ETG Development / Electromagnetic Projectile Accelerators / Forums

Forums

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

« Previous topic | Next topic »

ETG Development

1 2 3 4 next last

Move Thread

LAN_403

DYI

Wed Feb 11 2009, 02:15AM

Registered Member #1917 Joined: Fri Jan 09 2009, 02:38AM
Location:
Posts: 62

This is going to be an ongoing log of my research involving electrothermal guns. For those who prefer to skip the big words and go right to the pretty pictures, they're at the bottom of the post.

I'll start off here with a little background: I've been interested in ETGs since before Larda posted his 29kJ monster on Spudfiles. My interests have always been directed toward propelling things at the highest speeds I can, and ETGs offer an interesting opportunity in that field: other than the mechanical properties of the chamber, projectile, and barrel, there is really very little limiting the speeds that can be achieved with them, unlike railguns, coilguns, or any chemical propellant systems - when you input more energy, you get more out. My first attempt at an ETG used a mechanical switch and a few camera flash capacitors connected in parallel and couldn't even push the wadding out the barrel. I later obtained a 450V 4900uF electrolytic capacitor and a 1.2kV 7.2kA SCR, and this allowed the old gun to shoot 0.12g airsoft rounds at around 180m/s after tweaking. At this point, my interests and money were directed elsewhere, and when the SCR fried, I put away the ETG.

In December, I regained my interest in the project, and came armed with slightly more knowledge than before. I ballasted a MOT with a 300W lightbulb as a power supply, found four old 400V 3900uF 'lytics, and put together a new airsoft ETG with a more solid design (1/4" machined yellow brass nipples and coupling, 0.257" ID seamless SS tube for a barrel, 1/4" SS rod for the electrode with copper busbar and 4AWG conductors). After the pain and suffering of getting the trigatron switch to trigger at such low voltage, I tested at 400V for a while before I got my 20kOhm 12W bleeder resistors. This produced interesting, but not altogether impressive results. With some work, the gun could just barely pierce 3/16" plywood with one of the 0.12g spheres. After getting the resistors, I rearranged the bank so that all four caps were in series, and tweaked the design extensively. I've tested salt water, tap water, Al, Cu, and steel as "propellants" (I only have 1.6kV right now, not enough to bridge the 3mm gap between the electrode and the chamber wall, so I need something to start up the arc) as well as various wadding methods. Aluminum seems most effective at the moment, as well as being much less messy to work with than salt water. On the better shots (performance degrades quickly via the deposition of oxide layers on the electrode and chamber and the ablation of the spark gap electrodes, and the whole thing needs to be cleaned regularly) the 6mm diameter projectiles will punch 15mm diameter entry holes into plasticine targets, which they can travel about 16mm through at the maximum.

Although this project is still in its infancy, I've gleaned quite a bit of potentially useful information already:
-Plasticine is a VERY GOOD target choice for SMALL guns (the 100kJ muzzle energy gas gun I built in the summer would have gone through a lot of plasticine...). The impact marks on the targets can tell a lot about what happened during the impact. Some pellets will shatter on impact and leave very wide but shallow holes, others will penetrate almost three times their own length and remain somewhat intact (although they do expand by about 1mm around their diameter and flatten into disks). As I lack a chrony, or any other way of accurately measuring speeds at the moment, I can tell which loading parameters extract the most efficiency by observation of the crater size. It's relative measurement, but that is better than no measurement at all.
- El Cheapo 0.12g "Airsoft" rounds are not particularly consistent in their manufacture and not effective as high speed projectiles. 0.25" Delrin rod or spheres would be more suited to the purpose, and I intend to acquire some as soon as is possible (Delrin has extremely high impact resistance and shape retention, and was the choice of insulator in the "sparkplug" for my HE gun project).
-Low voltage ETGs are very finnicky things - while a variation of 1% in the ratio of the fuel:oxidiser mix injected into a combustion gas gun makes no significant difference, small changes in ETG loading can affect muzzle energy by a factor of more than 2. Oxide buildup after a few tens of shots on the center electrode in this test device dropped its performance to the point that it was barely more powerful than a high-end commercial airsoft gun.
-Electrolytic caps should be avoided in favour of pulse capacitors for ETG use. There's no way that any shot I have taken yet has topped 5% overall efficiency (60J). That would equate to 1000m/s. Conveniently enough, I still have a target from the summer that was hit by an airsoft pellet that was measured to be traveling at roughly 1000m/s, so the comparison can be made directly, simply by firing at that target and comparing the effects.
-The "propellant", while important, is really only there for the purpose of starting up the plasma production. After the plasma is established, the remainder of the energy probably goes toward making it hotter. I obviously can't determine this for certain, but a very small (2mm x 15mm) strip of foil (the smallest that I can get to "stick" in the chamber and make solid contact with the electrode) is sufficient for full performance, and performance decreases noticeably with large amounts of conductor present in the chamber.
-For some reason, a slightly damp chamber and electrode is superior to a completely dry chamber or a very wet/completely full of water chamber, assuming that a constant amount of foil is used. Theories in regards to this phenomenon would be appreciated.
-Slight leaks in the chamber cut muzzle energy dramatically. The pressure isn't there for very long, but it's also very hot in the chamber, meaning that the plasma can escape through places that it isn't supposed to very easily.
-While there is probably ringing (I don't have the instrumentation to tell), there is no voltage reversal to speak of in this design. Caps always end up charged to between 200V and 300V after firing. So it would seem that I'm discharging at least 96% of the stored energy through the circuit. How much is being dissipated in the switch? I've no idea. There seem to be a lot of ~50% efficiency figures for trigatrons floating around, so perhaps I'm actually putting about 600J into the chamber of the gun. If these numbers are correct, then 1000m/s would be more like 10% efficient conversion of electricity to kinetic energy.

