# Electromagnet help

Discussion in 'The Projects Forum' started by shortbus, Jun 6, 2012.

1. ### shortbus Thread Starter AAC Fanatic!

Sep 30, 2009
4,096
1,692
OK, can't seem to find answers on Googling so will ask here.

When building an electromagnet does the way the coil is wound on the core/armature make a difference to the magnetic lifting force from the magnet? I have a drawing attached to show what I mean.

Some assumptions for all the illustrations-
1. The cores in them all are laminated and the same size.
2. The windings are all the same number of turns, same gage of wire and the same volts and amperage on each.
3. The voltage is DC.

The drawings are-
A. The coil wrapped around the center leg of core.
B. The coil split into two parts with one part wound clock-wise the other counter clock-wise.
C. The coil in two parts but wound in a single direction.

Which one would make the most magnetic force? Or doesn't it matter how its wound?

File size:
77.5 KB
Views:
44
2. ### amilton542 Active Member

Nov 13, 2010
494
64
If you follow the path of conventional current through B and C, then each set-up will have identical magnetic polarities. If you wanted B wound how I think you do, then the two coils will weaken each other.

Assuming leakage flux and tightly wound coils, winding arrangment A will produce the strongest flux density in the air gap.

3. ### praondevou AAC Fanatic!

Jul 9, 2011
2,936
489
It's not clear how B is wound.
In C, I assume the two windings to be wound in the same direction (one is the continuation of the other).

I also assume that when you say all windings have the same number of turns you mean the total number of turns in each picture is equal.

My guess (!):

The winding in A generates a certain flux, it is concentrated in the middle leg and divided (equally) between the two outer legs. Ideally, that means the flux in the outer legs is half of that in the middle leg.

Like here:

How the flux distribution looks like in C is hard to say. Each winding generates now only half of the flux , the way of least resistance would be through the middle leg, then the two fluxes in the middle leg would cancel each other out... What does that mean for the attracting force on a piece of steel attached only to the middle leg? I would say there is no force. But since the total flux has to be the same, I would say that the two fluxes add and give the total flux through the outer legs... At least this what would happen if you do a superposition with an electric circuit, replacing the windings with a voltage source and the air gaps with resistors.

4. ### amilton542 Active Member

Nov 13, 2010
494
64
Praondevou beat me to it. I sat down and began to drink a can of lager, then realised the flux divides in A!

This time (assuming a lossless core with infinite permeability) and thinking outside the box. If I depict the bottom length of the core as a weight- lifting barbell, then I will do more work with my arms positioned in the centre; in other words it will be harder for me to lift and bring to my chest.

On the other hand, if I attempt to lift the bar-bell with my hands gripped either side; it will be easier for me to lift. For this reason I'll go with C, too.

Build each set-up, conduct the tests then write the report.

5. ### wayneh Expert

Sep 9, 2010
12,396
3,246
There are some factors not yet addressed, such as whether the coils will be layered in A but not in C, whether saturation of the core is approached in the middle leg of A but not C, and maybe a few other things.

That said, I don't see why A wouldn't be, on paper, the same as C. Same amp turns, same surface area for flux flow, and so on. In the real world I think C would lift more reliably, because the failure will always be when the bar falls away from one side or the other. I think C puts more flux to the outer legs where it can do more good. Hmmm... Then again, all the field lines still must return through A. My brain hurts. Forget this paragraph.

Winding direction does matter, and that's what I assume you are asking with B.

Last edited: Jun 7, 2012
6. ### praondevou AAC Fanatic!

Jul 9, 2011
2,936
489
In B, assuming the windings are made in a way to create opposite flux, there should be half of the flux in each outer leg and in the middle leg they would add since they are in the same direction.... I guess.

7. ### shortbus Thread Starter AAC Fanatic!

Sep 30, 2009
4,096
1,692
Thanks guys for the answers so far.

In A, I'm sure about the flux being through the center and split equally in the outside legs. Like in a transformer, that was the only diagram I could find, showing the flux in a core.

In B, with the windings going in opposite directions. My thinking was that it would put say, a south pole in the center leg and two north poles on the outer legs. When the "keeper" or metal was across all three legs I thought 'maybe' it would make like two separate but equal magnets from it.

In C, with the coils wound the same direction. I thought maybe the center leg would not have much strength but don't know.

A crane lifting magnet is made like the one Praondevou showed. Much like drawing A is.

8. ### crutschow Expert

Mar 14, 2008
13,509
3,385
You don't need a laminated core for DC.

