Here is an example of an advantage. This is a shot of the capacitor C2 voltage and the boost converter PWM bursts at 5 kHz 20% duty to keep the input voltage to 17 volts or so. We can see that the voltage of the capacitor C2 is now about 24 volts and now does not reach zero volts. Because there is not enough time due to reducing the pulse width.

Boost converter is voltage controlled by the 14M2 picaxe chip and the duty changed by a voltage divider on an analogue digital converter input to it. I'm writing new code but it's tough going when I have to teach myself new things while doing other stuff.

Cheers

At no time did Skycollection  measure any output voltage in his last video.

Mh, sorry I I wanted to talk about this video when skycollection speak with Flux4Energizer:
http://www.youtube.com/watch?v=a28DqHF5tGM&list=UUWqI0GXGprm81g3gkVa5v_w&index=2
Skycollion said he measured out the voltage of the output.

Here's a circuit for a single battery pulse motor folks.

How it works is the spike charges the capacitor C3 briefly because of the
inductance/impedance of the coil L1, C3 has the least impedance I think ( capacitors in parallel will help lower the cap impedance),
then after C3 is charged the inductor L1 discharges C3 into C2 to be reused.
In my setup the inductive spike charges C3 to about 25 volts for about 400 microseconds
before C3 discharges through L1.

D4 is a blocking diode so that the rotor magnets don't try to charge C2.

Doing this has greatly improved my motor I can run it from 300 mA at 1900 rpm
right up to 1.7 amps Amps for 3500 rpm, my mosfets don't have heat sinks
and stay cool at 300 mA input they feel cold. Performance is a fair bit
better than when there is a charge battery. But the charge battery can still be
used with this return circuit just by putting the battery in series with the flyback diode. 
The inductive energy return still catches the 50 volt spike and makes it 400 uS wide then put's it in C2.
Works a treat and very simple.

And some wave forms.

Cheers

Yes Milehigh, That is how it works.    I was beginning to think no one got it, but you nailed it pretty much.  Thanks
I could see it working in my head no problem. The uses of "resonance" in time delays is actually quite simple.
Well what I call resonant effects anyway, that is coils take time to charge and so do capacitors, they shift phase ect.

The last drawing with the inductive return is a thing many people have been trying to do, the answer was simple,
usually the answer is a capacitor, a coil a diode or all three.  With all the complicated circuits I've seen to do it I must say
I don't see why it needed to be difficult. Doing the hard yards to understand resonance in Tesla coils has taught me much.

The motor is working very well now and I've learned to use it fairly well so as to keep the phases correct for the rotational speed.
I'm beginning to be able to hear when the motor is "singing" and running efficiently. The torque is impressive for what was originally
a single coil motor with a charging circuit now it is much more efficient, faster and has a lot more torque, with another set of coils on
the opposite side of the rotor it would be faster and more powerful because the "on" times can be shorter if desired due to the
greater forces exerted on the rotor, the two charging coils could swing on an see saw arrangement to adjust the angle of both
charging coils in unison in a "four" coil setup. And it could be done by mechanical or manual or controlled by a servo and the picaxe.
Same with the pulse timing that can also be varied by a servo and picaxe to move the photo reflector or by software in the picaxe
I don't like the software approach to the pulse timing because of the code required and using hardware pulse processing it's much quicker.
For low frequency motor coils that's ok but my micro's are only the beginner ones and I have to teach myself how to code from scratch by
using the picaxe manual and intuition.

Here is sample of some some code I wrote for a 10 power level boost converter the battery alone and 9 levels of boost, the levels can be voltage controlled levels or current control levels for input power control. if the boost is controlled by the voltage level of C2 then the picaxe can work to keep the motor at a constant speed under load so that the pulse width and timing need not be changed when a load is added. The voltage is sensed from a voltage divider and the PWM switched off to limit voltage, I'm working on the automatic down or up staging to change levels based on the sensed voltage.

I's all open source the code the circuit and the idea. For those wanting others not to use their idea's or the idea's of others without giving credit, they now must give credit to me if they use my idea's. the others need not  .  As far as I am aware no one is doing just what I am doing here. The "two phases from one motor" is my design and the "inductive energy return" method I just showed is my idea as well. 

Innovation won't come from copying other people's work and thinking.


If we all ignore the fake OU claiming "Guru's" with agenda's and work together, we could advance in leaps and bounds. But I am always trying to explain that I do not care what Thane says or Utkin or Don Smith, what they say is obviously not true if we investigate properly.  The argument I get at times is madness, people talk at me in a condescending way like I know nothing.
We are all capable of the same things if we use our own hard work to do the experiments it will pay off in the end.

When I get a generator coil set up it should be a piece of cake to get an accelerating rotor under load using resonance principals
and load switching.

Thanks for the compliment MH, appreciated. 

