Choosing Capacitors for Railgun [was "Greetings..."] / Electromagnetic Projectile Accelerators / Forums

Forums

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

« Previous topic | Next topic »

Choosing Capacitors for Railgun [was "Greetings..."]

previous 1 2 3 4 next

Move Thread

LAN_403

Dave Marshall

Mon Feb 20 2006, 05:12PM

Registered Member #16 Joined: Thu Feb 02 2006, 02:22PM
Location: New Wilmington, PA
Posts: 554

Joe Doh, your line of reasoning, while logical, doesn't pan out in the real world.

Higher is NOT better. The higher the capacitance of an electrolytic capacitor, the higher the ESR (equivelant series resistance). In very large capacitors, ESR can approach and exceed half an ohm. This isn't the case with inverter grade capacitors, but its still far greater than any energy discharge capacitor.

Capacitor failure is also FAR more likely with electrolytics. Energy discharge caps are constructed to withstand the forces of intense discharges. The plates of the capacitors in electrolytics are typically thin aluminum. When the capacitor is subjected to a high amplitude pulse, similar in principle to a ring launcher, the plates can experience physical stress. This can often lead to a plate being sheared from its terminal, a punch through of the diaelectric, even failure of a capacitor in a nearly explosive manner. I had several such failures in a large coilgun I constructed in high school.

Its true that energy discharge capacitors often can have too brief a discharge, reducing efficiency and turning the railgun into a glorified spark shooter. However, this can be countered with a pulse shaping inductor to make the pulse behave precisely the way you want it to, with minimal overall loss of energy.

Arguing that electrolytics are more desirable is like arguing that a rubber mallet is a better tool for driving nails than a claw hammer. You might still drive some nails with the rubber mallet, but its usable life is dramatically reduced, and its far more effective to use the claw hammer, even if the claw hammer costs a bit more.

Every successful railgun that I've read about has used energy discharge capacitors. I've seen three using electrolytics, and the results were far from impressive. Lytics are more affordable, yes, and they are likely safer for a beginner's first try, but they are in no way the most desirable tool for the job, and I'd imagine Erg would like to know the benefits and disadvantages before sinking his money into components.

Dave

Dr. Shark

Mon Feb 20 2006, 05:29PM

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

Well, theory i fine, but I like to check with reality ocasionally. The only working amateur railgun I am aware of

is using electrolytics, and the author explains very well why he is doing it (sorry, its in german).

Edit: Link fixed

Dave Marshall

Mon Feb 20 2006, 05:48PM

Registered Member #16 Joined: Thu Feb 02 2006, 02:22PM
Location: New Wilmington, PA
Posts: 554

I should clarify some. I gave the impression that a good lytic railgun isn't possible. Its been done by a few different individuals/groups. None of them were simple systems though, they required extensive experience and considerable resources.

In any case, big general purpose electrolytics are not in the least bit suitable for the very high peak currents you want for a railgun. They literally tear themselves apart. An ideal bank of electrolytics for a railgun would be a 'cube' constructed of somewhat smaller pulse rated electrolytics. Several banks of caps in series to get the voltage up, then several of those banks in parallel to bring ESR down. This gets rid of most of the major issues with electrolytics in pulse duty, HOWEVER this is no inexpensive proposition. Such a bank of appeciable energy capacity would cost many hundreds of dollars.

General purpose caps can be used, and would probably work to some degree, but in the end, you'll pay more replacing failed capacitors, and suffer the disadvantages of crummy peak current capability.

Dave

badastronaut

Mon Feb 20 2006, 06:02PM

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

Dave, you seem to have an obsession with increasing the working voltage of the capacitors. Please explain your rationale for using a higher voltage. For example, explain why it would be better to wire two identical real capacitors in series rather than in parallel.

You're also wrong about the relationship between electrolytic capacitance and associated ESR. A higher capacitance actually is associated with lower ESR for individual capacitors. Just look at some of the datasheets on Cornell Dubilier's electrolytic capacitors.

