Maximizing electromagnets for levitators / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Maximizing electromagnets for levitators

Move Thread

LAN_403

uzzors2k

Thu Mar 12 2009, 07:35PM

Registered Member #95 Joined: Thu Feb 09 2006, 04:57PM
Location: Norway
Posts: 1308

I took apart some floppy and CD disks recently and found they have several hall-effect sensors in the spindle motor, and, having some spare time I thought making a hall-effect levitator would be neat. It wasn't until half-way through winding the electromagnet that it dawned on me I never put any thought into whether fat coils of skinny coils are best. I've read winding on a short, fat core gives the most strength. Is this true for a practical coil?

Example:
For a short and fat coil each successive layer will increase the length of wire required to make a turn, meaning more and more resistance/turn as the coil is widened. The first, say 5 layers will use a reasonable amout of wire per turn, but once you start on layer 20 it'll take maybe 4-5 times as much wire for the same turn. At the same time a fat coil will have less area to dissipate heat from. On the bright side the magnetic circuit is short.

For a long and skinny coil it takes less wire to get the same number of turns, and the area is greater so it can dissiapate more power. The only drawback I see is magnetic circuit being longer.

So how should I wind an electromagnet to get the best ratio of force and power dissipation? The picture below is what I've made as of now. I'm considering moving the winding over to the bolt seen beside it, but I'm not sure if it's worth it. (which is what this post boils down to.)

Sulaiman

Thu Mar 12 2009, 08:53PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

More copper = better
more iron = better

Force = Area x (flux density)squared

For maximum efficiency a very short airgap is required,
that is, a complete magnetic circuit is required,
e.g. Horseshoe, pot etc.

StevenCaton

Fri Mar 13 2009, 06:58PM

Registered Member #1845 Joined: Fri Dec 05 2008, 05:38AM
Location: California
Posts: 211

Hi Uzzors.

That's funny, I was wondering the EXACT same thing as you.

I am currently designing my own circuit that will turn an electromagnet on and off.

Anways, i have two bolts just like you. One is 3/8'' thick and long, and the other is 1/2'' (short and fat)I'm not sure which one to wind for my electromagnet though. My circuit is going to place 9 volts across the electromagnet and cause 1 amp to flow. (I'll cut the wires to a 9 Ohm lengths)

I'll let you know which electromagnet ends up having more force, (I'll try both) but as Sulaiman pointed out that "Force = Area x (flux density)squared" I'll assume the 1/2 inch bolt will have more kick than the 3/8 inch bolt.

Both coils will have 9 Ohms and pretty much the same amount of turns so I'lll see what difference the bolt size makes.

aonomus

Fri Mar 13 2009, 07:27PM

Registered Member #1497 Joined: Thu May 22 2008, 05:24AM
Location: Toronto, Ontario, Canada
Posts: 801

One thing I've noticed is that most bolts from the hardware store are stainless steel of varying grades. For some reason some stainless steels are still mildly ferromagnetic, while other stainless steels are nonmagnetic.

The slight vanadium and chromium content does it, but its still something to consider.... if you can find a iron rod or bolt you would be better off to use that, and perhaps electroplate it with another metal to prevent corrosion.

Coronafix

Fri Mar 13 2009, 08:30PM

Registered Member #160 Joined: Mon Feb 13 2006, 02:07AM
Location: Melbourne, Australia
Posts: 938

Check out Tesla s patent for a better electromagnet.

Marko

Fri Mar 13 2009, 08:30PM

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

For a magnetic core, I wouldn't use any steel that isn't made specially for the purpose. I would try to use silicon steel transformer lamination (even if drive is DC) simply to assure maximum permeability and saturation flux density.

Marko

Sulaiman

Fri Mar 13 2009, 11:07PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

That's an excellent suggestion,
If you take apart a transformer core and re-assemble it as all "E" laminations together and all "I" laminations together, the lifting force will be huge.
Wind the coil on the center leg of the " E " and attract the " I "

The 'standard' device for this kind of application is an 'electromechanical solenoid' - worth googling.

OR if you get two steel bolts and screw them into a steel plate to form a ' U ' shape you can have a coil on each and the magnetic 'pull' will be many times stronger than just one bolt.

StevenCaton

Sat Mar 14 2009, 05:31PM

Registered Member #1845 Joined: Fri Dec 05 2008, 05:38AM
Location: California
Posts: 211

I tried two different bolts yesterday to see which one gave the most strength. Each coil carried one amp of current.
I would say the 1/2 inch bolt worked SLIGHTLY better than the 3/8 bolt.

So how should I wind an electromagnet to get the best ratio of force and power dissipation? The picture below is what I've made as of now. I'm considering moving the winding over to the bolt seen beside it, but I'm not sure if it's worth it. (which is what this post boils down to.)

I read somewhere (I think Barry's coilgun site) that a short fat coil towards the end of a bolt is superior to a long skinny coil along the length of a bolt.

uzzors2k

Sat Mar 14 2009, 06:09PM

Registered Member #95 Joined: Thu Feb 09 2006, 04:57PM
Location: Norway
Posts: 1308

Thanks for the tips peeps. I've learned enough of how to use Femm 4.2 to model some coil shapes with an object nearby to see what works. It appears short and fat coils are indeed better as they give a larger area for the field to emit from, consistent with everyone's observations. Modeling E-cores didn't look promising however. I have a MOT electromagnet which is an E core, but the field doesn't extend as far as with a I core. The force on an object which closes the magnetic circuit (placed so it closes the E, like an I-section would) is impressive however, which the simulations showed as well. So for levitators I think I cores are better suited. I'm going make some square cores with MOT laminations and see how well they work in the real world.

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