Magnet charger design 2

[Boris A]

Hi everyone
My hobby is antique engines which frequently come with magnetos. These magnetos can be over 100 years old and sometimes need thier permanent magnets "recharged". I have designed and built a magnet charger that has a single pole rather then two poles that is the more common design for chargers (see attachment). It has 160 coils of 12 ga wire wound on a 1" x 4" low carbon steel rod. It draws around 100 amps from a 12V car battery for a brief moment to charge the magneto magnet.
I would really like to know if there is a formula that could estimate the charging capability of this configuration. I know that amp-turns is a basic measure of its performance but how does the core rod dimension and material affect its performance? Also how does the armature structure design come into play?
The formulas B=µNI/l and F=(NI)^2μ0A/(2g^2) (where g is the air gap). seem to imply that core cross section is important as well as length and material. Is there an easy way to tell at what point the core saturates? And how does the rest of the structure affect its performance?
Is there a basic formula that I could use?
Thanks much!
Robert

 

[MagMike]

The necessary equations aren't quite straightforward, you might have better luck borrowing a gaussmeter and measuring the field on the face of your electromagnet while pulsing.

Better yet, can you add a few more pieces of steel to your armature structure to make a U-shape? This will produce more of a closed-circuit configuration which will be much more efficient from a magnetizing standpoint. In addition, the magneto will magnetize more evenly if both poles are in contact with the charger.

 

[dgallup]

You can get a pretty good idea what's going on with the free 2D magnetic simulation software FEMM.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[MagMike]

I agree with dgallup, FEMM could easily model this.

That said, I ran a model in 3D FEA. Assuming a 1" x 1" x 4" long 1018 CRS bar wrapped along the 4" length, producing 16000 Ampere-Turns: 3400 Gauss (0.34 Tesla) on the face of the steel. Field strength drops to 1875 Gauss (0.1875 T) at 0.5" (12.7mm) away from pole face.

That probably won't be enough to properly re-magnetize your magneto.

Another option that I've used successfully on large Alnico 5 horseshoes (sometimes called 'Horns') is to use steel to make a closed magnetic circuit and then wrap as many turns as feasible about the magnet. You might need to pulse it 2 or 3 times, but it has a good chance of working.

 

[Boris A]

Thanks dallup, FEMM sounds like just what I need. Unfortunately I have a Mac, so it won't work for me unless I fire up my old Win 98 laptop.

So thanks MagMike for running a sim for me! And thats just the measurement I need.

When you say "face of the steel" are you referring to the ends of the core rod or the gap between the top armature blocks? How close to saturation (I think thats the right term) does the 1 x 4" rod get to (assuming no air gap)? Would a shorter or thicker core rod help much? How does the length of the flux path through the armature to the magnet being charged affect the strength delivered?

For my next charger I will change the armature blocks from cast iron to low carbon steel, and the core rod to a high purity "magnetic" iron. Would that help much? Or just pump in more amp-turns?

Thanks guys, you are assume!

 

[MagMike]

Boris, is the picture in your original post a picture of your charger? I modeled the core rod by itself, without including the armature. I think I misunderstood your setup and, if so, I apologize. I can re-run a model over the weekend.

That said, low carbon steel will help. In the FEA model, the 1018 CRS core was already at 2.1 Tesla, so it is at or close to saturation.

 

[Boris A]

Yes thats the charger. Those black blocks are cast iron 1.25 X 2.25 x 3" I tried to reduce the flux path length as much as possible to the magnet being charged. Is that beneficial?
Your finding that the core may be saturating is very helpful thanks!

 

[MagBen]

If you can make taped pole pieces toward the gap, it not only concentrates flux with less fringe loss, but more importantly increases your field strength B proportionally to the contact area.

 

[EdStainless]

A process related tip.
After you magnetize the piece put steel keepers across the poles before you remove it from the fixture.
When you then install it slide the keepers out of place as you slide the magnet back into the assembly.
This will minimize the self demagnetizing effects.
This process is important with older very low coercivity materials.


= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[MagMike]

Sorry for the delay in replying. I modeled up an approximation of the magnet charger shown in the initial post.

~1000 Gauss on the pole faces
~730 Gauss at the center, midway between the two pole faces

It would depend on the coercivity of the magneto material, but I suspect this isn't enough to adequately magnetize the magneto

 

[Boris A]

Thanks MagMike for the sim results. Looks like its not strong enough.
One final question: how does the length of the armature plus coil affect the strength of the flux?

 

[Compositepro]

I can no longer resist pointing out that magnetizing something requires a high magnetic field strength, but only for a very short time. This is usually done using a high voltage discharge from a large capacitor.
Also, calculating the magnetic field strength in the air gap is not very useful, as there is no air gap when the horseshoe magnet is in place.