This brings me to where I am now. I hope to acquire a Chrony in a few weeks for measurement purposes. Pulse caps are definitely in the works, and I'd like to try some non-conductive propellants as well as play with the barrel lengths after I get some real speed data. One interesting idea I've come up with is to use a chamber pressurised with helium, with an insulated electrode in the middle. The pressure puts the breakdown voltage of the gas just above the length of the gap, and to fire the gun, the pressure is dropped via a crazy-fast burst disc valve into the barrel. This would have the beneficial effects of eliminating the switch (and by extension, the switching losses) altogether, as well as pre-accelerating the projectile to some extent before the effects of the pulse hit it, hopefully helping to keep it relatively intact. I am aware that this will require a long barrel, and a LOT more power than I currently have at my disposal to work properly.

And now for the pretty pictures (which should be pretty self-explanatory). No vids as of yet, because without a high speed camera or a massive muzzle blast, there's really not that much to see. All of the impacts shown were done with 0.12g airsoft rounds.

1234318369 1917 FT0 Airsoftcarnage1resized

1234318369 1917 FT0 Airsoftcarnage3resized

1234318369 1917 FT0 Airsoftcarnage4resized

1234318369 1917 FT0 Airsoftcarnage5resized

rp181

Wed Feb 11 2009, 02:54AM

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

As far as electrolytic capacitor's being "slow", i do not think that is your main problem. The main goal right now should be to get a high density LOW temperature plasma in order to conduct the power more effectively. It would probably help if the explosive was enclosed in a low thermal conductivity capsule, but weak enough that the force of the bank will break it. Carbonization of the propellant is bad, it lowers its yield. Look for some research paper's, they are very helpful.

As for a damp chamber making it much more power full ( I have found that the best is to wash it with water, dry the outside, and shake it to get the interior water out. Water in the barrel helps too), It may have something to do with hydrogen oxygen production from the temperature. It also may efficiently convert thermal energy to pressure.

When you use SCR's, beware of the di/dt. I suggest making a rogowski coil.
You need to add some small diodes on there (not to refer to anything). Ring back sometimes happens. I have had a 50v reverse voltage with 6 of those capacitor's. when you move on to higher power version's, clamp the wire's securely.

On another note, This has nothing to do with electromagnetic stuff =p

DYI

Wed Feb 11 2009, 04:02AM

Registered Member #1917 Joined: Fri Jan 09 2009, 02:38AM
Location:
Posts: 62

You don't seem to have read the post very well

There are no explosives involved here.

I have found the exact same that you did in regards to the water - wash chamber interior (removing oxide buildup and assorted other crap), then shake out the chamber to remove any water lying around, then insert Al. foil strip.

As to reverse voltage: I don't intend on going any bigger than this with electrolytics. The only way that I see reverse charging happening with such low voltage caps is using a conductor in the chamber so large that it doesn't even vaporize. Larda hasn't had reverse charging problems, and he's running about as much energy as both of our designs added together and multiplied by ten. I also don't intend on messing about with semiconductor switches - a design like this should concentrate on internal ballistics, and eliminate as many other potential problems as possible. In my book, when compared to solidly built mechanical things, semiconductors count as "potential problems". The pulse capacitors I intend on using for the next stage of the design are rated for enough voltage reversal and a long enough life that I won't be investing in monstrous diodes unless I find some really good deals.

