Hilden-Brand Magnet Motor

[gyulasun]

Hi Rob,

Thanks for showing the simulations, very interesting setup that you embedded the permanent magnet into the body of a 7" OD toroid core. I am not really familiar with the magnetic properties of the M47 steel material you used for the ring core: how high frequency (some hundred Hz, or a few kHz?) can be used for switching it with neglectable eddy current losses?

Wrt to the output switching I may not have been clear enough so I have a go at it again:

The rectifier and the energy storing capacitor is ALWAYS connected to the output coil permanently,  the switch will connect the useful load to the capacitor (and not directly to the output coil), right?  This way "only" the  cap charging current loads the output coil permanently and by not discharging fully the cap later with the real load, the counter effect of the cap charging current on the main flux could be kept at the minimum, that is what I think.  Reading again my previous letter now may make more sense. 

Regards
Gyula

[MeggerMan]

I am still not sure where you are going with your idea, if you charge up the output capacitor via a diode or FWBR then at some point you are going to use this stored energy to light a bulb for example, and therefore there will a load on the output coil.
I think what you may be saying is that we need to limit the amount of current generated in the output coil.
What may happen is it will reach a state of equillibrian where the switch part switches.

Only testing with some laminate steel and a Neo magnet will prove wether this can work.
Simulations are OK to a point

M-47 Steel:
Electrical conductivity= 4.3 MS/m
Thickness = 0.635 mm
Fill factor = 0.98
Saturates at what appears to be 1.4 Tesla

If you get a copy of Femm 4 it tells you in there.

Bit of info here too:
http://www.mapessprowl.com/MAPESWEB.NSF/PublishedTopic/Electrical%20Specs?OpenDocument

Looks like I should use a lower number like M-3

Regards

Rob

[gyulasun]

Hi Rob,

Ok, lets assume you operate the magnetic valve with 50Hz (50% duty) square wave, right?
It will mean the valve is on for 10 ms only out of the full 20 ms cycle, right?
(I could have said a sine wave too but for practically switching an electromagnet on or off a square wave driven switching device is better.)

So the output cap should charge up in 10 ms to a certain voltage, then you have got 10 ms time to discharge this cap through the switch to a load.  So you have a max of 10 ms switch-on time for the switch at the output and during this 10 ms periode the valve is NOT switched on, right? Of course the input and output switching times must be syncronised and probably the input duty cyle should be made variable to find the best operating point possible.  Also the output capacitor's value needs experimenting too.  Also it is very probable that the output capacitor charges up to a few hundred volts when not loaded this would involve choosing an appropiate load etc so these all are factors to be considered.

Thanks for the data on the core material.

Regards
Gyula

[MeggerMan]

Here's another core based on the Magnetic Power Inc design, this allows switching both ways so may be even better than Jacks design.
There is no ouput coil shown, but you can get the idea.




The femm file is in the yahoo group that I think you are already a member of.
http://groups.yahoo.com/group/pes_flynn_pp/files/Robs Files/Example by Rob Generator rev 2.FEM

Regards

Rob

[Light]

Here's another step in studying of Jacks valve.
Just played in different variations and got this speed (see attach).
Anyway the valve works; the matter of OU is a matter of time and practice...