170 watts in - 1600 watts out - looped - Very impressive build and video

[hartiberlin]

Here is a proposal on how this device could produce 60hz output.

Viewing the video at .25x speed, the timing of one complete cycle of the control board's leds appears to be 1s.  It also appears that 2 switches are on at any one time, so this would be equal to .025s in real time with a total of 18 intervals.  Therefore, each switch interval is .025/18 = 1.389ms.

Calculating the total number of intervals at 60Hz is (1/60)/1.389 = 12.  Since there are 36 total coils, 36/12 = 3 coils per switching interval.  However, to produce an even magnetic field across the stator, 2 pairs of 3 coils each positioned oppositely from each other on the stator would be required and driven with opposite polarities.  So, we have two 3-coil pairs sequentially switched on for a period of 1.389ms for a total of 12 intervals to complete one cycle which would equal 16.667ms which is 60Hz.

Regards,
Pm

Yes, I could be wrong with 16 Poles, maybe only 12 then ?
Could you please draw up a drawing how you see it ?

Well, yes as the flux lines must go out into the air and not stay inside the stator iron, he has to use some kind of trick to force the fluxlines go out into the air...
So he might use 3 coils to do this, so the middle one of each pole is the one which flux lines are forces out of the core....Do you mean it like this ?
Many thanks.

[seaad]

Stefan, all
N-N or S-S ends opposing each other makes the Flux Lines to "escape" from the iron core where they collides!  Is a perpendicular (iron) core nearby it sucs in the flux.  ,  Or maybe Electromagnets in a row arranged as that FIX-magnet-array making the field to be stonger/ strengthened on one side (Halbach array).

Regards  / Arne

[partzman]

Yes, I could be wrong with 16 Poles, maybe only 12 then ?
Could you please draw up a drawing how you see it ?

Well, yes as the flux lines must go out into the air and not stay inside the stator iron, he has to use some kind of trick to force the fluxlines go out into the air...
So he might use 3 coils to do this, so the middle one of each pole is the one which flux lines are forces out of the core....Do you mean it like this ?
Many thanks.

The stator has 36 slots and 36 poles.  It appears he has wound each coil to encompass 5 poles with a total of 36 coils.  I propose he has then connected three consecutive coils in series for a total of 12 coil sets.  He then sequentially switches these 3-coil sets along with each opposite 3-coil set until one cycle is complete and then repeats the process.  The stationary rotor can then be placed basically anywhere in the stator to complete the magnetic path of the sequenced 3-coil sets.  The 3-coil sets that are adjacent to the ends of the rotor would produce the highest or peak voltage of the sine wave output.  The basic question is, can the energy recycled from the field collapse of the coil sets back to the super caps exceed the input energy?  It would appear so!

I'll have to dust off my AutoCad skills to produce a drawing!

Regards,
Pm

[pmgr]

There are some other observations:


1. Based on wall plug and the fact his English is not very good (and based on his French name), he is most probably located in the French region of Canada.


2. The first video states the home made transformer is 36V 30amps. With a 120V main line, this means the ratio of windings is 120/36=4/3. Driving a 2.5A current in from the mains side is equivalent to a current of 4/3*2.5=8.33amps on the secondary. This is rectified and goes through a 4ohms 50W resistor. The power generated in the resistor is I^2*R=(4/3*2.5)^2*4=278W! This will explain why in video 4 the resistor is very hot, BUT, I am not sure how a resistor can survive this much power for so long if it is only rated for 50W.


3. I notice a thin red wire going from the diode bridge into the same bundle as the wires coming from the cap/resistor. Then a black thin wire pops up at the end of this bundle together with the red wire, but I don't see where it is coming from (maybe it is spliced into the thicker black wire somewhere). Another intermediate size red wire is spliced onto the resistor and this wire appears to go to (or come from) the diode recovery board. He states this is "the return wire of the coil that will charge the supercapacitor". If it is charging the supercapacitors, not sure why it is by passing the resistor as it would expose them to any high voltage BEMF which could possibly damage the capacitor boards.


4. He is using three super-cap boards which appear to be wired in series (not parallel). This is to increase the max voltage that can be applied to them. Each of the board has 6 caps in series, the caps on each of these boards are 500F with a working voltage of 2.7V. So 18 of them in series will allow for a max voltage of 18x2.7V = 48.6V. Total capacity is reduced to 500/18=27.7F. Energy stored in them at 24V applied voltage is 0.5*C*V^2=0.5*27.78*24^2=8000 Joules. So that is 0.00222 kWh, or 2.22 Wh or 79.2kWs. This means he could run a 1100W microwave for about 72secs max.


5. Motor stator appears to come from a three phase motor. Identification is 36A01W462 F1009/1759 (or 1760). I did a search for the stator number and found this:


http://attachments.temcoindustrialpower.com/Data_sheet/Baldor-M3613T-50.pdf


https://www.easa.com/resources/booklet/typical-failures-three-phase-stator-windings


I probably didn't wind this stator himself. Extremely difficult to do that. Maybe he just used an off the shelf stator and then wired each of the 36 windings up to his control board to the way he needed it.


6. It does look like he is driving the stator in a 3 phase fashion. In his first video, notice how he has three connectors from the control boards hooking up to the stator.


7. His arduino is from www.sainsmart.com. But Googling I can't find any reference on Arduino 2556 (I do find 2560).


8. The diode recovery board is grouped into three groups: 2x group of 27 diodes, 1x group of 18 diodes for a total of 72 diodes. If it is driven 3-phase, why are they not grouped into three groups of 24?


PmgR

[e2matrix]

There are some other observations:


<snip


7. His arduino is from www.sainsmart.com. But Googling I can't find any reference on Arduino 2556 (I do find 2560).

snip <

I found reference to an Arduino Mega 2556 in a University of Wisconsin document for what it's worth.  Apparently not a common Arduino board but there are a few references to them around.