Claimed OU circuit of Rosemary Ainslie

[TinselKoala]

@TinselKoala,

I have uploaded the Rosemary Ainslie patent. Can be found here:
http://www.overunity.com/index.php?action=downloads;sa=view;down=290

Also, I noted that she state in an article that she did add inductance to the heating resistor. See snip. If you read the patent then you can see in one of the drawings that she put the heating resistor (load) in parallel with the coil. I think there is more to this circuit than just the load resistor.

Groundloop.

That patent application (not patent!) says completely different things wrt "duty cycle" than does the original pdf that started this whole thing, and also makes much more conservative claims as to efficiency. In the cases cited in the application the margin is so small that measurement error is the first suspect, and the devices described in the application are intended to work at completely different frequencies and duty cycles than the circuit in the pdf. As far as I can tell, that is.

In the pdf she represents the load symbolically as an inductor in series with a resistance because she uses a wirewound resistor of significant inductance. I'm using the same total inductance as far as I can tell.

[TinselKoala]

Grrr. See what happens when you open these cans of worms?

In the pdf of the EIT paper, on page 8, she describes a "control" experiment where she just hooks the 10 ohm wirewound load resistor up to a battery. And she calculates 17.7 watts as the average power dissipation here.
Then she cites and tabulates the results of an experimental run where she estimates the power dissipated in the load resistor, over the 997 minutes of the experiment to average 17.5 watts and total power 1.22 Megajoules.
And then that figure is compared to the calculated 67.6 kiloJoules calculated to have been delivered by the battery.
And this is where the COP>17 comes from.

BUT:

She states in the pdf that the circuit is ON for 3.7 percent duty cycle. However I have determined that the 555 circuit posted here produces a 96.3 percent ON duty cycle, and that it appears that she may be mistaken about what the true ON duty cycle is in her experiments.

Just roughly looking at the input parameters using this circuit, I get a 600 mV, nearly rectangular pulse shape, representing the voltage drop across a 0,25 ohm shunt. Ignoring the spikyness for the moment just to get ballpark values: that gives a current of 2.4 Amps, and times 24 volts that's 57.6 Watts, times 0.963 (actual duty cycle) gives about 55.5 Watts average power drawn from the battery. No wonder my load and mosfet heat up so quickly.
55.5 Watts times 997 minutes times 60 minutes per second gives about 3.3 MegaJoules input energy.

So the effect of the duty cycle mistake, if it is such, is to bring the COP down from >17 to about only 37 percent or so (COP<1/2), which is just what is expected in this kind of circuit. A substantial portion of the input power is going to the mosfet as heat, and some more is being radiated as EM waves, and some is even being reflected back into the battery.

What's not happening is "more energy out at load than in from battery".

[fritznien]

well done TK!

[poynt99]

Wow I certainly hope it wasn't this little "innocent" mistake that caught Rosemary, although it would not surprise me.

Many people invert their scopes or swap the scope leads around to make the wave form "look better" only to later fail to realize that all their measurements will be inverted as well, as TK has just shown can easily happen.

.99

[TinselKoala]

well done TK!

Thanks, but we need confirmation of several things before we can put the issue to bed.

First, did I build and connect my 555 circuit correctly, and am I right about its performance?
Second, did Ainslie really make the same error that I did at first? That is, is she really using 3.7 percent OFF instead of 3.7 percent ON?
And third, does her power calculation method (which seems a bit screwy from the outset) come up with the same input figures, proportionally, if the duty cycle correction is made?