Barrel material / Electromagnetic Projectile Accelerators / Forums

Forums

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

« Previous topic | Next topic »

Barrel material

1 2 next

Move Thread

LAN_403

Maj

Thu Jan 10 2013, 01:02PM

Registered Member #9128 Joined: Sat Dec 29 2012, 02:23PM
Location: Sydney, Australia
Posts: 7

The HvWiki states: "To avoid eddy currents you should use a non-metal barrel".

So how about carbon fibre? It's non-magnetic, but conductive. Does this behave like a metal barrel or is it OK?

BigBad

Thu Jan 10 2013, 04:44PM

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

Fiberglass is cheaper and non conductive

Barry

Thu Jan 10 2013, 06:33PM

Registered Member #90 Joined: Thu Feb 09 2006, 02:44PM
Location: Seattle, Washington
Posts: 301

Maj wrote ...

So how about carbon fibre? It's non-magnetic, but conductive. Does this behave like a metal barrel?

A carbon fiber barrel will be okay IF eddy currents do not consume significant energy.

Think of "coil over barrel" as being a transformer with a one-turn secondary winding. Here, a carbon fiber barrel is essentially a resistor in series with an ideal inductor. You'll have to compute your own equivalent resistance value from your particular material's properties and your barrel's diameter and thickness. From this and a little math, it is possible to figure out the I[sup]2[/sup]R power loss.

Also note that eddy currents will introduce a small delay in how fast the magnetic field will penetrate the barrel. This is why the barrel's thickness should be less than the skin effect depth. Read more about calculating skin effect depth here.

My stupid-assed wild guess is that carbon fiber will work great, especially if it allows you to use an extremely thin barrel. But it's just a guess; I haven't tried it nor done the math.

Personally, I prefer thin clear plastic barrels. I like being able to see inside the firing tube.

Cheers, Barry
I went outside once. The graphics were nice but game play was horrible.

Maj

Sat Jan 12 2013, 12:46PM

Registered Member #9128 Joined: Sat Dec 29 2012, 02:23PM
Location: Sydney, Australia
Posts: 7

Thanks Barry. Based on the resistivity and permeability figures I've been able to dig up for carbon, it seems that skin effect depth shouldn't be a problem.

However I2R losses seem to be about 10% of total energy if I've calculated correctly.

Not sure the barrel will stand up to me trying to slot it, so I'll live with the losses for now.

klugesmith

Fri Jan 18 2013, 01:50AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

Maj wrote ...
Thanks Barry. Based on the resistivity and permeability figures I've been able to dig up for carbon, it seems that skin effect depth shouldn't be a problem.
However I2R losses seem to be about 10% of total energy if I've calculated correctly.

Would you mind sharing the resistivity values you found for carbon? Were they for bulk graphite, or for fiber-reinforced composite structural material? I would bet (guess) that the circumferential conductivity and eddy current magnitude are less than 1% of what you would get with aluminum or brass.

Maybe I can find a scrap at home and measure it.

Maj

Fri Jan 18 2013, 10:50PM

Registered Member #9128 Joined: Sat Dec 29 2012, 02:23PM
Location: Sydney, Australia
Posts: 7

klugesmith wrote ...

Would you mind sharing the resistivity values you found for carbon? Were they for bulk graphite, or for fiber-reinforced composite structural material? I would bet (guess) that the circumferential conductivity and eddy current magnitude are less than 1% of what you would get with aluminum or brass.

Maybe I can find a scrap at home and measure it.

I saw a few, but I recall basing my calc on this on:

It shows graphite carbon to be between 3 and 60 x 10^-5 ohms per metre

I realise it's not an accurate value for carbon fibre, but gives me an approximation that I figure is close enough since I'm comparing with materials that have a different order of magnitude (ie Copper, Plastic)

BigBad

Sat Jan 19 2013, 12:51AM

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

Seriously, you can buy fiberglass reinforced polymer tube, it's a standard part, not very expensive and you get no losses.

Ash Small

Sat Jan 19 2013, 01:46AM

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

Carbon fibre is a lot more conductive than you'd think, even in an epoxy matrix.

I once supplied the stainless rigging bushes for a carbon fibre yacht, and even though the bushes were passivated in boiling nitric acid, galvanic corrosion (rust) appeared within a very short time, surprising both myself and the owner (an olympic silver medallist).

I found a link here

pertaining to 18/8 stainless steel and CFRP used for body implants.

I'm sorry I can't find any figures that are more relevant to the topic under discussion here. (I don't think galvanic corrosion, and de-lamination will be an issue here

)

Maj

Sat Jan 19 2013, 07:12AM

Registered Member #9128 Joined: Sat Dec 29 2012, 02:23PM
Location: Sydney, Australia
Posts: 7

BigBad wrote ...

Seriously, you can buy fiberglass reinforced polymer tube, it's a standard part, not very expensive and you get no losses.

I have a bunch of different sized carbon fibre tubes already so was trying to use them, but might see if I can order some of those online. I'm trying to keep the barrel wall thickness to a minimum but I'm sure fibreglass reinforced polymer will be quite strong/thin. Thanks BigBad.

klugesmith

Tue Jan 22 2013, 10:21AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

Maj wrote ...

I saw a few, but I recall basing my calc on this on:

Good numbers, except the unit is ohm-meters, not ohms per meter. I'd love to pursue this in more detail, since you have a good assortment of carbon tubes. (for remote-controlled flying things?)

I found several references

about eddy current testing of carbon composite materials (a nondestructive way to find defects and damage).
An oft-quoted range for conductivity is 5000 to 50000 S/m parallel to the fibers, and 10 to 100 S/m transverse to the fibers. Your hyperphysics link values translate to 1700 to 33000 S/m.

Before that, I did some measuring. Had scrounged a scrap of carbon fiber bicycle tube, 1.125" OD, 0.08" wall, 0.93" long.

Was unable to get a stable resistance measurement and potential gradient in the circumferential direction. Sanded two longitudinal strips, on opposite sides, down to black stuff. Squeezed aluminum-foil strips against them, with resilient backing, all clamped in a vise. It seemed like I was dealing with an unexpectedly great anisotropy.

Had no trouble making an axial resistance measurement. Used the same vise jaws and backing material to squeeze aluminum-foil rectangles against the sanded-flat ends of the tube specimen. Measured 63 mV drop with 1.00 amps flowing. That works out to 2200 S/m (4.5e-4 ohm-m), not inconsistent with the parallel-to-fibers ranges given above. Copper is 27,000 times more conductive.

I bet the resistivity is much higher in the circumferential direction. Easily explained if the CF woven mat came as a prepreg sheet, wound around a mandrel with poor electrical contact between the fibers in consecutive layers. Not easy to measure the resistivity. Could saw the tube in half lengthwise, sand flat, and press against aluminum foil electrodes -- but that would short all layers together, for apparent conductivity that could be much too high.

Any comments on measurement by eddy current, perhaps done as follows? Make a thin cardboard sleeve that slides over CF tube. Wind a solenoid of magnet wire (20 AWG?) on that sleeve. Measure its L and R with and without the CF tube inside, then figure the CF shorted turn resistance assuming, say, 90% coupling. Same result could be obtained by connecting a charged capacitor to the coil and measuring the LCR discharge pulse waveforms.

1 2 next

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