Rosemary Ainslie COP>17 Circuit / A First Application on a Hot Water Cylinder

[gyulasun]

Hi Fritz,

You wrote a very good post, I agree with it in every respect and unfortunately I surely misunderstood Rosemary's earlier post by assuming the batteries are connected by a controlled switch. This is why I answered that way in my previous post.

Thanks,  Gyula

[fritz]

a controlled switch would make sense 
especially with an inductive load.

[Rosemary Ainslie]

Gyula and Fritz,  yet again thanks for the explanations.  I see where I was wrong.  There's far less resistance in the battery than I realised.  And however hard one switches - it's going to be increasingly difficult to keep that potential difference between the batteries as those switching cycles develop.  Therefore we would need to use that inductance on the resistor as per the usual circuit.  I've also had it on good authority that the two battery arrangement may not necessarily be any better than one.

That leaves us with the same problem.  How do we use a MOSFET which is known to be brittle at higher wattages.  One can put them in series - as explained by just about everyone.  But my own experience of this is disasterous.  One blown and the others follow in an expensive cascading series where we're left with nothing but that acrid smell of burnt metal (very distinctive). 

By the way Gyula - the preferred 'oil bath' calorimetric test is acknowledged as preferred.  It just was not really required as we were not measuring 'marginal' over unity results.  Had we been doing so then - of course - it would have been required to immerse the resistor in some sort of oil or somesuch liquid.  Those protocols that we applied to all our tests were designed by academics and considered as sufficient - albeit not one academic was prepared to attend a demonstration to accredit those results.  But their objections in those days - were largely based on the 'understanding' or, dare I say it, 'belief' that the circuit - however it was configured - would not be able to 'recharge' itself.  Our measurements and battery performance put paid to that 'belief'.  Indeed - as shown by Bedini et al - the battery rather prefers to be recharged with that heavy duty spike.  It's just that we didn't even bother with the a diode to take that energy back.  It seems that the only path required was supplied by the internal body diode of the MOSFET itself.  In any event.  Our learned and revered were entirely satisfied that a battery could not recharge itself - however it was configured.  They were wrong.  LOL

I was told by sundry transistor manufacturers that the cost of developing a MOSFET at the tolerances we require - would cost in the region of 500 000 Euros.  That's a little outside our budget.

I'm going to see if I can improve that 'give and take' battery configuration to see if I can find some solution.  I still feel that it's a means to get that extra energy.  But it will, undoubtedly, need to be more closely defined and designed.  Perhaps one can use it in conjunction with MOSFET - somehow - to get the benefit of both.  Anyway.  I see more sleepless nights ahead.  LOL

Thanks for your input.  Much appreciated.
Kindest regards,
Rosemary
http://www.scribd.com/aetherevarising

[Rosemary Ainslie]

Guys - this is a big appeal to you all.  I've been approached by a broadcaster to do a documentary on this 'over unity' movement that's gaining momentum all over the place.  I only really know my own circuit. It's a little off topic but I'd be very grateful if you could give me some names of all those 'inventions' that have been proven to give over unity.  Especially as it relates to motors.  Steorn is the one that springs to mind.  Are there others?  I'd quite like to forward this so that he can do the required investigation.

Thanks,
Regards,
Rosemary

[fritz]

If we assume 3 identical 12V batteries  20Ah capacity, 25mOhm internal Resistance, identical charge condition (1+1 == 24V, 1 == 12V), I would expect the following:
Battery Voltage each 12.6V, 3times internal resistance = 75mOhm, Voltage loss diode+switch 2V-> voltage difference==10.6V, resistance 75mOhm
Well, that would give a theoretical value of 140Amps. The major impact will be the wiring and contacts involved (which is missing but would play the dominant role). I would expect something around 30 Amps.
The problem with 140Amps diodes is that they are not the fastest ones.

I would expect the transfered charge as u(diff)(10.6)*i(==u(diff)/(internal resistances+wiring)) *t(on)
->@ 320W (@30Amps) - adding some losses, heating up - we transfer maybe @250W

The diodes would heat up (@30Amps, 1V drop) with 30W each.

So I would assume that you blow any diodes and switches using 20Ah lead-acid with fast diodes at the first pulse.

To be correct - I would estimate these figures for current pulses >5ms.
In your circuit you have pulsed DC with  500us pulse duration.

I don´t know your batteries - nor do I have my own data or data from a battery manufactorer at hand so - nothing we can base on.

If we reach down to 1us - I would estimate the internal resistance in the area of few ohms. For 500us - a current from 2-5Amps would sound reasonable.

This is why I heavily suggest to determine the AC input/output resistance of batteries used with such setup.

How to determine AC impedance of a battery ?
Use 80´ties DC coupled HIFI-AMP+sinus generator+output capacitor.
(or industrial servo amp or similar)
Determine the internal resistance as function of frequency using low inductive precise test load (1Ohm).
Measure damping as function of frequency if feeding to battery terminal.
The battery impedance can then be calculated from that damping function corrected with the values from the 1Ohm test.

rgds.

fritz