Storing Cookies (See : http://ec.europa.eu/ipg/basics/legal/cookies/index_en.htm ) help us to bring you our services at overunity.com . If you use this website and our services you declare yourself okay with using cookies .More Infos here:
https://overunity.com/5553/privacy-policy/
If you do not agree with storing cookies, please LEAVE this website now. From the 25th of May 2018, every existing user has to accept the GDPR agreement at first login. If a user is unwilling to accept the GDPR, he should email us and request to erase his account. Many thanks for your understanding

User Menu

Custom Search

Author Topic: Pierre's 170W in 1600W out Looped Very impressive Build continued & moderated  (Read 423789 times)

konehead

  • Sr. Member
  • ****
  • Posts: 462
Hi dog one
Ok fwbr a.c. legs across coil
Dc side of fwbr into cap ( such as big caps already pre charged) as in DZ gen
Now two approaches:
Have switch on one of the a.c. legs of fwbr that  connects the recovery to caps
Time this switch to have delay time approx 5 degrees retarded
Works like an echo..i call this "active"recovery circuit
"passive" if no switch.on a.c. leg of fwbr.
Delay time longer if low frequency
delay time less if higher frequency...tune delay time to system as it operates.
Power in recovery output now huge as no reflection back to primary plus I will guess actual backemf forces coheree with flyback spikes energy all comes out same direction in dc from fwbr this the lamebrain theory but this really works.
2nd approach is shorting induced or pulsed inductor at peak...
If pulsed this means close switch that shorts out  coil just before saturation point.
If induced such as sweeping magnet past coil then close switch to short inductor very briefly  at sinewave peak period.
Have fwbr a.c. legs across inductor and dc side to cap only
Discharge cap by itself to load for power....
I suspect DZ generator in its crawling rotation of A AB B sort of
Coil energizing sequence perhaps causes A peak shorting type of event when one coil moves "over" the adjacent coil if polarities reversed maybe does not need to be reversed and get effect from constant movement and two coils being energized at once and remember power in magnets is at the edges and  so leading edge has reversed polarity as to trailing edge same with electromagnets maybe overlap of leading "top" edge over trailing "bottom" edge  of rotation causes brief coil shorting event if rotation speed and iming is right and so then voltage in super caps rockets up like crazy...maybe backemf inherent forces manifest in coherence too don't know.
Murder of Eugene Mallove vonvinced me 15 years ago  dark forces will kill to protect their power with no problem to them to do so
they don't care they don't have to.



« Last Edit: June 19, 2018, 01:33:28 PM by konehead »

T-1000

  • Hero Member
  • *****
  • Posts: 1738
With parametric resonance and ways to make nuclear reactor from the core while there is simple alternator function going on... That is what I am calling overthinking.
Guys, again, grab magnet, move over coil , see what parameters change is creating current on wire. Then make same with coils. Simplicity have to be here not overcomplication.

Cheers!

Jeg

  • Hero Member
  • *****
  • Posts: 1532
Hi :)
I am in the process of building my switches and i just received my main switching components (IGBTs).
 https://www.infineon.com/dgdl/irg4pc50u.pdf?fileId=5546d462533600a401535644832d22f2

Datasheet mentions that this IGBT is optimized for high operating frequencies 8-40 kHz in hard switching, >200 kHz in resonant mode.
Does that mean that i will have problems if i operate it at lower frequencies in the order of 100-1000Hz?


MenofFather

  • Hero Member
  • *****
  • Posts: 943
Hi :)

Datasheet mentions that this IGBT is optimized for high operating frequencies 8-40 kHz in hard switching, >200 kHz in resonant mode.
Does that mean that i will have problems if i operate it at lower frequencies in the order of 100-1000Hz?
No. Lower frenquency=less loses.

Jeg

  • Hero Member
  • *****
  • Posts: 1532
Thanks Menof
Good to see you ;)

listener192

  • Sr. Member
  • ****
  • Posts: 272
In a mechanical generator, you have momentum and twisting of the magnetic fields (opposed fields are generated, armature reaction) This is not increments or steps in momentum, armature reaction, and field intensity.  These parameters do not like to be incremented/decremented. 

To simulate momentum, you use overlap in the code to never break the fields rotation.  Again, its not perfect, there is room for improvement here.  I don’t see PWM on each stator coil as the answer, you do not want to shut off the field in rotation.  Hold it as it goes around.  The answer lies in the wire.  If you change the wire as it rotates, and change it in a VERY dramatic way as it rotates (stronger the intensity), you will have something.