Cheers

P.S. Milehigh, With this arrangement I think I can actually show that when the rotor is loaded or accelerating  under certain
conditions the inductive energy voltage spike completely disappears because the energy is all given to the rotor. I see it on the scope. The statement made by some that the mechanical energy is free and all the energy put into the coil can be recovered
without loss due to driving the rotor is a flat out lie in my opinion. the people making these statements are either misinformed
deceptive, or just lying to forward their own agenda's, likely hype and money. Time for them to fess up, prove it or risk their reputations. Well that's my opinion anyway.

As far as all the diodes go with respect to power dissipation, the 1N5822 Shottkey diodes are great, three 1N5822 diodes should dissipate less power than just one FR307 fast recovery diode, the flyback diode sees too much voltage to use a 1N5822 there I think they are rated at 40 volts from memory, I could use a better diode there though it only needs to handle 1 Ampere or so.

..



 

Back to coils, I have a question. What happens if we take a pancake coil bifilar or not and place it in close proximity to the rotor magnets, so that the spiral windings are in the same "plane" as the rotor (circle to circle) and the spiral coil is a band of windings right next to the side of the diametrically magnetized tube magnets, so that the magnets rotate right next to the spiral coil windings continuously ? Nothing ? What if I shield part of the coil  from the magnetism ? So that the coil gets pulses of magnet flux in the windings ? I might try that.  That will be my idea as well, I'll call it "PLANAR FLUX GATING". If it works that is. hehehe,  I'll make a drawing.  Maybe it has already been done.

Cheers

Hi Farmhamd,

Get a large ring magnet, angle it about 24-25 degrees but make the shaft axially symmetric. When it spins it should look like a wobbling plate with forced precession. Place your coils so they align to the same plane of the shaft but through the center of the magnet. As the magnet wobbles it's plane will cut the coil generating power. I've tested this before it works, not OU but interesting.

Prof. Eugene Butikov has a really informative site on precession physics.

http://faculty.ifmo.ru/butikov/index.html

This is the Java applet shown in the picture and setting.
http://faculty.ifmo.ru/butikov/Applets/Precession.html

Playing with the applet will show you a lot clearer the motion of the precession and fixed axis.

Have fun and keep experimenting.

Edit: Forgot to mention to select "List of Examples" and select the first option "Spherical".

Hi Farmhand,

Thanks for the drawing, it is clear now. Regarding the "shield", unfortunately you would have to use copper windings in series in a pattern as you drew and indeed either DC bias or pulse those windings or terminate with a load resistor because if the shields are not made of ferromagnetic material but passive metal plates then maybe eddy currents may have a "shielding" effect. And if the "shield" is a ferromagnetic material, then your tube magnets would attract to it of course (maybe causing eddy losses too) but in general this passive ferromagnetic shield would not cause much loss or drag and would prevent flux entering the pancake coil part it covers. Unfortunately I do not know any material (metal or non-metal) which blocks flux while it has no magnetic attraction to a permanent magnet.
All in all, only an actual test can give answers whether such setup has any advantage.

I understand your valid questions on Thane's acceleration under load setups. Years ago Thane was active on his thread of this forum on that topic and showed several videos he deleted since then from youtube but I do not recall scope shots you ask. I know he has again appeared occasionally on this forum and has again a youtube channel albeit with less number of videos. I think what you have found by your tests has given you an impression and answer on what really happens so you may wish to step over it. I do not think there is any secret Thane may keep under a 'shield'...

rgds, Gyula

Hi Farmhand,

No problem on the 'rant' on Thane, he really did not care much about the input power consumption at the time he showed the acceleration under load effect.  And later with his BITT transformers he argued that making the input power factor near zero the Watt-Hour meter would not measure input power anyway so he thought he had achived COP > 1 ...

I assume you stacked some magnets 'in series' on the rotor, this way the flux at the ends facing the coils become more narrow hence stronger for induction or interaction.  I believe that bringing more and more flux from permanent magnets into a system has benefits. I would show you an old thread where a member demonstrated with a simple experiment the advantage of using stronger magnets at a fixed input power. (Of course the focus of the flux by magnet stacking is also a good step if you really did that because eventually it also increases flux.)  This is the post by Ian, includes a 10 minute video http://www.overunity.com/1754/pulse-motor-video/msg35080/#msg35080   
His idea of using more and more coils in series and also in parallel combinations so that the resultant coil inductance and DC resistance remains the same as for any of the single coils sounds good (albeit it would not insure excess output), efficiency would increase for sure.  Back then I toyed with the idea of using a rotor disk of 30-40 cm OD and fixing many small but strong magnets on it while the many stator coils would be arranged like Ian described, using a moderate number of windings of thicker wire but I did not have the mechanical means for the bigger rotor mechanics.  While I do not fully agree with Ian on all his statements as a final outcome, the more flux involved can give more output torque for sure versus the single same coil - single same magnet setup (albeit the COP > 1 is still a question with it).

rgds,  Gyula