Yes, individual photoflash caps have high ESR, about 1 ohm, but large can capacitors have ESR's around 50-20 milliohms.

Marko

Mon Feb 20 2006, 06:44PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Yes, individual photoflash caps have high ESR, about 1 ohm, but large can capacitors have ESR's around 50-20 milliohms.

Thats what I talked about for long, nobody listened

max current depends much more on ESR than actual resistance of entire system (for electrolityc caps).
People told me also to wire all in series and get most voltage, but actually more peak current can be gained by wiring in prallel (to some limit) because then they can give twice max current of 1 capacitor.

Normal dielectric pulse capacitors have incredibly small internal resistance and thus are very good for this kind of use, actually they are more benefical in series as ESR is very small compared to external resistances.
Here we can get real boost in peak current by seriesing caps and getting more voltage
(for railgun pulse can even sometimes be too short so this is not always good)
...

Dave Marshall

Mon Feb 20 2006, 07:22PM

Registered Member #16 Joined: Thu Feb 02 2006, 02:22PM
Location: New Wilmington, PA
Posts: 554

I'm no railgun expert, but from the research that I've done, the key to getting a railgun to do its thing is to keep the projectile from sticking to the rails, a feat more easily accomplished if you can get a small portion of the projectile to vaporize. This is usually accomplished with an injector of some sort, and high voltage/peak current. Railguns thrive on high peak current, there is no denying that. High peak peak current is achieved with high voltage, and low resistance.

Capacitance has little to do with ESR overall, HOWEVER, physical size has everything to do with it. A 1000uF, 100v capacitor the size of a film canister is going to have light weight plates, thin connections to the terminal, and small terminals. This lends itself to a high ESR. A 1000uF cap the size of a soup can would have the room for heavier conductors, bigger terminals, and more robust connections to the terminal, equating to a lower ESR. You could have a 1000uF 10v capacitor the size of a film canister, and a 5,000uF 500v capacitor the size of a soup can that have the same ESR, but I have never seen a high value capacitor the same size as a lower value cap, with a lower ESR.

Edit: I think maybe some of the confusion stems from the fact that there are infact different types of railguns. Plasma aramatures are the easiest to construct, and require the least specialized resources (precision machining equipment etc). I expect, being his first try at a railgun, Erg would probably lean towards a simplistic topology. Plasma armatures are as I described above. A small portion of the projectile (and rails, though that isn't really desirable) vaporizes, and the projectile essentially rides a cushion of plasma down the rails, preventing your railgun from turning into an expensive tack welder. This is more easily achieved with voltages higher than typical electrolytics can handle repeatedly. Like I said, electrolytics can work, even for a plasma armature, but general purpose electrolytics will not. You need specialized inverter grade electrolytics that are not exactly cheap. Small ones in the 200j range that I've looked at in the last couple hours start at about $30 each, and only go up.

Dave

Steve Conner

Mon Feb 20 2006, 07:36PM

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

Here is my $.02 worth. I'm no railgun guru but I know a lot about capacitors:

It seems that electrolytic capacitors can be used to make a good performing railgun. Jason Rollette and Thomas Rapp both use electrolytics and have probably the two best railguns on the net. Rapp's gun shoots no sparks whatsoever and launches projectiles at the speed of sound. BUT......

Not any electrolytics will do. Both of their guns use the best inverter grade capacitors. These are specifically designed to carry heavy currents, and are bigger than ordinary lytics of the same rating, with very big hefty terminals. These are expensive new, but you can usually find them quite cheap on eBay. They are used in DRSSTCs too. If you use smaller capacitors like you find in a camera flash or TV set, their ESR will probably be too high to shoot anything other than their own innards.

