Selfrunning cold electricity circuit from Dr.Stiffler

[Loki67671]

@Dr.S
I usually work with TTL level stuff not used to high frequency high voltage certainly not used to cold electricity or scalar stuff! So a quick question about the SEC boards; is it safe to stick a 10x probe anywhere on the board or do I run the risk of zapping my scope (1Mohm 15pF <=400v)? I have a nice 100x probe on order which should definitely be safe right?

@Loki
I'm impressed with that array of 6x1W LEDs you're lighting up there. Out of interest, to get a similar luminosity, how much DC power would you have to feed them? Probably much more than 139mW I'm sure. Probably more than 139mW even if you fed them low markspace ratio PWM at high freq, SEC looks like a very interesting circuit for sure. Wish I had the parts in stock, to proto it, can't wait for Dr.S's boards to arrive!

@Yucca,
The LED show, it is cool, literally. You talked me into it and I'm going to do that and see.  Lets take a comparative snapshot here. What is a fair comparison, 5 diodes in series, seems to me would be a "close" and I use that term lightly, approximation of the AV plug??? Since I don't have a luminosity meter presently this is going to be an ugly and subjective comparative study.

The Series string, call it branch 1, will consist of:

VCC feeding,
1. 1N4148 diode
2. ASMT-MB00-NAE00 1W High Power LED, Blue
3. ASMT-MB00-NAE00 1W High Power LED, Blue
4. ASMT-MB00-NAE00 1W High Power LED, Blue
5. 1N4148 diode
Terminated at GND

At the junction of #1 cathode and #2 anode from above the next Series string, call it branch 2, will consist of the following in parallel with 2 through 5 to GND:

6. 1N4148 diode
7. ASMT-MG00-NGJ00 1W High Power LED, Green
8. ASMT-MG00-NGJ00 1W High Power LED, Green
9. ASMT-MG00-NGJ00 1W High Power LED, Green
10. 1N4148 diode
Terminated at GND

Just open the AV plugs at the cathode of the second 1N4148 and tie them to ground. I'll feed this with a DC supply and monitor power input with the same Flukes I always use. 1 reading current and the other voltage. If that description isn't clear I'll post the diagram and results tomorrow. Take a peek at the data sheets for both the 1N4148's and the High Power LED's (Avago Technologies) If I find the need to do some current limiting I will take the liberty of doing so but the diagram will accurately reflect the circuit measured. We'll know within a day or so.  If I were putting this circuit together for a job I'd be planning on more than 139mw that's for sure. 

Best regards,

Jim

The above circuit does not function like an AV plug. The first string of LED's become blinding bright! The second string barely lights. Pushing the supply any higher will destroy parts which is of course a bad thing. Tough to compare apples and oranges 

A Very subjective performance comparison, visual indication only, shows similar subjective illumination with input of 9.56vdc and 21ma. applied to a series string. No current limiting so if you push this you will kill your LED's. Everybody take a look and see why subjective interpretation is beyond difficult. Never rely on this method. Now I have to wait on the wife to go to bed so I can raid her kitchen for suitable stainless steel items......  

Best regards and back to SEC,

Jim

EDIT to clarify pictures: Loki

Top picture is the Breadboard SEC exciter running at what I posted previously, I believe 5.78vdc and 24ma from the supply, single leg current measurement.

Second picture is a strait +,Diode,LED,LED, LED,Diode,- series circuit running strait from a DC supply. I DO NOT RECOMMEND ANYONE TRY THIS WITH COMPONENTS THEY WISH TO KEEP!  Don't blame me if you let the smoke out of the parts. 

[Gustav22]

...
If I could decouple them we would all have stable self runners. No matter how long the delay line if the end is returned to ground or back to the exciter it all dies. I have used lines up to 25 chokes long and the same effect. I have missed something big that is causing me many sleepless nights. The power is there, but can not use it unless the load remains floating.....

A related train of thought::

What is fed back to the input side should be of the exact same quality as the primary signal which was used before. In other words: what is fed back to the input should be straight DC, such as delivered from the primary DC source (the battery).
From this i deduct, that the output signal should first be processed in such a way, that it will become DC, before it can be successfully fed back to the input.

As long as a fed back signal contains oscillations, these will interfere with the rhythm of self oscillation of the working system.

This would mean, that the output should be converted to straight DC, before it can be re-used as input.

How?

Maybe by feeding it into the primary side of a (small) transformer. The output of this transformer should consist of 3 secondaries, that is to say it should consist of 3 phases; same signal on each phase, each phase shifted by 120? vs. the other two phases.
Maybe a toroidal core wound with a primary and with three appropriate output windings (secondaries) might be the right setup.
Then these 3 signals should be rectified and added up. The resulting signal will be straight DC.
I am not sure how to best achieve the rectification step.
However i think, "transformation" of the output signal into 3 phases (equal but phase shifted signals) and then into DC may provide a way to close the sack ... ?h, i mean the loop.

[hartiberlin]

@Fausto,
you use about 20 Volts and about 100 mA of input current.
Then, why do you say you are using 13 Watts of input power.
Isnt this just about 2 Watts only ?
How did you hook up your battery to get it charging ?
I also want to use my coming SEC to charge selfmade NiMh batteries.
By the way, does somebody know,
where I can buy the raw materials for NiMh batteries ? I want to build BIG ones... 

[alan]

@Fausto,
you use about 20 Volts and about 100 mA of input current.
Then, why do you say you are using 13 Watts of input power.
Isnt this just about 2 Watts only ?
How did you hook up your battery to get it charging ?
I also want to use my coming SEC to charge selfmade NiMh batteries.
By the way, does somebody know,
where I can buy the raw materials for NiMh batteries ? I want to build BIG ones... 

Really like to see someone trying to charge a big cap instead of a battery using the SEC. Some have claimed OU effects using caps, though not very objective measurements were used.

[DrStiffler]

@Fausto,
you use about 20 Volts and about 100 mA of input current.
Then, why do you say you are using 13 Watts of input power.
Isnt this just about 2 Watts only ?
How did you hook up your battery to get it charging ?
I also want to use my coming SEC to charge selfmade NiMh batteries.
By the way, does somebody know,
where I can buy the raw materials for NiMh batteries ? I want to build BIG ones... 

Really like to see someone trying to charge a big cap instead of a battery using the SEC. Some have claimed OU effects using caps, though not very objective measurements were used.

@alan
I'm sure we would all be interested in your results, are you running SEC on a proto-board in your exploration of the circuit??