Permanent magnet assisted motor coil designs

[gyulasun]

Hi captainpecan, 

Jack Hildenbrand wrote about 8 to 10 W input to his electromagnet coils for his valve to operate as he expected if I recall correctly.  Your input power is around 2.3 W or so (considering 13.6 V and 80 Ohm coil resistance). Of course the iron sleeve, the magnet size and strength etc he and you used can be different (Jack never wrote about his coils data) but from your test observations I also think the input power is too small to open up the permanent magnet's  field locked into the washers. And when you increase the supply voltage beyond 13.6 V which seems to be a must, you will need to reduce the test time to as small as possible to prevent overheating the coil. At say 40 V the 80 Ohm coil may dissipate 20 W, so the intermittent testing is preferred, forcing you towards the pulsed operation (as you mentioned).   

Note also that the washers have air gaps between them (however tiny they are) and this may make overall "core" saturation even more input power demanding than in case of a solid sleeve, besides the fact that cores with an open magnetic circuit from the coil's point of view involve a certain input power demand in itself too. 

If you have a Variac, I suggest to build a simple variable DC supply using a diode bridge across its output with say at least a 100 VDC rated electrolytic filter cap (47 uF - 100 uF would be enough). I know a Variac autotransformer is galvanically not isolated from the mains but using it with care for such tests it may be okay.  If you have a step-down mains transformer say to 15 or 24V, it may be used in reverse, driven from the Variac and rectify its (primary) output, this would already be safe and galvanically isolated from the mains.   Unless you find other solutions to increase DC voltage

In case your magnet (enclosed inside the washers) can be removed relatively easily, then a kind of comparison test could be made between the electromagnet alone and the permanent magnet also alone. Check the magnetic field strength of the permanent magnet from a few cm distance.  Then do this with the electromagnet (with no magnet inside) for the same distance to have field strength comparable to that of the permanent magnet while varying the input voltage to the coil. Of course you need to act fast to avoid overheating when input voltage is increased to some ten volts and higher. This test is not an accurate comparison of course, perhaps good for a rough estimation on coil input power. 

Gyula

[captainpecan]

@gyulasun,

I agree with pretty much everything you said. It's good to hear your thoughts on the process. I think it's also possible to reduce power to get to the release of the internal pm field by decreasing the thickness of the core needed to get saturated. In other words, as nice and easy as using a stack of washers is, using a thinner pipe option may actually be preferred. The washers are pretty thick in width and could be overkill and just require more power to reach the "release" point of the pm. Jimbo had mentioned the pipe idea a bit. I would like to wind a coil using pipe as a test also but as of now I am having a hard time finding a ferrous material pipe that my current magnets fit properly in. I'll keep it in mind amd keep looking for options as I work forward with what I have.
I tried removing the magnet after the coil was completed because doing those type of split tests was already on mind. But now I can't get the magnet out of it. I may need to use much more force to remove it now. Which is frustrating because it slid in and out as easy as possible before. So before I accidentally screw up the coil I have, I'm going to try and get the power needed figured out to use it the way it is.  But of course it's all a work in progress and I learn best from stuff that doesn't work as expected. But it really is an intriguing concept and pretty exciting. Hopefully everything keeps looking better and better.

[captainpecan]

A little bit of research at K&J magnetics, it shows that for my 12mm by 12mm (1/2" basically) magnets, it only takes 2mm and above of plate to saturate one side of the field. That makes me believe if I cut the thickness surrounding the magnet to between 2mm and 4mm, it may greatly decrease the amount of energy needed to release the pm field from the core. It would easily make the core at least 2 or more times thinner. I think I will try wrapping a magnet with some layers of 1mm sheet metal I can easily cut with tin snips. Or maybe i can find a 12mm ID bushing i can work with that is megnetic. May be worth a little try before I go much further.

[captainpecan]

I decided to test my theory a bit and make another coil. I found some spacers that work great for my use and they are 3mm wall thickness. Kind of expensive buying them this way, but I only needed 2 to test my theory. I could not find 1 that was the length I needed or I would have done it all in one piece. I needed it slightly longer than the magnet because i need to put 2 or 3mm of steel on both ends of the magnet, and all need to be enclosed inside the spacer and copper windings.
I also knew I was not going to have enough wire from this spool to get 2200 turns like the other one. But I did want to have some center taps at 500 turns, and 1000 turns so I cam see how the effect of the added turns is. Maybe it no longer helps at a certain point? Just gives me more control to test things.
The coil looks a bit crappy, and 1 of my taps wasn't lined up where I wanted it. But the center is easily removed and I cam replace the permanent magnet with solid core for testing as well. After the coil dries a bit more, I'll hopefully get to test it out tonight.
The spacer I used is less than half the wall thickness. If there is any significant difference, it should show.

[captainpecan]

The coil looks a little crappy. But I made the coil with 2 center taps
500 turns 11.6 ohms
1000 turns 25.5 ohms
The spool was almost empty so I just ran the last of it on the end so it was 600 more turns.
The total of the coil is.
1600 turns 45.4 ohms.


Hopefully I can have some test results soon.