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

[seaad]

Suggestions to different step pattern to achieve the magnetic locomotion/ rotation.
1 = Now    2a, 2b = New

[partzman]

Suggestions to different step pattern to achieve the magnetic locomotion/ rotation.
1 = Now    2a, 2b = New

Seaad,

IMO, not enough thought has been given to the overall requirements for an efficient rotating electromagnetic field.  For example, take a 5mH coil with .1 ohm dcr we wish to make into an electromagnet at a current level of 1 amp.  To initially charge this coil from say a 10 vdc supply will require a duration of 500us (neglecting any dcr loss) which will consume 2.5mJ of energy.  Now, to maintain 1 amp in this coil for our electromagnet will require a dc voltage of 0.1v which over the same 500us period will consume only 50uJ or 1/50th the charge energy.

This is why Pierre uses a form of switching overlap in an attempt to maintain a constant but moving electromagnetic field.  However, several things to note here-  1) If the mmf or H field or inductor current is maintained at relatively constant level, then little is available for recovery to the power supply.  2) Overlap is not necessary if one considers the action of the collapsing electromagnetic field even in the N_S_N_S_etc configuration.  3) No consideration is being given to the coupling or K factor of the coil arrangement in the above conditions IOW, leakage inductance is ignored.

Also, in Pierre's 1st generator the stationary rotor is driven or induced from the stator's flux fields.  In his 2nd generator design, the rotor will be induced by the H field produced by bucking fields in the stator windings or he will have little or no output.

Regards,
Pm   

[T-1000]

Seaad,

IMO, not enough thought has been given to the overall requirements for an efficient rotating electromagnetic field.

True, most people just gone and made standard way 3 phase overlapping coils which will result in bucking mode on Pierre's way of driving.
Also after looking more carefully around Pierre's stator winding directions there are 31 coils with wire going CW and 5 coils coing CCW. Why is that no one know but the output coil core is still attached to 5 CW coils. Which allow to amplify magnetic field when turning on two coils in paralel.

Cheers!

[seaad]

Seaad,

IMO, not enough thought has been given to the overall requirements for an efficient rotating electromagnetic field. (Math.)

This is why Pierre uses a form of switching overlap in an attempt to maintain a constant but moving electromagnetic field.  However, several things to note here-  1) If the mmf or H field or inductor current is maintained at relatively constant level, then little is available for recovery to the power supply.  2) Overlap is not necessary if one considers the action of the collapsing electromagnetic field even in the N_S_N_S_etc configuration.  3) No consideration is being given to the coupling or K factor of the coil arrangement in the above conditions IOW, leakage inductance is ignored.

Also, in Pierre's 1st generator the stationary rotor is driven or induced from the stator's flux fields.  In his 2nd generator design, the rotor will be induced by the H field produced by bucking fields in the stator windings or he will have little or no output.
Pm   

Hi partzman
Thx for your math example. Your skills there are way above mine. I'm more of a try and error guy ( to lazy, or bad memory for math) and I let the nature tell me. Hope we , I can use that skill.
I understand though your point. It's like accelerating a car and burn fuel.

"" IMO, not enough thought has been given to the overall requirements for an efficient rotating electromagnetic field. ""

Agree! We have to make fundamental experiments about the principles first instead of building complex contraption in the kW-class. See my small experiments in the beginning of this thread.

"" This is why Pierre uses a form of switching overlap in an attempt to maintain a constant but moving electromagnetic field. ""
So did Clemente Figuera also, " always in contact with two collector bars" IMO this Pierre thing is, acts as a Figuera Machine!
But I'm still having heavy doubts about IF THIS PIERRE THING REALLY WORKS!  To slow repetition rate to feed a M.W. Oven!

"" 1) If the mmf or H field or inductor current is maintained at relatively constant level, then little is available for recovery to the power supply. ""
Se my pic. Yes, the coils between the fed, Switched coils have a short time of DC-rest (steady).

But how can we collect and forward the short cutted coils power??



"" 2) Overlap is not necessary if one considers the action of the collapsing electromagnetic field even in the N_S_N_S_etc configuration. ""
Overlap of what. Pulses , coils??    See my post http://overunity.com/17609/170-watts-in-1600-watts-out-looped-very-impressive-build-and-video/msg518124/#msg518124

But if some energy goes, transfers to the fix output "rotor" coil(s) from any of L1 to L5 in my example less power of course is looped back to the source.
In Pierre's case we don't have collapsing fields only shorted fields.   The principle of an OU unit is that the primaries should never "see, know" that the secondary steals power from them!

