Low speed coilgun efficiency (why it's so bad at high accelerations) / Electromagnetic Projectile Accelerators / Forums

Forums

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

« Previous topic | Next topic »

Low speed coilgun efficiency (why it's so bad at high accelerations)

1 2 3 next

Move Thread

LAN_403

BigBad

Sat May 21 2011, 11:08PM

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

OK, I think I've worked out why CG efficiency is very typically horrible, and I possibly know how to avoid it.

So, I'm talking about CGs with iron projectiles, where the material is running saturated.

The problem is simply the saturation.

Let me explain.

In a coil gun at low speeds the projectile is being pulled along by magnetic pressure. If you look up magnetic pressure, you'll probably find that it goes as a square law on the magnetic field strength. Well... but actually that's not entirely true. What it goes as is BH, where the B is applied field and H is the reaction of the object (or its own field if it's a permanent or electromagnet). For a superconductor, it can create a complete equal and opposite, proportional H field, so the pressure goes as B^2. Air has no H so no pressure. Metal gives you a pressure only when it has its own field. etc. etc.

So at low accelerations in a coil gun, i.e. low field, you're on the linear part of the magnetisation curve of iron, and so the magnetic pressure goes as B^2, because the induced H in the iron is proportional to B. That's good because B field is proportional to the current in the coils, and resistive losses go as I^2, so everything is happy and proportional- you can be very efficient at low accelerations- it's just you run out of bore length. If it was a rotary motor you could be a perfectly efficient 90%, but it isn't rotary. But it WOULD be efficient if you had a very long bore.

Ok, high accelerations, high field. Big problems. The iron projectile saturates, so magnetic pressure now goes as B, H is constant. BIG problems i^2 losses go up much faster than I and B. So the faster you want to go, for any coil gun length, the faster you accelerate, the worse the efficiency gets...

(Aside: Note that at VERY high speeds you might be able to get it to work, because, energy = force x distance, thus useful power= force x speed, so for any given force, at some speed, the ineffiiciency of the iron saturation gets overwhelmed by the fact that the coils are only on for a short period, it's just that the coil gun people here build are nowhere near that fast).

So this means that you're completely screwed?

Nope.

The trick is NOT to use iron projectiles at high accelerations. If you you use conductive projectiles (copper, aluminium) and run your gun as an induction motor, designed carefully, you should/may be able to get 90% (or at least a lot better than ~3%). The thing is conductive projectiles don't saturate (if you keep iron/ferrite completely out of the circuit), so the force will go as B^2.

Anyway, that's my theory. It's plausible, but doubtless incomplete and it may be entirely wrong.

Enjoy!

Ash Small

Sun May 22 2011, 12:28AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

It certainly sounds plausible, once you reach saturation efficiency will drop off significantly.

One point though, once you reach saturation, isn't efficiency around 3%? (ie the figure quoted for these devices)

I don't know enough about non-ferrous projectiles to be able to comment. I'm still a relative newbie here.

BigBad

Sun May 22 2011, 03:00AM

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

I just simulated an axial induction motor using FEMM 4.2, and the acceleration scaled as I expected, I doubled the current and the force quadrupled, and I was getting decent acceleration. I think it seems quite promising.

BenB

Sun May 22 2011, 05:56PM

Registered Member #3785 Joined: Sat Mar 26 2011, 03:37PM
Location: Surrey, UK
Posts: 16

Attempting to avoid sounding like that went entirely over my head, but when using a reluctance design, would your theory imply that using a projectile formed of a material with a higher b/h saturation curve would improve efficiency? I'm in the process of choosing a final projectile design at the minute and I'm quite interested in casting iron filings or powdered ferrite into epoxy or a similar medium.

Ash Small

Sun May 22 2011, 07:16PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

BenB wrote ...

Attempting to avoid sounding like that went entirely over my head, but when using a reluctance design, would your theory imply that using a projectile formed of a material with a higher b/h saturation curve would improve efficiency? I'm in the process of choosing a final projectile design at the minute and I'm quite interested in casting iron filings or powdered ferrite into epoxy or a similar medium.

That's pretty much what BigBad is suggesting. Using a projectile that doesn't, or only just, saturates.

Forty

Sun May 22 2011, 09:12PM

Registered Member #3888 Joined: Sun May 15 2011, 09:50PM
Location: Erie, PA
Posts: 649

electronics goldmine has some cheap solenoids (1.99 u.s.d. each) that come with an already pointed, presumably silicon steel, plunger in them. also the original solenoid bobbin makes a great coil form and, if your coils O.D. isnt too large, the metal housing makes perfect fitting external iron to cover 4 sides.
Surplus sales of nebraska has just the solenoid plungers for sale too if that' all you need.

edit: i'd shy away from trying to use ferrite or powdered iron. just think about how easily those materials break apart when disassembling transformers, you'll have shattered bits of your hard work all over the place if you shoot anything besides a pillow.

BigBad

Mon May 23 2011, 01:00AM

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

Yes, the highest saturation you can get.

If you do want to go with epoxy, adding a small amount of glass fibres in the mix would probably help a LOT with holding it all together; it massively reduces the brittleness you often get with polymers (a single layer around the outside would do unbelievable things, otherwise try mixing some in randomly). Other fibres like paper also help to a degree, but they're not a patch on glass fibre.

But the problem is that these composites all reduce the magnetic properties and add weight.

Marko

Mon May 23 2011, 02:20AM

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

Hi BigBad

I think the most important concepts regarding increasing coilgun efficiency have been proposed before - most important of which is using iron in full magnetic path, and secondly using relatively large, slow moving projectile. I'll never have time or resources to invest into something like that, but the general concept is to design the gun using the theory of linear motors, which can be either reluctance type (synchronous or SLM's) or induction(LIM's). Both are capable of very high accelerations and efficiencies and I think that with some tweaking a pretty good coilgun could be derived.

Marko

BigBad

Mon May 23 2011, 03:03AM

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

I think the theory basically says that at low speeds that reluctance-type motors cannot achieve, simultaneously, high acceleration and high efficiency, although you can get one or other fairly easily; but induction motors ought to be able to do both.

I'm pretty sure that adding iron or magnetic core in the path of a reluctance type (moving iron/core) coil gun can't get you higher efficiency at high tesla; once the core has saturated, the relative permeability falls to 1.0; increasing the field beyond that, gives you no advantage from the core, and may even make it worse.

Shrad

Mon May 23 2011, 09:49AM

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

have someone ever tried to use those cheap ferrite antenna bars as a projectile?

1 2 3 next

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