I have also shown in my post why the code is not the same through loop.  It is not a glitch. When you learn variation, intensity, and momentum of the magnetic field through the wire only, the code now has purpose.

All while this is going on, the opposed generated field from the intensity is always reflected back and captured. You now see why 72 diodes were used instead of 36.  I believe there is a better way.  I’m not sure on best method moving forward at the moment, but active rectification on coil pairs makes sense to me, at least at this moment.

Remember that the load as generating its own opposed fields for more recovery.

Can anyone of you recommend a proper method for ACTIVE A/C rectification?  When coil is North, you recover differently than when coil is in south.  The stronger your simulated magnetic field, and load, the stronger your recovery needs to be as well.   There are new chips on the market for active rectification, but something tells me there is a simpler method.  So if anyone has experience in this area.  It would be nice to see help.

Jerdee
________
Fr.
Dans un générateur mécanique, vous avez un momentum et une torsion des champs magnétiques (des champs opposés sont générés, une réaction d'induit) Ce ne sont pas des incréments ou des pas dans l'impulsion, la réaction d'induit et l'intensité du champ. Ces paramètres n'aiment pas être incrémentés / décrémentés.

Pour simuler l'élan, vous utilisez le chevauchement dans le code pour ne jamais casser la rotation des champs. Encore une fois, ce n'est pas parfait, il y a place à amélioration ici. Je ne vois pas PWM sur chaque bobine de stator comme la réponse, vous ne voulez pas fermer le champ en rotation. Tenez-le comme il va autour. La réponse réside dans le fil. Si vous changez le fil pendant qu'il tourne, et le changez d'une manière TRÈS dramatique pendant qu'il tourne (plus fort l'intensité), vous aurez quelque chose.

J'ai également montré dans mon post pourquoi le code n'est pas le même à travers la boucle. Ce n'est pas un problème. Lorsque vous apprenez la variation, l'intensité et l'élan du champ magnétique à travers le fil seulement, le code a maintenant un but.

Pendant tout ce temps, le champ généré opposé de l'intensité est toujours réfléchi et capturé. Vous voyez maintenant pourquoi 72 diodes ont été utilisées au lieu de 36. Je crois qu'il y a un meilleur moyen. Je ne suis pas sûr de la meilleure méthode pour l'instant, mais la rectification active sur les paires de bobines est logique pour moi, du moins en ce moment.

Rappelez-vous que la charge génère ses propres champs opposés pour plus de récupération.

Quelqu'un d'entre vous peut-il recommander une méthode appropriée pour la rectification ACTIVE A / C? Lorsque la bobine est au nord, vous récupérez différemment que lorsque la bobine est au sud. Plus votre champ magnétique simulé et votre charge sont forts, plus votre récupération doit être forte. Il y a de nouvelles puces sur le marché pour la rectification active, mais quelque chose me dit qu'il existe une méthode plus simple. Donc, si quelqu'un a de l'expérience dans ce domaine. Ce serait bien de voir de l'aide.

Jerdee


Unless I turn coils off completely I see very little recovery in fact it is only seen into the small caps on the bridge boards not back into the cap bank. Turn coils off so no coils are on then you see 20 % or so recovered into the cap bank. Changing the rotor load makes hardly any difference to recovery only to DC input current.


L192











listener192

  • Sr. Member
  • ****
  • Posts: 272
Hi dog one
Ok fwbr a.c. legs across coil
Dc side of fwbr into cap ( such as big caps already pre charged) as in DZ gen
Now two approaches:
Have switch on one of the a.c. legs of fwbr that  connects the recovery to caps
Time this switch to have delay time approx 5 degrees retarded
Works like an echo..i call this "active"recovery circuit
"passive" if no switch.on a.c. leg of fwbr.
Delay time longer if low frequency
delay time less if higher frequency...tune delay time to system as it operates.
Power in recovery output now huge as no reflection back to primary plus I will guess actual backemf forces coheree with flyback spikes energy all comes out same direction in dc from fwbr this the lamebrain theory but this really works.
2nd approach is shorting induced or pulsed inductor at peak...
If pulsed this means close switch that shorts out  coil just before saturation point.
If induced such as sweeping magnet past coil then close switch to short inductor very briefly  at sinewave peak period.
Have fwbr a.c. legs across inductor and dc side to cap only
Discharge cap by itself to load for power....
I suspect DZ generator in its crawling rotation of A AB B sort of
Coil energizing sequence perhaps causes A peak shorting type of event when one coil moves "over" the adjacent coil if polarities reversed maybe does not need to be reversed and get effect from constant movement and two coils being energized at once and remember power in magnets is at the edges and  so leading edge has reversed polarity as to trailing edge same with electromagnets maybe overlap of leading "top" edge over trailing "bottom" edge  of rotation causes brief coil shorting event if rotation speed and iming is right and so then voltage in super caps rockets up like crazy...maybe backemf inherent forces manifest in coherence too don't know.
Murder of Eugene Mallove vonvinced me 15 years ago  dark forces will kill to protect their power with no problem to them to do so
they don't care they don't have to.


