using them.

I don't use an inductor, and I can't prove that they are helpful at all except to reduce didt, which can be a problem for solid state switches. The resistance of the inductor would also hurt performance dramatically. For my design, I actually tried to reduce inductance as much as possible so that the current won't oscillate given the low circuit resistance. The total inductance is about 0.2uH. My ideal railgun model says that the performance will be the same inductor or not. Real life factors such as finite rail length and friction changes this answer.

This railgun is actually a modification of a previous railgun. The length was chosen 4 or 5 years ago based on material and funding availability. I didn't want to have another one built, so it stuck.

I think the use of an inductor is another myth created by Sam. He says using them would help, but he doesn't prove why they might help or even give an explanation. Pulse length has no bearing on performance assuming the rails are infinitely long, frictionless, and superconducting. Only the action integral matters to any significant degree. It can also be proven that inductance does not change the value of action.

previous 1 2 3 4 next

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