large coil gun / Electromagnetic Projectile Accelerators / Forums

Forums

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

« Previous topic | Next topic »

large coil gun

first previous 3 4 5 6 next last

Move Thread

LAN_403

Shrad

Thu Feb 26 2015, 07:18PM

Registered Member #3215 Joined: Sun Sept 19 2010, 08:42PM
Location:
Posts: 780

of course with a ferromagnetic projectile ;) or of a construction which would induce a reverse field with the small electrical path involved

Uspring

Thu Feb 26 2015, 07:56PM

Registered Member #3988 Joined: Thu Jul 07 2011, 03:25PM
Location:
Posts: 711

DerAlbi wrote:

-> So what exactly do you think would be the propellant force ? At least pushed or pulled? Magnetic or electric?

A transmission line does not have to be a straight piece of wire or a pair of such. Think of a number of inductances connected in series. Between each inductance a capacitor is connected to ground. The first inductance is connected to a pulse generator. This constitutes a transmission line in the limit of many caps and inductances.
The inductances are solenoids arranged just as in a multi stage gun. The behaviour of this line is, that a pulse applied will travel down the line with the speed of 1/sqrt(L*C), where L and C are the inductances and capacitances per unit length. A projectile inside of the solenoid will see a travelling magnetic field and be accelerated if located at the tail end of the field pulse.

-> how do you think to jump on a wave that travels near light speed? (You really think thats like suring on water??)

I've given the speed above. It can be adapted to any speed you like by choosing L and C appropriately. Speed can be made non uniform to account for the acceleration of the projectile.

- even more interesting: what makes you think this would be specially high efficient???

The reason is, that I (perthaps falsely) believe, that only a small percentage of the energy in the magnetic field is conferred to the projectile. Possibly this is due to the fact, that the field has to be quenched in order to avoid suck back. In a transmission line type circuit, quenching wouldn't be needed.

DerAlbi

Thu Feb 26 2015, 07:57PM

Registered Member #2906 Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 727

I dont see how a water capacitor can be used in a homade design.
The paper/book you linked describes a 80kJ pulse with a Cap made of 1670gal of Water. This makes an energy to weight ratio 80g/J. Electrolytics are 40x better than that. Foils not as much, but too lazy to dig for data.. i suspect a factor of 10..25 over water.
Unfortunately that makes it impractical.
btw: 1.7cm/ns is 1/60th of the speed of light.

Uspring:
First of all, a transmission line consosts of DISTRIBUTED components. There is no L and no C. ther is only a L/m and C/m. You are talking about a discrete delayline - ok so far.
Second: the reach the last delayelement the Pulse must go through every delayelement before. So in terms of source impedance the last coil is supplied with the summed ESR of all delayelements before. This cant be benefical. Specially given the point, that the last coils must actually conduct more current, than the first coils... thats kind of backwards.

Uspring

Thu Feb 26 2015, 08:36PM

Registered Member #3988 Joined: Thu Jul 07 2011, 03:25PM
Location:
Posts: 711

Second: the reach the last delayelement the Pulse must go through every delayelement before. So in terms of source impedance the last coil is supplied with the summed ESR of all delayelements before.

There is definitely a loss of energy per stage, a sort of exponential decay. But it has to be seen in relation to the loss of energy due to energy imparted to the projectile.

This cant be benefical. Specially given the point, that the last coils must actually conduct more current, than the first coils... thats kind of backwards.

Why does the last coil have more current?

DerAlbi

Thu Feb 26 2015, 09:52PM

Registered Member #2906 Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 727

Coilgun Basics: you need to keep the Amp*Turn constant.
Later coils need to energize faster due to higher projectile velocity. Therefore less inductance is allowed which can only be achieved by decrase of turns. If you want to keep your pullforce=Amp*Turns constant, you have to compensate this with higher current.
Anyway, if you go through the design, you end up in a conventional multistage design.
Because your exponential decay ... really.. its not driven by the energytransfer to the projectile.. its driven by ESR. Coilguns struggle with bad Efficiency, remember?