9. ### praondevou AAC Fanatic!

Jul 9, 2011
2,936
489

This is a quick drawing which is the cause for the irregularities. It shows only how the flux is supposed to be divided, not absolute values. However , the simulator would be capable of calculating it... attracting forces too.... if someone wants to put the time and effort into it.

If this is true, there would be no attracting force on the middle leg in C... if this is true in reality, no idea.

EDIT: I replaced the picture, B and C were inverted.

• ###### e-core flux.jpg
File size:
191.6 KB
Views:
56
Last edited: Jun 7, 2012
10. ### shortbus Thread Starter AAC Fanatic!

Sep 30, 2009
4,096
1,692
@Praondevou - WOW that is an amazing program you have!!! Thank you very much for running it! It shows everything I wanted to know. I guess my guesses weren't too far off.

Is the program called FEMM ? Or something else.

This is all leading up to my building a SRM (switched reluctance motor). Already bought the E-laminations and now working out the electrical winding part. Then the frame/housing, but had to know the best way to wind it to go farther. Thanks again.

11. ### shortbus Thread Starter AAC Fanatic!

Sep 30, 2009
4,096
1,692
They will eventually end up in an electric motor, so thats why I'm using laminations.

12. ### praondevou AAC Fanatic!

Jul 9, 2011
2,936
489

It's freeware (and 2D). For a 3D simulator you probably will have to pay.

There are some tutorials outthere that show how to do these simulations. I did it with permanent magnets, but it can also be done with coils. It essentially treats coils like permanent magnets, it is mainly for low frequency and DC analysis.

strantor and shortbus like this.
13. ### shortbus Thread Starter AAC Fanatic!

Sep 30, 2009
4,096
1,692
Thank you again for the help!

Now another question about electromagnets. I've used the wire 'ampacity tables' for wire. But in a transformer, motor or electromagnet, they seem to defy the ampacity of wires? How does this work in a coil? Have been able to find an answer to this, or a ampacity chart for coils.

In a thread on Electro-tech-online ( http://www.electro-tech-online.com/...ent-voltage-general-confusion-all-around.html ) they were saying, if I understand correctly, that maximum magnetic force is made when the coil has the same resistance as the internal resistance of the supply. How then do DC motors work that are high amperage? Like a car starter motor. Isn't the formula for amps- I = E/R ? If a battery has say 2 Ω internal resistance at 12V doesn't that make a 6A motor the most power on that battery? Doesn't line up in real life, or am I not thinking right.

14. ### wayneh Expert

Sep 9, 2010
12,396
3,246
I haven't read that link, but it's true that you'll get maximum power in your magnet coil when the impedance (not just DC resistance) of the coil matches the impedance of the power supply. It's just like choosing the right gear on a bicycle. The power supply - the rider - is happiest when the load is matched.

But that's optimizing just one parameter. There are other considerations that might dominate, and perhaps that answers the motor question.

15. ### wayneh Expert

Sep 9, 2010
12,396
3,246
What do you mean? Never heard of ampacity.

Sep 30, 2009
4,096
1,692
17. ### strantor AAC Fanatic!

Oct 3, 2010
4,302
1,989
i believe those ampacity tables are from NEC code and have several safety factors included, with derating. That's what size wires you can run through a wall without losing your electrician's license. As far as "how much current are you allowed to pass through a given guage of wire in a wall/conduit/etc per NEC" versus "how much current can you pass through a given guage of wire before it becomes a fuse" there is a big difference. Devices like motors and transformers (i believe) are not goverened by NEC - they're goverened by the court of public opinion. So if through R&D they find that they can get away with winding their motor with 20AWG wire and pass 50A (vs the 9A in the ampacity table) through it reliably without burning up and it makes their customers happy then they can do it.

18. ### amilton542 Active Member

Nov 13, 2010
494
64
In such a case, it's a matter of heat-dissipation. A length of chosen wire guage will generally dissipate heat into open-air with no 'sweat' at all. Wrap that wire gauge into a tight coil of a thousand turns or more, however, it will be close to taking your skin off if you press it against your face (it's a coil in a sleeping-bag).

Typically, a power transformer will make use of oil and forced air cooling mitigation techniques in order to keep the heat dissipation threshold amid tolerable limits.

In a similar way, the cooler you keep a motor - the harder it can work; simple.

19. ### shortbus Thread Starter AAC Fanatic!

Sep 30, 2009
4,096
1,692
But where do you find that information?