PhysicsProf Steven E. Jones circuit shows 8x overunity ?

[xee2]

ENERGY CALCULATION USING VOLTAGE CHANGE

10,000 uF capacitor
start voltage = 1.366
end voltage = 0.936
time = 21 minutes = 21 * 60 seconds = 1260 sec.

Joules = 0.5*C*(V1^2 - V2^2) = 0.5*10000e-6*(1.366^2 - 0.936^2) = 0.0049493

watts = Joules / seconds = 0.0049493 / 1260 = 3.9 uW

[JouleSeeker]

Steven

i made an interesting discovery - the tertiary winding for the o/p in my previous circuit is  redundant!  (see below for updated schematic)

the latest circuit, using a 1000uF 35V cap as C2 (all other other parts as posted above) takes 687 seconds (11min 27sec) to discharge a nominal 1000uF cap from 2.55V to 1.5V

C2 1000uF (nominal)
2.55V => 3.251mJ
1.50V => 1.125mJ
                    -------
           Ein: 2.126mJ

Pav: 2.126/687 = 3.1uW (including cap leakage)


thanks
np


PS  if the 'blanking' voltage is different between two different coloured LEDs and nothing else changed in the circuit, presumably then it's related to the excitation levels involved in the different turn-on and forward-voltage drop characteristics of different colour LEDs?


http://docsfreelunch.blogspot.com

Remarkable achievement, NP!   Hope you keep experimenting... you might get to a self-runner yet   .
(Has the 3.1uW been corrected yet for cap leakage?)
I like the simplicity of this circuit of yours; and of Xee2's circuit.

@Nick:  Yes, increasing the R in series with the LED increases the voltage at which the LED turns off, then back on.  For 21 ohms = R, the green LED goes off at ~ 1.694V and back on (with less power consumption) at 1.664 V.  Interestingly, the Pinput calculated using the cap/time method is about the same for R= 21 ohms (9.8uW) as for R=1 ohm.

In any case, NP's circuit is still the front-runner with regard to the LOWEST Pinput.  (NP, have you found a 1N4148 yet?)

[JouleSeeker]

Just saw your post, Xee2 -- 3.9uW is also remarkable!  Thanks for using the cap/time method.

ENERGY CALCULATION USING VOLTAGE CHANGE

10,000 uF capacitor
start voltage = 1.366
end voltage = 0.936
time = 21 minutes = 21 * 60 seconds = 1260 sec.

Joules = 0.5*C*(V1^2 - V2^2) = 0.5*10000e-6*(1.366^2 - 0.936^2) = 0.0049493

watts = Joules / seconds = 0.0049493 / 1260 = 3.9 uW

Now, NP is going from about 2.55V down to 1.5 V and you're going from 1.366V to  0.936V.  (My best Pinput to date is ~7.2mW, not in the running...   )
To compare circuits, it is important to have consistent starting and ending voltages.

I've observed with my Xee2 replication (as explained above) that below a critical voltage, the Pinput consumed goes down dramatically, so this circuit should be tested below your 1.37 volts. 

1.  Can you both test from 1.37 V down to 1.10 V, using the cap/time method?  That should give plenty of time to get an accurate reading.  You may use different caps as you wish, but please give the correction for cap leakage (done by seeing how the cap voltage drops over the same time as the run -- but without cap connected to the DUT. Subtract this "effective cap-power loss" from the measured Pinput. This leakage correction will be smaller for shorter time in the run).

2.  AND, if you have a 1N4148 diode available, pls repeat the test with this "standard diode" in place of the LED. 


We still have some days left in the "competition" and you guys have done remarkably well, and I htink we've learned a lot (I know I have).  Actually, I'd like to move the "end date" for the contest up to July 1st, 2011, if no one objects.  Fast recent progress suggests this change.  And my wife says we have a trip to see grandchildren a bit later in the month, so would like to get closure on the contest before we go.

Thanks, guys!  any others want to jump in?   Lowest Pinput (input power) consumed (with a few conditions stipulated above) gets the prize.

[xee2]

Thanks, guys!  any others want to jump in?   Lowest Pinput (input power) consumed (with a few conditions stipulated above) gets the prize.

I am not in the contest. My posts were to show what a standard Joule thief circuit can do.

[JouleSeeker]

   I understand, Xee2 -- thanks for contributing significantly to the discussion.  The purpose of the "contest" was for learning and to have a little fun while doing it. 
Do you have any suggestion of why your JT draws so little power?  did you find that one element was "most" important to this effect?

BTW -- Where are Kooler and Clanzers? I sincerely hope they are all right...  I rather hoped they would contribute to the discussion also.

   Anyway, I have checked my "Xee2 replication" and found that with a 1N4148, the power usage (Pinput) INCREASES by about 40% with the 1N4148 compared to a green LED, and also would not have allowed me to SEE the change in energy usage as the voltage from the cap dropped through a critical value.  Further, a red LED I have draws more power than the green LED I have... a bit surprising, but observed.

SO -- I'm dropping the request of using a 1N4148 in place of the LED.  The LED is an active part of the DUT and the experimenter is allowed therefore to select the LED of his/her choice.  Likewise the voltage range of choice. I'm just asking that the cap/time method be used -- and I thank you fellows for using this method here to test your devices, to allow some comparisons using this method.