Hilden-Brand Magnet Motor

[MeggerMan]

I am just working out the fine details of this core using 4 stacked magnets (N38 50mm height x 25mm x 25mm) and 0.9mm (20SWG) copper wire for the 500 turns on each of the 3 main limbs.
Using much thicker wire allows the volts drop to go down and power consumption to also go down dramatically.
Using more magnets increases the flux slightly whilst still using the same power to switch the field, but more important gives space for the switch windings themselves.
The total weight of copper for the windings is about 3.5 to 4 kG or 600m of 0.9mm diameter wire.
I will also include 2 small slots for the tiny linear hall effect probes, this way I can monitor when and how well the switch is working.
The output windings can then be added once the switch is proved to work.

Regards

Rob

[MeggerMan]

Sorry the file is in rar format, but you should be able to un-rar it.
This has the following fixes:
1. Winding wire is much thicker (0.9mm).
2. Power consumption is about 0.04 watts (1/10 of the previous).
3. Depth of core is now correct - 50mm.
4. Double height of magnets (12 magnets).
5. Better overall core shape, squarer windows.

Just needs the probe windows adding.

+50mA



-25mA



Regards

Rob

[hartiberlin]

Hi Rob,
many thanks for posting your new FEM circuit.
I did not know that one could define a "circuit property" also and
apply to a magnet wire ( copper wire) a current in this program.

That is a very neat feature!

Are you sure the setting "series"  is right ?
As in this coil space the wires  are all in parallel  if you look from the sideview, aren?t they ?
Or how is this defined in the program ?

What is the complex operator J x has to do with it ?
Can you also apply phase relationsships there ?
So can you also simulate AC timings with phase shiftings this way ?

Is it also possible to see with FEMM how much voltage an output coil would get ?
Probably not, as this would need a dnamic ssimulation as the output voltage
is just proportional to the d(phi)/dt inside the core piece of the output coil, right ?

[hartiberlin]

Hi Rob another question is,
if you will get so much wire into such a small coil space ?
Maybe simulate different coil spaces ( different big areas for them)
and have a look, if the simulation
will change..

[dutchy1966]

Sorry the file is in rar format, but you should be able to un-rar it.
This has the following fixes:
1. Winding wire is much thicker (0.9mm).
2. Power consumption is about 0.04 watts (1/10 of the previous).

Hi Rob,

Good job putting all the effort in to create a nice simulation of the hildenbrand valve.
Just a little thought about the power consumption you mentioned. The power consumption is in my opinion only valid when there is no output current (coils). As soon as power is generated in whatever output coil there will be created a counter flux. This will make the valve to switch back as this counter flux is much stronger than the flux of the switching valve. Because of this, I think, your switching power should be much higher to prevent this from happening. (Or is here a chance that the whole thing starts to resonate....hmmm)
What do you think of this?

greetz,

Dutchy