04-23-2023, 05:47 PM
(04-22-2023, 08:45 PM)tihomir Wrote: Hi Joel,
First I want to say "Thank you" for the immense amount of work you have put into these videos and explaining your experiments in text as here.
Yesterday I found your YouTube channel and went back to the back-EMF experiments I've been doing a year ago, but without any success in utilizing it. I'm a self-taught amateur and I'm still struggling with many technical issues, although I think I understand the principles. I've also found that all these have an exact relation to mechanical oscillators and work the same way, however, I still can't manage to figure out something from the back-EMF recycling.
I'll use the example with a simple low-side switching of a coil. In this video of yours you seem to be switching the coil with an NPN transistor on the high side. This is probably why your peaks get negative and you use a reversed diode. Is that correct?
In my tests I'm switching on the low side (just as Bedini and most of the NPN circuits) and the back-EMF peak is positive which is why I put the diode in a forward biased position right before the transistor (as a new branch). It's not working the other way (and it shouldn't be, in my opinion). I have the following:
(+) -------UUUUUUUU----o---(NPN)----(-)
At the "o" point between the coil and the NPN transistor I create a new branch with a diode to one leg of a capacitor.
--o--(NPN)-----(-)
|
|----------------------->|------------| capacitor | --------(?)
This is where I'm puzzled: what do I do with the other leg? If the first leg (let's call it) is the one with the peak, the other leg, at first I thought, should go to the "-". I did it that way, but nothing special happened. Yes, the capacitor gets charged, however, I can't feed it back to the input, because I don't know how.
In your explanation in another thread it seems that the "other leg" should go back to the "+". This sounds like "splitting the positive" (in the E.V. Gray motor) where we have a "+" that is "more positive" than another "+" and current can flow downhill. In this case the "other leg" of the capacitor should probably go back to the "+", the initial source? Maybe I have to put a diode between the "other leg" and the "+" so no current is fed back before the capacitor is charged?
But what do I do with the original minus when I turn the source off? Do I replace it with ground? I tested both ways with no success.
In order to test this I'm connecting a few LEDs in shunt to the capacitor using a potentiometer to limit the current draw. For a moment I thought it was working, but got disappointed when I found that the LEDs were dimming in about of 10-15 seconds.
I haven't calculated any coil or capacitor values. No resonance, just a random frequency with short duty cycle where I see back-EMF peaks on the oscilloscope . I'm using an air-core bifilar coil and a 50uF capacitor.
I had the same issue a year ago and could not resolve it. Could not find anyone to ask, because most "specialists" deny anything about the possibility of using the back-EMF for something useful, so I was on my own. Finding your YouTube channel gave me hope I could find the reason for my frustration. What is the mistake I'm making?
I hope I was somehow clear in my explanations.
Kind of difficult to understand your setup sitting from this angle. How ever if your convinced that your method generates a dc pulse. Back emf or other method. You can just isolate the output with a isolation transformer. Maybe a 10 to 10 windings to start with. Rectify the output of isolated side and send back into your dc input circuit. Best of luck!