"" 3) No consideration is being given to the coupling or K factor of the coil arrangement in the above conditions IOW, leakage inductance is ignored. ""
Agree! As I call it "transformer effect" See efficiency in my two pics with the same function but different coil arrangement.I Have tried different K-values in my simulations, but it's impossible there to simulate a Genny-Stator.

Regards Arne

[partzman]

Hi partzman
Thx for your math example. Your skills there are way above mine. I'm more of a try and error guy ( to lazy, or bad memory for math) and I let the nature tell me. Hope we , I can use that skill.
I understand though your point. It's like accelerating a car and burn fuel.

"" IMO, not enough thought has been given to the overall requirements for an efficient rotating electromagnetic field. ""

Agree! We have to make fundamental experiments about the principles first instead of building complex contraption in the kW-class. See my small experiments in the beginning of this thread.

"" This is why Pierre uses a form of switching overlap in an attempt to maintain a constant but moving electromagnetic field. ""
So did Clemente Figuera also, " always in contact with two collector bars" IMO this Pierre thing is, acts as a Figuera Machine!
But I'm still having heavy doubts about IF THIS PIERRE THING REALLY WORKS!  To slow repetition rate to feed a M.W. Oven!

I have my doubts as well in this regard and for the following reason.  The transformer used in the typical microwave oven for economics sake is designed to operate near maximum flux density of ~1.5T at 50 or 60 Hz depending on location.  If this transformer is run from say a frequency 20% lower than the specified operating frequency, the laminated iron core will saturate and draw a considerably larger current than normal.  If operated at a frequency of 10-20 Hz, the saturation current would be limited only by the dc resistance of the primary and would be huge!  This raises a huge red flag for me and IMO has not been addressed by anyone satisfactorily.  However, for the time being I will give Pierre the benefit of the doubt until evidence points otherwise.

"" 1) If the mmf or H field or inductor current is maintained at relatively constant level, then little is available for recovery to the power supply. ""
Se my pic. Yes, the coils between the fed, Switched coils have a short time of DC-rest (steady).

But how can we collect and forward the short cutted coils power??

I don't have an answer for you on this at the moment due to what I believe is a far more complex operation than what is apparent.  Pierre says recovery to the cap bank is important and also says that maintaining the field strength is important but I fail to see how both can be accomplished together.  However, there will be a parametric change (increase in inductance) in the coils that are closest to the rotor and if this occurs when the electromagnetic field is held constant, this can provide a gain mechanism based on my previous bench tests so this may provide a means for recovery to the supply.

Also, the coil combinations that are operating in the buck mode may produce an H field in the direction of the rotor which would reinforce the available output energy.  If true however, this would be reflected via Lenz to the stator windings so no gain would be realized.

"" 2) Overlap is not necessary if one considers the action of the collapsing electromagnetic field even in the N_S_N_S_etc configuration. ""
Overlap of what. Pulses , coils??    See my post http://overunity.com/17609/170-watts-in-1600-watts-out-looped-very-impressive-build-and-video/msg518124/#msg518124

The coils would all be in series but the pulses driving the coils would not need to be overlapped if recovery diodes are connected from coils to the supply rails.  Certain overlap timing can be used with some recovery current seen at the supply and I have a partial simulation which proves this point.  I'm trying to do a complete 30 slot sim with rotor load but the task is huge due to the coil coupling factors which if incorrect, prevent the simulation from converging or completing.  If overlap can't be prevented in the switching circuitry, it should be at a minimum to provide maximum flux transfer to the rotor.

But if some energy goes, transfers to the fix output "rotor" coil(s) from any of L1 to L5 in my example less power of course is looped back to the source.
In Pierre's case we don't have collapsing fields only shorted fields.   The principle of an OU unit is that the primaries should never "see, know" that the secondary steals power from them!

"" 3) No consideration is being given to the coupling or K factor of the coil arrangement in the above conditions IOW, leakage inductance is ignored. ""
Agree! As I call it "transformer effect" See efficiency in my two pics with the same function but different coil arrangement.I Have tried different K-values in my simulations, but it's impossible there to simulate a Genny-Stator.

Ah yes, so you know exactly what I'm talking about regarding the couplings.  There is a way to simulate any electromagnetic topology in LtSpice and that is to model the core and windings with gyrators and capacitors using behavioral sources but the degree of difficulty for this type of device would be so complex that it would be easier to just build it!

Regards,
Pm

Regards Arne