Very interesting but how would you accomplish this with the hardware that Pierre was using?


L192

listener192

  • Sr. Member
  • ****
  • Posts: 272
With parametric resonance and ways to make nuclear reactor from the core while there is simple alternator function going on... That is what I am calling overthinking.
Guys, again, grab magnet, move over coil , see what parameters change is creating current on wire. Then make same with coils. Simplicity have to be here not overcomplication.

Cheers!


What about offering the simple solution yourself?


Jerdee has said the same thing but I see nothing concrete from either of you!


L192

listener192

  • Sr. Member
  • ****
  • Posts: 272
The answer is in where the strongest position of each magnet pole is when you introduce paralel same pole magnets. With single magnet it is middle of the magnet diameter. With 2 magnets is on junction position between magnets. Same happen with coils. As soon you do that there is "movement" of magnetic pole. And if you point magnetic pole towards core of the coil it is not same when magnetic pole is matching middle of the coil core and where it "moves" a bit the the side. With air core coils it is also same relative position to the coil wire. If the magnetic field movement happen to cross wire and also is in middle of change polarity from one to another there is induction in the wire. This is required for power generation.
If you just weaken or strengthen the magnetic field density in PWM the resulting change of flux is like in any transformer only.


Yes you would have a pulsating wave and a traveling wave. Both would flux link as the flux through the rotor does not cut the rotor windings at 90degs. In a generator the static flux from the rotor splits into two paths around the stator and cuts the stator windings at the point of entry into the stator. As the rotor rotated different wires are cut.


There is no equivalent to this using switched coils on the stator as they just produced discrete stepped flux that gives the appearance of movement when looking at rotational force I.e the rotating magnet in place of the rotor but there is no physical movement of flux through space just a series of discrete fluxes at different angles relative to the center of the stator. logically putting coils directly in line with the path of this switched flux would not result in flux cutting just flux linkage.


L192

jerdee

  • Newbie
  • *
  • Posts: 49
L192, look at the field groupings that I’ve shown. DON’T cross your fields!!!  This is why you have no recovery!!!   The wiring of your stator needs to be different.  Your fields are all locked up with no recovery.  Also, use the coil groupings to create the leading/trailing edge of a magnet.   This is were I’m going in the research.

@ Konehead, I like your thinking on using switch on a.c. leg of FWBR.  I believe you use solid state relays for this.  Am I correct?   The body diode on our H-bridges will possibly be a problem still.  We are using the IBT-2 boards. For a 6 field rotation system.  I want to keep it down to 20 H-bridges with the field groupings.  This would require 20 MORE switches for the SSR.  Keeping MCU pin count down will be tricky. Can you recommend a good SSR?  We need the switch to be bidirectional.  I have a very high quality bi-directional switch that we can use, however, not 20 of them, and will cost more to build.  Hopefully I'm thinking in the right direction.  Correct me if I'm wrong.

As for the leading/trailing edge of the magnet.  Konehead!!!!You are bang on!  BRAVO!!!!   If you combine coil groupings in a very specific way you can simulate a leading/trailing edge of a magnet.  Run SAME field groups in parallel with SAME OFFSETS to create your leading/trailing edge of a magnet.   As the rotation changes from N to S or vs…you still have the leading/trailing edge combination of field groups.  Use the field groupings in parallel to create the stronger leading/trailing edge of a magnet.
This is for sure the right direction.