Shrad

Fri Feb 27 2015, 07:32AM

Registered Member #3215 Joined: Sun Sept 19 2010, 08:42PM
Location:
Posts: 780

a water capacitor of a couple nanofarads can be done with two aluminum cylinders and nylon endcaps, as in the laser video I linked

if you use that capacitor and an inductor with the formed PFN designed so that the capacitor discharge time matches the line propagation, you have a discharge which is sustained during the whole travel of the projectile, if the projectile itself shorts the turns of the inductor you allow it to accelerate and keep flux strength high by discharging a decreasing energy in a decreasing value inductor

that's just an idea of course, which is far enough of classical designs so I can believe people would criticize, but it is primarily meant for discussing pulse compression and high power PFN

Uspring

Fri Feb 27 2015, 10:46AM

Registered Member #3988 Joined: Thu Jul 07 2011, 03:25PM
Location:
Posts: 711

Later coils need to energize faster due to higher projectile velocity. Therefore less inductance is allowed which can only be achieved by decrase of turns.

As said, the TL should be non uniform in speed to accomodate for the acceleration, i.e. L*C should decrease to the end of the TL. Also the impedance of the line should be constant to avoid energy reflection. Impedance is sqrt(L/C). Together this implies a reduction of L and C by the same factor towards the end of the line. For a given pulse energy, this implies larger currents at the end and also lower energizing times. So a correct design automatically implies less inductance at the end.

If you want to keep your pullforce=Amp*Turns constant, you have to compensate this with higher current.

I don't think a constant pull force is required.

Anyway, if you go through the design, you end up in a conventional multistage design.

If you'd be able to recover the energy left in a stage after the projectile has gone through it, it would be the same.

Because your exponential decay ... really.. its not driven by the energytransfer to the projectile.. its driven by ESR. Coilguns struggle with bad Efficiency, remember?

Possibly just because you don't recover the energy left in the coil after the projectile has gone through it. Is the Q of the coil without a projectile that low?

DerAlbi

Fri Feb 27 2015, 01:49PM

Registered Member #2906 Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 727

Uspring you are missing an important point

Coilstages get more and more efficient with higher Projectile velocity. So basically the first stages suck. Even with a halfbridge the injector coils optimal switching points (Based on Force-IÂ²-Correlation and optimization to EffÂ²*TransferedPower) gives a switching scheme that is basically that of an SCR-circuit. Efficiency is simulated to be about 8%.
In your discrete Delay-Line the propagating pulse energy can only be the left energy after the pulse. So connecting your coils in series could never achieve a higher eff than that of the initial stage! Ind it doesnt matter if you use a seperate injector system. The priciple is allways the same.
Usually the efficiency increases with projectile energy. Your eff. decreses with every stage. I would call that a design flaw.

Shrad:
the propagation delay is dictated by dielectric constant

How do you plan to slow down the pulse?
Getting down from 1/60 of the speed of light to even Mach 20 is a difference of 3 orders of magnitude.
you would maximize sqrt(L*C). The problem is, that C maximizes when L shrinks. (and vice versa) There is a practical limit to this. nothing you could influence or design with such freedom.

BigBad

Fri Feb 27 2015, 08:19PM

Registered Member #2529 Joined: Thu Dec 10 2009, 02:43AM
Location:
Posts: 600

Uspring wrote ...

If you want to keep your pullforce=Amp*Turns constant, you have to compensate this with higher current.

I don't think a constant pull force is required.

It depends whether you want a constant acceleration, or constant power acceleration. If the latter, performance is worse.

Uspring

Fri Feb 27 2015, 08:57PM

Registered Member #3988 Joined: Thu Jul 07 2011, 03:25PM
Location:
Posts: 711

In your discrete Delay-Line the propagating pulse energy can only be the left energy after the pulse. So connecting your coils in series could never achieve a higher eff than that of the initial stage!

You lost me here. What is the "left energy after the pulse"?
Say the first stage will convert 5% of the electrical energy into kinetic energy and say another 5% are lost due to ESR, eddies etc. The next stage will then get 90% of the initial pulse energy. Assuming similar efficiency and losses, pulse energy will be down to 81% after the second stage and so on. After e.g. ten stages pulse energy will be at 35%, i.e. 65% will be gone and half of that is kinetic energy.

first previous 3 4 5 6 next last

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