For instance: I recently bought 12 3300uF 420V capacitors for a DRSSTC project. They are slightly bigger than soup cans and have an ESR of 0.026 ohm IIRC. Hence each one would deliver about 16,000A into a short circuit, or 8,000A into a matched load. If I connected all 12 in parallel I imagine I could launch something somewhere

I paid about $100 for the bank of 12: they were surplus stock from a company that made HF spot welders that went bust.

BTW- Rapp's clever projectile design (a plexiglass bullet with thick copper wire brushes) may help the efficiency since it doesn't shoot sparks. A higher voltage power supply might show more of an advantage with regular projectile designs that lead to the formation of a plasma armature with its extra volt drop.

Eric

Mon Feb 20 2006, 07:58PM

Registered Member #69 Joined: Thu Feb 09 2006, 07:42AM
Location:
Posts: 116

I think electrolytics can be made to work in a railgun. The thing you have to look at when selecting capacitors is the RC constant of a cap, its ESR times its capacitance. This gives you a measure of its minimum discharge time (when short circuited). Just because a cap has a lower ESR doesn't make it better, the RC measure is what essentially tells you how fast you can extract energy from the cap. There's no way to arrange caps to improve this time limit.

The highest power is available in 2*ESR*C, with a load that has the same resistance as the ESR, although in that case half of the energy is wasted heat. Ideally you want the cap RC to be much smaller than the timescale of your planned discharge, that way more of the cap energy will be delivered to the rails instead of internal heating.

Big oil caps are great but they are actually much higher performance than you need for a railgun. These things are made for *really* short discharges, the railgun timescale is much longer.

Oh, and one other thing, as far as voltage goes, 300-400V is an ok range to be in for applied voltage to the rails, you do not want/need to go higher assuming you have a somewhat optimized design.

Jasonr

Mon Feb 20 2006, 08:57PM

Registered Member #167 Joined: Mon Feb 13 2006, 06:41PM
Location: Waterloo, WI
Posts: 54

Well let me jump in here with my 2 cents. You are all correct when only talking about a small portion of the over all system.
First you have to think of the cap bank all by it self. The limiting factor of discharge current is ESR. So the one way to lower this value is to parallel caps.

The next factor is if you are using electrolytics then they do not go much higher than 550volts so if you need a higher voltage then you need to series them. But when you series them you end up with the problem of adding ESR! So now you have the chicken or the egg scenario.

The next is the over all resistance of the entire system. If your system has a .5 ohm resistance from the bank to the to the rail and back. ( this includes the armature as it skips down the rails so this is low!)

So if you have a bank of qty 10, 1500uf 500Vdc caps 1875 joules and your system has a resistance of .5 ohm. Your peak current is around 1000 amps. But if you half your capacitance and series parallel them, in the 2 series and five of those in parallel arrangement then your current goes to 1850 amps. Now you say I did not take into account ESR. Well you are right?!?!?! Because the esr of the first scenario is virtually nothing because the ESR is lowered with every cap you add. So if you double your ESR that was virtually nothing it does not really affect your current too much.

The limiting current tends to be your rails and connections, to over come this you use a higher voltage.

pulslaser

Mon Feb 20 2006, 09:38PM

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

Hi
like Steve and Joe told I have made very good experience with electrolytic capacitors and unless you going to use a plasma armature 200.. 500 V charging voltage is sufficient. My results with high voltage pulse capacitors where very dissapointing. Of course you get very high puls currents but only for some 10..100 microseconds, useless to make the rails longer than some centimeters, even with the speed of sound the projectil moves just 3mm to 3 cm. On the other hand the extremly high currents predicts difficulties with arcing, switching. Lower currents but longer duration are easier to handle. I found the most important point ist the switching of the current. Selfswitching with any type of injector results in arcing again, and may change youre railgun in a thermo electric accelerator. So my advise: use electrolytic capacitors as much you can get ( I bought mine in ebay 5 Euro each) and use a switch (ignitron, thyristor) to close the current path. By the way, german scientist build a working rail gun during II world war with a battery of lead acid cells

previous 1 2 3 4 next

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