Muller Dynamo

[gyulasun]

...
Anyway, i built 5 romero driving circuits (need to do 3 more) and i collect the bemf from all 5. The 22000uF/35V cap is filling up in 5-6 seconds and all this with less curent consumed  and a gain in rpm

Hi Marius,

Just realized that you do not need 5 isolating transformers but one.  It is just enough to use one such 1:1 transformer to feed the DC/DC converter to insure there can be no short circuit between the rest of the stages when you close the loop.  (The supply wires of the in-between stages do not disturb each other in any way till looping so you can wire the stages as you showed in your schematic yesterday and use a 1:1 transformer in the last driving stage to feed the DC/DC converter fully isolated from the other stages.)  What do you think?

Gyula

[mariuscivic]

Hi Gyula

I have tryed one transformer but is not working.I have almost nothing in the other coil.It behaves just like an resistor. In my curent setup i'm not recycling the bemf but, i store it into this large new cap(47000 uF) from all 5 driving circuits.The cap charges up to 50V  and  discharge it with no drag. But the voltage must be always over 15V in order to see no drag. 
This is my new video where you can see that the super cap is being charged with no drag
http://youtu.be/N0mviNIk34I

[gyulasun]

Hi Marius,

Now I think I understand your setup and how you collect back emf from each coil pair into a common puffer cap via individual rectifier diodes.  At the moment all I can suggest is to use an isolating type of DC/DC converter like joefr showed yesterday in an ebay link, maybe you can obtain such or similar in your country too. 
I will think on this setup how the isolation of the input supply pins from the puffer cap supply pins could be achieved. 
Do you happen to have an L meter? It would make things easier a bit    

Gyula

[mariuscivic]

Hi Gyula

For now i'll try to adapt this circuit.The secondary of transformer should go to the battery

[konehead]

Hi Joe

Nice experiments - easy to see everything going on in your video too...

My two cents worth are first your "primary" pulse width seems sort of wide to your motor coil - it looks around 50/50 on off duty cycle too...and the pulse "flat tops" for awhile too - I think you will get better perecentat of power recvoered vs power spent in backemf/recoil recovery if you dont let those motor coils "saturate" like they do (as soon as that voltage-peak starts to go horizontal it is sort of saturated   I mean by thiis)
Someone told me long ago that the primary pulse will work like big snowball rolling down the hill if it stays on too long - and it will snowball that backwards spike and backwards bemf into going the way the primary wants to go wheich is backwards to the way the bemf stuff wants to go - so anyways if you can somehow clip that morot-pulse much shorter pulse width I think it will really help....if you want the same amount of power and rpm as you have now, then do a cluster of pulses say 4 or 5, and pull out bemf/recoil after each one too (!) Last thing bill Muller told me is you wouldnt beleive the power you get when you do a cluster of 4 or 5 pulses...

OK there is that - also I think that when you do disconnect cap from "source" during that time the bemf/recoil circuit is not connected so you aremt gathering anything during that time and it is also time for spikes to squeak through blow up a diode or swtichg, or maybe cause some lugging to the rotor since they oppose direction of rotor...

This is one reaosn you want to use diode-plug circuit where you use single diode into cap A other diode into cap B and cap A discharges to load when cap B fills up...

All that said, what I like to do with BEMF/recoil is SWITCH it out - put a 2nd swtich that connects the coil to the "steering diode" or AC leg of FWBR if you are doing it like that....so you can "turn on and off" the recovery circuit.
you will need bidirectional mosfets if you put swtihc on AC side of things coming off coil...
The only time recovery-circuit needs to be ON is right when that motor coil switch turns OFF - that is when everything happens - other time is at that negative spike flipover point at bottom of scope shot (SHORT IT THERE INTO CAPS !...other subject)

Anyways if you swtich-out your backemf do the same pulse widht as your motor coil pulse width, and delay it about 5 degrees....put a resistor (6ohm say) across the cap being filled and adjust that delayed timing until you get a good speed up effect (I thnk it wants to work like an electronic ECHO) ...substitute the resistor with a huge DC cap....run 2nd mmotor coil circuit with huge cap....recover eveything off 2nd motor coil circuit too, and run back into 1st motor coil circuit and loop it see if it will run forever eh...