Jerdee

listener192

  • Sr. Member
  • ****
  • Posts: 272
L192, look at the field groupings that I’ve shown. DON’T cross your fields!!!  This is why you have no recovery!!!   The wiring of your stator needs to be different.  Your fields are all locked up with no recovery.  Also, use the coil groupings to create the leading/trailing edge of a magnet.   This is were I’m going in the research.

@ Konehead, I like your thinking on using switch on a.c. leg of FWBR.  I believe you use solid state relays for this.  Am I correct?   The body diode on our H-bridges will possibly be a problem still.  We are using the IBT-2 boards. For a 6 field rotation system.  I want to keep it down to 20 H-bridges with the field groupings.  This would require 20 MORE switches for the SSR.  But I'm thinking you can turn all 20 SSR on at the same time.  Keep the MCU pin count down. Can you recommend a good SSR?  We need the switch to be bidirectional.  I have a very high quality bi-directional switch that we can use, however, not 20 of them, and will cost more to build.  Hopefully I'm thinking in the right direction.  Correct me if I'm wrong.

As for the leading/trailing edge of the magnet.  Konehead!!!!You are bang on!  BRAVO!!!!   If you combine coil groupings in a very specific way you can simulate a leading/trailing edge of a magnet.  Run SAME field groups in parallel with SAME OFFSETS to create your leading/trailing edge of a magnet.   As the rotation changes from N to S or vs…you still have the leading/trailing edge combination of field groups.  Use the field groupings in parallel to create the stronger leading/trailing edge of a magnet.
This is for sure the right direction.

Jerdee
Well my coil scheme is the 5 coil version of the 6 coil version that Pierre described and used. Sure there are overlapping fields N and S just like Pierre had. He never showed us any recovery waveforms but the only way you see appreciable recovery with this scheme is if you turn coils off completely which he shows at the end of his code sequence. I am not so fast to step away from this as Pierre mentioned there is still a missing element. I have seen little effort to find out what this is and now you are wanting to proceed down a developmental path to improve on his original design which nobody yet understands.


L192

T-1000

  • Hero Member
  • *****
  • Posts: 1738
There is no equivalent to this using switched coils on the stator as they just produced discrete stepped flux that gives the appearance of movement when looking at rotational force I.e the rotating magnet in place of the rotor but there is no physical movement of flux through space just a series of discrete fluxes at different angles relative to the center of the stator. logically putting coils directly in line with the path of this switched flux would not result in flux cutting just flux linkage.
All coils must be have magnetic flux path over same output core just on different angles to the center like the moving magnet usually does. And in case of moving magnet there is flux cutting on the wire in transition from one magnetic pole to another. Also if magnetic pole positon changes position too fast there will be almost no current on the output coil. And when switching from one coil to another you need smoth power down in previous coil and power up in current coil which will slow down magnetic pole position change. One of possible solutions for that is to have capacitor across each coil for this purpose. And have all coils also connected with 3 plate capacitor (2 capacitors in series with center tap) from the power source. Then you can make coil power transition smooth as possible from one coil to another. And obviously it is a bit different approach than everyone are doing Pierre's case.

P.S> Pierre's 3 phase aligment have 1 strong and 1 weak pole facing output core. Which makes perfect sense when you want to have single North or Souh pole going across whole coil. Then changing smothly as possible to opposite pole.

pmgr

  • Full Member
  • ***
  • Posts: 184
    • Stop organ harvesting from Falun Gong practitioners
All coils must be have magnetic flux path over same output core just on different angles to the center like the moving magnet usually does. And in case of moving magnet there is flux cutting on the wire in transition from one magnetic pole to another.
T-1000 again, as L192 has already mentioned multiple times, there is no flux cutting. Why don't you take another look at the FEMM simulations I did for this device:

https://www.youtube.com/watch?v=bVARO4DH8uU

All the flux is contained inside of the stator and rotor and the only position where it crosses the air is in the gap between the rotor and the stator. It's a discrete flux cutting scheme.

PmgR

ariovaldo

  • Sr. Member
  • ****
  • Posts: 374
« Last Edit: June 20, 2018, 03:35:58 AM by ariovaldo »

listener192

  • Sr. Member
  • ****
  • Posts: 272

This is just the controller to simulate the magnet "passing" in front the coil.
I will rewind one of the stators that I have.
https://www.youtube.com/watch?v=nCReDKa6Fn8


Let's to see the results..


If this were true it would show up in a FEMM simulation.


L192