Bifilar pancake coil overunity experiment

[TinselKoala]

Here are a few others. I haven't yet found a datasheet with a different pinout than ECB.


http://www.unisonic.com.tw/datasheet/2SC945.pdf
http://cdn2.boxtec.ch/pub/diverse/C945.pdf
http://www.elenota.pl/datasheet-pdf/163457/N_A/BC945?sid=a7267847d499df42fd11eee01ca6b6a7     (searched as BC945)

http://www.ges.cz/sheets/2/2sc945.pdf

[ayeaye]

Sorry, something was wrong with the svg file in the simple format, it should be the following.

<svg version="1.1"
xmlns:xlink="http://www.w3.org/1999/xlink"
xmlns="http://www.w3.org/2000/svg">
<path d="m 100,682.36218 80,90 80,50"
    style="fill:none;stroke:#000000;stroke-width:1"/>
<path d="m 260,1022.3622 70,-90.00002 60,-40 80,-20 120,-10 210,-10"
    style="fill:none;stroke:#000000;stroke-width:1"/>
<path d="m 100,1162.3622 100,-20 60,0"
    style="fill:none;stroke:#000000;stroke-width:1"/>
<path d="m 60,852.36218 790,0"
    style="fill:none;stroke:#000000;stroke-width:1"/>
<path d="m 60,1352.3622 790,0"
    style="fill:none;stroke:#000000;stroke-width:1"/>
<rect width="10" height="10" x="60" y="600"/>
<rect width="10" height="10" x="60" y="1400"/>
</svg>

So how to do it, use the Python scripts that i posted last in this thread. The content of the svg file in the simple format above, is for testing purposes only. In the real life, draw your traces with Inkscape, save the drawing in the normal svg format, somewhere where you can find it in the file manager. Then open that file with wordpad, and copy all the text from there, instead of copying it from here.

IMPORTANT When copying code from here, hold down the mouse button and slide the mouse down. But after going over the last line, don't go outside the quote box, before pressing ctrl-c. Then there will be no additional spaces before each line.

Open ideone at the following link.

https://ideone.com

Select Python, always before pasting, delete the first line in the source code box, what it has there, "# your code goes here", delete it.

Copy and paste the Python script that extracts data from Inkscape, to the source code box, and the content of the svg file here above, to the stdin box. Click on Run, what you should see is the following.

https://ideone.com/dhjcFA

Open the wordpad with no text in it, copy and paste the content of stdout from there, into wordpad.

Go back to the ideone main page (go back one page in the browser) and paste the Python script that does the power calculation, to the source code box. Change the values of the units in the code to the ones that you had, these should be the oscilloscope scales divided by 1000, in ns and mV, if they were any different. Change the resistor values there to the real ones that you had, in ohms. Change the frequency there to the real frequency used, in Hz.

Copy and paste the text that you copied into wordpad, to the stdin box. Click on Run and what you should see is the following, with the power in and power out values in stdout.

https://ideone.com/1tPuBc

I'm sorry for writing it so long, but this is important. I did everything above myself to check it, and it worked for me. Please say when something would not go as it should.

Ahh, install Inkscape, from the following link, for your os, what it is, windows, mac or linux.

https://inkscape.org/en/release/0.92.3

I posted here an online vector editor too, if you don't want to install anything in your computer. It has svg format, so should work the same. But Inkscape is much better for the purpose. You might like it, as it is also very good for any technical drawing.

[TinselKoala]

Seems like a lot of trouble. In the old days we would simply trace the traces onto tracing paper, then cut them out with scissors and weigh the paper pieces on an analytical balance.

Something that is not going as it should is the transistor pinout.


Do you have a fake transistor? Or a bad transistor? Or something strange about the connection? Or do the datasheets apply to the wrong transistor?  I know in JTs one can often get NPN transistors to work "backwards" that is with C and E reversed. But here we are dealing with B being on the right (as all the data sheets I've found say it is) or in the middle as with your transistor (apparently).



Can you post a legible photo of the transistor you are using?

[AlienGrey]

Seems like a lot of trouble. In the old days we would simply trace the traces onto tracing paper, then cut them out with scissors and weigh the paper pieces on an analytical balance.

Something that is not going as it should is the transistor pinout.


Do you have a fake transistor? Or a bad transistor? Or something strange about the connection? Or do the datasheets apply to the wrong transistor?  I know in JTs one can often get NPN transistors to work "backwards" that is with C and E reversed. But here we are dealing with B being on the right (as all the data sheets I've found say it is) or in the middle as with your transistor (apparently).



Can you post a legible photo of the transistor you are using?

Interesting when I was a school kid I first used a china graph pen cell from the art shop then a deacon dado pen came out and then some bright spark invented transfers
but since the  mid 80s it's been computers and paper tasers or special ink.

[ayeaye]

Seems like a lot of trouble. In the old days we would simply trace the traces onto tracing paper, then cut them out with scissors and weigh the paper pieces on an analytical balance.

In these days they could do that, today they use computer. This is a very imprecise method as well, as the weight of the paper is never exactly even, very difficult to precisely weigh it as well.

Then you likely had to make the oscilloscope to show the multiplication of two traces, then find the area to calculate power. In this case this is more complicated, the oscilloscope has to show two multiplications of two voltage values, and then they have to be divided by different resistor values, one has to be subtracted from the other, and only in the first (positive) part of the cycle. I don't know whether any oscilloscope can do that. If it cannot and it is a digital oscilloscope, waveforms data has to be downloaded from it, and still some code is necessary to calculate from this waveform data.

Maybe by using two analog multipliers, then the oscilloscope will show the output of both, and calculate the area of them both separately during the positive part. If the oscilloscope can calculate the areas of two traces. Maybe it can be done, though i wouldn't say it's simple. And today they rather make the electronics simpler by using software than doing the opposite.

But if you can figure out some simpler method that people with digital oscilloscopes can use, that would be great of course. I think you now know what should be done. Me with my analog oscilloscope have no really better way.

Ok, calculate the area of Vr2 * Vr3 during the first part in one measurement, then calculate the area of Vr3 * Vr3 during the first part in the second (separate) measurement, can it be done, like with Rigol? I don't exactly know what the digital oscilloscopes can do, as i don't have one.

Do you have a fake transistor? Or a bad transistor? Or something strange about the connection? Or do the datasheets apply to the wrong transistor?  I know in JTs one can often get NPN transistors to work "backwards" that is with C and E reversed. But here we are dealing with B being on the right (as all the data sheets I've found say it is) or in the middle as with your transistor (apparently).

I have a c945 transistor and the pinout appeared to be as i did show, and indeed i couldn't find any datasheet that did show c945 with that pinout, all were with the pinout that you referred to. I found the pinout by testing the transistor with a multimeter, and no, i didn't put C and E backwards, because the voltage drop on the emitter is greater than the voltage drop on collector. The transistor worked well too, as much as i saw.

Can you post a legible photo of the transistor you are using?

Yes i can, the photo is on the figure below. This was rather difficult thing to do though, i almost had to hold the flashlight between my teeth.