PhysicsProf Steven E. Jones circuit shows 8x overunity ?

[JouleSeeker]

Speaking of humor... see attached.  We do need a little lightening up now and then.

Scarecrow invited me to be on his show tomorrow, hopefully it will work out.
He will have some time for Q&A if you wish to show up. 

[JimU]

600+ pages for 20 bucks -- I'm going to spring for this book.  Thanks, Jim.
I discussed the experiment with a physicist last evening at some length.  A few modifications arose from that discussion, but the conclusion was the same (apparent non-conservation of momentum).

1.  Have the current on in coil A for some period of time at the start, so the B-field at B is established.
2.  Turn A off at the same time that the current in B is turned ON. 
In this way, B is immersed in the field from A when it turns on, so receives an impulse to the right,
and A will be off (and open so no effective eddy currents) when the field from B arrives.


Keeping an open mind is key to scientific progress, IMO -- thanks, Jim.

Hi Dr. Jones & all,

Per an open mind and relativity, have you perused the corpus of work
done by the Process Physics group at Flinders U in Adelaide, AU, which
is Prof Reg Cahill's group?

http://flinders.edu.au/science_engineering/caps/our-school/staff-postgrads/info/cahill-r/process-physics/
http://flinders.edu.au/science_engineering/caps/our-school/staff-postgrads/info/cahill-r/process-physics/papers/

They have re-analyzed the Michelson-Morley experiment, and the various
subsequent interferometer experiments and find they were not null results
after all.  In fact, they all agree on a preferred reference frame and our
velocity w/r/t to it.  More recent experiments of the one-way velocity of
lght have shown the same results, and the same velocity vector in space.

From this, they derive a new gravity theory which explains various anomalies,
such as dark matter (no need for it), etc.  There is a lot more and quite
interesting.

----------------------------------------------------------------

Per measuring the output power of your circuits, I've not been carefully
following all the great work done here, so this may be redundant, but
JL Naudin on his 2SGen project uses what looks like a simple and
effective approach:   A diode bridge on the ouput feeds, on each
leg, a parallel cap & resistor of suitable sizing, which then go to ground
or the circuit return.   So, the output pumps the caps up to a voltage level
on each leg that sustains the drain through the resistor.  Simply
measuring the steady-state voltage on each cap lets one compute
the output power on each leg of the bridge.

Section 7 of Mr. Naudin's 2SGen project shows a good schematic:

http://jnaudin.free.fr/2SGen/html/s2genep7en.htm

Actually, this section 7 of JL's 2SGen project, when I analyze it for
power, seems to show a clear 2:1 out/in power ratio.  Curiously,
JL himself never analyzed in this way, instead preferring to compare
the ratio of the two output legs of the diode bridge, per magnetization
and demagnetization, focusing on a theory by N. Zaev.  But, maybe
he's already got the tiger by the tail here!  Top level link to 2SGen:

http://jnaudin.free.fr/2SGen/indexen.htm

Anyway, perhaps Mr. Naudin's output measuring method would be
helpful here, if it has not already been tried out.

Regards,     Jim

[JouleSeeker]

actually i already used that method on a previous test

   link-->http://www.overunity.com/index.php?topic=10773.msg290346#msg290346

and it didn't seem to tally at all with the previously reported DVM-based results (which suggested 'n' = 1.3 approx, iirc)

hence my latest test looking at the possibility of increasing or maintaining the state of charge of the supply cell


trying to 'simulate' a load level across a capacitor can be counter-intuitive sometimes - it's possible that you don't end up with an equivalent 'load resistance' but instead you form a 'potential divider' arrangement between source impedance and 'load resistor' - in which case the capacitor terminal voltage will tend to change towards the 'o/p' voltage of the divider


taking up your comment about the LED & Diode, above, i agree the two will cause different behaviour of the test circuit - hopefully, their target loads (AAA NiMH and 2200uF capacitor) will dominate over any differences in the forward transfer characteristics of the two components


thanks
np


http://docsfreelunch.blogspot.com

Thank you for these insights, NP.
I mentioned your work on the Smartscarecrow broadcast this evening -- hope that's OK.  I really appreciate your work.

@JimU, note the comments by NP above, which jive with my experiments. 

Experimentally -- I placed a 10mF cap across Rout and found 5.1 V across the cap when the variable resistor was 2.51 Kohms.  This gives Pout ~ V**2/R = 5.1**2 / 2.51 K = 10.4 mW, which is not unreasonable, but evidently n<1 for this case.  (And not counting the Power dissipated in the LED, which was very bright, not sure how to do that reliably.)
  However, I found that the waveform (power) is significantly distorted -- and see caveats above.

Thanks for the URL's - I hope to get to those tomorrow.

[JouleSeeker]

  While it is very difficult (at least for me) to measure Pin and Pout reliably and simultaneously, I think much can be learned just from the effort to reduce Pin in this circuit, or ANY JT-type circuit.

I'm proposing a little "contest" to this end, to see which circuit can draw the LEAST POWER INPUT and still light an LED with reasonable visibility.  (That's a little hard to define, but say -- visible in a lighted room -- and visible in a photograph.) 

By varying Cb, Rb, the wound-toroid, I've reached 0.17 mW Pin -- see photo showing the set-up.

It will be necessary to measure Pin reliably -- I used the Cap/stopwatch method.
P = Ein/time, measuring Ein using a cap, from 2.55 V to 1.5 V, so around 2Volts in.
Best result (to date):  12.7 seconds using a 1000 uF cap for Piin, so
P = 1/2 10mF (2.55**2 - 1.5**2)/12.7s

= 0.17 mW = 170 uW.

Conditions: Rb = 47 Kohms
Cb = 223 pF (ceramic cap)
MPS 2222 transistor
Ro =  220 ohms
Lo ~ Lb ~ 130 uH
No CSRout or Co or Rr or CSRin.

[JouleSeeker]

(could not get this portion to post, so adding here)

I believe I can measure Pin quite accurately, to roughly +/- 5%, with this method -- once I measure C, which I need to do.  (At a colleague's in town.)   

It would be helpful to know the frequency of operation of the devices, but not necessary since not everyone has an oscilloscope (or other means to determine frequency).

I would like to encourage replications (and learning), and so I'm proposing a small "contest" -- to see who can reach the lowest Pin for any JT-type circuit.  (See, e.g., attached)    Pin to be measured by this method, cap + stopwatch for Pin.  LED present, visibly glowing.
Incentive (wish I were richer; this is just to make it fun; I just found $100 tonight I didn't know I had!):  Lowest Pin in one month -- on July 17th,  will receive $100 minus ($microwatts/10). 
THus, my entry today would be $100 - 170/10 = $83.

However, please announce results as you go along (as I'm doing above), so we can see the progress.

If someone reaches Pin = 0 in a self-runner, that goes to $200 PLUS I'll be glad to help you get this forum's OU prize which now reaches roughly $20,000.

Just for fun -- and learning!