I have to disagree on the pulse length thing considering that, even now, with relatively low muzzle speeds, the projectile is only in the barrel for about 1ms. Electrolytic caps usually have discharge times in the millisecond range. Now what happens if you want, say, 2km/s? A quarter millisecond. I don't know enough about electricity to know for sure, but I'd be willing to place money on the fact that this gun will be more efficient with a 7.5kV pulse cap than it is right now.

And finally, the section this is placed in: I know it isn't strictly electromagnetic, but it is a projectile accelerator, and there have been numerous other ETG threads in this section.

rp181

Wed Feb 11 2009, 04:10AM

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

Its just soooo long =p
My excuse will be i was giving suggestion's for your future ETC =p
As for power, look at what i got:

So lardas is only 6 times as powerful =p

If you are interested in pulse caps, check out general atomics energy products. Very good quality, and price is not that bad contrary to what people think (where else can you get a 201kJ pulse capacitor for 20,000$ and 500$ for shipping in the US?)

Electroholic

Wed Feb 11 2009, 06:48AM

Registered Member #191 Joined: Fri Feb 17 2006, 02:01AM
Location: Esbjerg Denmark
Posts: 720

if you think they are slow, why dont' you put them in parallel?

DYI

Wed Feb 11 2009, 10:06PM

Registered Member #1917 Joined: Fri Jan 09 2009, 02:38AM
Location:
Posts: 62

GAEP's prices do get quite reasonable at higher power, but have you ever asked them how much a 2kJ capacitor costs? The per-joule cost goes up by a factor of ten. And I really hope that you try that 5.6kJ bank in an ETG too - considering that you now have 4.5 times as many caps as I do, you might be able to get an airsoft round over 1km/s

And to answer Electroholic: while it would seem as though the reduced ESR at 400V would allow for better performance, this is not the case. All four in series is noticeably superior. Perhaps when they're all in parallel the arc doesn't conduct as well? I really don't know the reason behind it.

mnr

Wed Feb 11 2009, 11:04PM

Registered Member #996 Joined: Sun Sept 09 2007, 06:17PM
Location: Canada
Posts: 16

DYI wrote ...

And to answer Electroholic: while it would seem as though the reduced ESR at 400V would allow for better performance, this is not the case. All four in series is noticeably superior. Perhaps when they're all in parallel the arc doesn't conduct as well? I really don't know the reason behind it.

Your circuit can be modeled as a simple RC discharge circuit (stray L will not have a very pronounced effect on your current setup). When the capacitors are in parallel, this will increase the total capacitance, resulting in a longer discharge time constant. With capacitors in series this will give a higher peak power by allowing for a shorter discharge time (assuming the capacitor ESR is much less than the rest of the loop resistance). It can be inferred from your experimental data that the resistance of your connections, bus bars, chamber, and propellant is much greater than all of your capacitors ESR's added in series.

rp181

Wed Feb 11 2009, 11:46PM

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

All the parts for a railgun arrived (garolite enclosure!) so that will occupy my time. Ile try it when i get tired of sanding though.

EDIT: I tried it with my setup i told you about earlier, at 330 volts. My camera died right before doing it, so no video. This was LOUD. After the shot, the vent pipe's where audibly shaking. The 12 AWG copper electrode completely vaporised, and before one electrode could be used for a long time with sanding.

DYI

Thu Feb 12 2009, 01:16AM

Registered Member #1917 Joined: Fri Jan 09 2009, 02:38AM
Location:
Posts: 62

Excellent!I hope you do it again, and get a video of it. What were the target and projectile for that test, and how did they fare? Would that be 3.8kJ of energy you discharged there?

I'm surprised that my caps have lasted so long, considering the fast discharges and use at 430V... I suppose that the infrequent use and lack of voltage reversal probably contribute to their continued survival.

rp181

Thu Feb 12 2009, 01:24AM

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

Dry shot, just foil dropping into the electrode, And yes, 3.8kJ. After the shot, there was 73v left in the bank. Camera is charging right now.

Your capacitor's will last quite awhile, The fact that there not being rapid pulsed helps alot.

On a offtopic note, certain people have been wondering where you where, so if anyone asks, your in narnia =)

1 2 3 4 next last

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