Selfrunning cold electricity circuit from Dr.Stiffler

[k4zep]

Hi Dr. Stiffler,

Circuit very resonant around base frequency of 5.8 MHz, tune for max lighting/which shows up as max differential across resistor.  Bandwidth is a few hundred Khz wide.  We are looking at the pure IR drop across resistor @ RF frequency.  Further down in circuit waveforms identical to yours as shown in your scope shots.  Much higher peak to peak voltages after L/C circuit before and feeding BF open ended coil.  Can show scope shots if you need.

Ben

Hi Dr. Stiffler,

Took me a while to find a way to make viable measurement with LED as indicator/diode in front end and get rid of capacatance effects and to think through exactly what is going on voltage and duty cycle wise in the circuit..

Found with common ground of signal generator/scope, all floating allowed good measurement.

Circuit is identical to basic circuit with L/C on input except a resistor added directly to sig. generator input then a diode connected to L/C circuit all in series..  Thats is all.  See abbreviated schematic attached......quick hand draw

Resistor 150 ohm carbon non inductive.
Signal:  5.8MHz square wave.
Signal input TP-A:  27 V Peak to Peak square wave +/- around ground.
Signal after Resistor TP-B = 23 V Peak to Peak.
Difference across resistor 4V
Actual difference seen from load due to diode 2V
I=E/R=2/150=.0133 MA.
P=EI=2X.0133=26.6 MW/2 for duty cycle=13.3 MW input.

LED Output" 18V across LED's with 10uf also across them for filtering.
Series Current with DVM in series with LED's= 1.8 Ma. (lousy coil!)
P=EI=18X.0018
Power out=32.4 MW.

I'm pretty confident of these readings.  It only gets better with more LED's on the output.  This thing seems a constant current source till it runs out of voltage..........

Your knowledge of how this thing works I Will not question as to bridge mucking up output.  I haven't tested that..............but I suspect you are right.  There is some funky phasing due to the AV plug working both ways and the virtual ground.

Ben

@k4zep

I may stand corrected, but connecting both of the secondaries to a bridge should remove all chance of any amplification.

You just may have hit on something I have not thought of, with your LED in the input, try two with a 1 ohm in between. Try measuring across the resistor as the impedance is lower this way. I did a similar approach with 20 LEDS split with resistor in between. That allowed some confidence in the measurement. Would be interested in the result. If your scope is grounded it of course will not work. An associate used an ac isolation xformer and did a fair job on readings, buy there is still the capacity. Anyway take the worst case, because you know it is really better.

I put the model number of my solar cells on my site. Beware they are expensive as they are commercial.

Do not be afraid to ask, this is why I am here and what I am asking you all to do, that said lets look at what you have.

If the generator will not drive the LEDS direct and you have no inductance, one thing comes to mind. Place 4.7K ohm resistor from your generator to you ground connection, exclusive of the LEDS. In other words the generator Hot lead to one end of the resistor and the other end of the resistor to your ground connection. With a VOM or DVM or DMM measure the voltage across the resistor. Measure both A.C. and D.C. Don't worry about the meters frequency response, I expect to see a high A.C. voltage across the resistor. Even though you tried the inverter arrangement I suspect low level A.C. versus the generator signal.

How all of what I just said is rubbish if you can adjust the frequency to either side and the LEDS dim or go out. Should this be the case I might try a different outlet strip as the one you are using contains harmonic suppression inductors or varistors which could be bleeding into the A.C. hot side.

[amigo]

I cannot say how happy I am to see all this activity happening and the good doc back as well.

@Dr.Stiffler,

does your site server support php/mysql because you could run a CMS (Content Management System) on it and greatly ease the way you update your pages? I could help you out with the setup if you need assistance...

@Stefan,

please post your oscillator circuit here so that others may use it, if it proves to produce necessary drive. We really want everyone to be able to easily replicate all the findings so that naysayers have nothing to say.

I have built a new simple square wave oscillator (another hack job) by using a combination of two NAND (from 74HC00) and couple of hex-inverter schmitt triggers (from 74HC14) as the output boost. I find it working better in this configuration than flip-flop on its own or the hex-inverters on their own. I'm attaching a mock-up schematics...with this values I get around 10MHz, use your own judgment on filtering the DC power signal, I'm pretty bad at it. Matter a fact if someone can improve on this would be greatly appreciated.

Has someone tried that LTC based oscillator that was posted here couple of pages back yet?

Lastly could someone please tell me how to (properly) wind a 1.1uH L1 because I have not had much luck doing it so far with various online air coil calculators? I find it that using a commercial choke (of unknown inductance value) I had, makes the circuit work, while my own coils don't, grrrr.

Thanks.

[DrStiffler]

Hi Dr. Stiffler,

Circuit very resonant around base frequency of 5.8 MHz, tune for max lighting/which shows up as max differential across resistor.  Bandwidth is a few hundred Khz wide.  We are looking at the pure IR drop across resistor @ RF frequency.  Further down in circuit waveforms identical to yours as shown in your scope shots.  Much higher peak to peak voltages after L/C circuit before and feeding BF open ended coil.  Can show scope shots if you need.

Ben

Hi Dr. Stiffler,

Took me a while to find a way to make viable measurement with LED as indicator/diode in front end and get rid of capacatance effects and to think through exactly what is going on voltage and duty cycle wise in the circuit..

Found with common ground of signal generator/scope, all floating allowed good measurement.

Circuit is identical to basic circuit with L/C on input except a resistor added directly to sig. generator input then a diode connected to L/C circuit all in series..  Thats is all.  See abbreviated schematic attached......quick hand draw

Resistor 150 ohm carbon non inductive.
Signal:  5.8MHz square wave.
Signal input TP-A:  27 V Peak to Peak square wave +/- around ground.
Signal after Resistor TP-B = 23 V Peak to Peak.
Difference across resistor 4V
Actual difference seen from load due to diode 2V
I=E/R=2/150=.0133 MA.
P=EI=2X.0133=26.6 MW/2 for duty cycle=13.3 MW input.

LED Output" 18V across LED's with 10uf also across them for filtering.
Series Current with DVM in series with LED's= 1.8 Ma. (lousy coil!)
P=EI=18X.0018
Power out=32.4 MW.

I'm pretty confident of these readings.  It only gets better with more LED's on the output.  This thing seems a constant current source till it runs out of voltage..........

Your knowledge of how this thing works I Will not question as to bridge mucking up output.  I haven't tested that..............but I suspect you are right.  There is some funky phasing due to the AV plug working both ways and the virtual ground.

Ben

@k4zep

I may stand corrected, but connecting both of the secondaries to a bridge should remove all chance of any amplification.

You just may have hit on something I have not thought of, with your LED in the input, try two with a 1 ohm in between. Try measuring across the resistor as the impedance is lower this way. I did a similar approach with 20 LEDS split with resistor in between. That allowed some confidence in the measurement. Would be interested in the result. If your scope is grounded it of course will not work. An associate used an ac isolation xformer and did a fair job on readings, buy there is still the capacity. Anyway take the worst case, because you know it is really better.

I put the model number of my solar cells on my site. Beware they are expensive as they are commercial.

Do not be afraid to ask, this is why I am here and what I am asking you all to do, that said lets look at what you have.

If the generator will not drive the LEDS direct and you have no inductance, one thing comes to mind. Place 4.7K ohm resistor from your generator to you ground connection, exclusive of the LEDS. In other words the generator Hot lead to one end of the resistor and the other end of the resistor to your ground connection. With a VOM or DVM or DMM measure the voltage across the resistor. Measure both A.C. and D.C. Don't worry about the meters frequency response, I expect to see a high A.C. voltage across the resistor. Even though you tried the inverter arrangement I suspect low level A.C. versus the generator signal.

How all of what I just said is rubbish if you can adjust the frequency to either side and the LEDS dim or go out. Should this be the case I might try a different outlet strip as the one you are using contains harmonic suppression inductors or varistors which could be bleeding into the A.C. hot side.

You have no idea how much I appreciate your work. Great professional job.
When you feel comfortable with what you have done and your understanding which is very, very close to being right on, then you need a decent coil, because the next phase is to move from milli-watts to watts. Just imagine how it will feel to see your calculations showing watts 'OU' ?  But, are not the milli-watts a thrill?

The next phase is driving the incandescent. Then the cream on the cake.... Boy am I happy for you and my selfish self.

[DrStiffler]

I cannot say how happy I am to see all this activity happening and the good doc back as well.

@Dr.Stiffler,

does your site server support php/mysql because you could run a CMS (Content Management System) on it and greatly ease the way you update your pages? I could help you out with the setup if you need assistance...

@Stefan,

please post your oscillator circuit here so that others may use it, if it proves to produce necessary drive. We really want everyone to be able to easily replicate all the findings so that naysayers have nothing to say.

I have built a new simple square wave oscillator (another hack job) by using a combination of two NAND (from 74HC00) and couple of hex-inverter schmitt triggers (from 74HC14) as the output boost. I find it working better in this configuration than flip-flop on its own or the hex-inverters on their own. I'm attaching a mock-up schematics...with this values I get around 10MHz, use your own judgment on filtering the DC power signal, I'm pretty bad at it. Matter a fact if someone can improve on this would be greatly appreciated.

Has someone tried that LTC based oscillator that was posted here couple of pages back yet?

Lastly could someone please tell me how to (properly) wind a 1.1uH L1 because I have not had much luck doing it so far with various online air coil calculators? I find it that using a commercial choke (of unknown inductance value) I had, makes the circuit work, while my own coils don't, grrrr.

Thanks.

One note that I must make for you, your IC drivers are basically current drivers and the output impedance my not be good enough. I tried using SN7400 and no cigar. If you drive from the drain with an inductor by MOSFET may be better. Keep it in mind in case you get poor response.

[amigo]

One note that I must make for you, your IC drivers are basically current drivers and the output impedance my not be good enough. I tried using SN7400 and no cigar. If you drive from the drain with an inductor by MOSFET may be better. Keep it in mind in case you get poor response.

I notice that I can connect only so many LEDs until the light gets dimmer than I'd like it to be. Perhaps a combination of the 74HC00/74HC14 for the signal generation and 2N7000 output buffer might work better?

Thanks