Self running coil?

[wattsup]

@gotoluc

Saw your last video (10) with thanks. You sounded kind of pessimistic but don't let that get you down cause you just looped your first one so CONGRATULATIONS are in order. That one step is major.

I am not convinced you have the right or best frequency to look at this.

First of all, don't get all bogged down about the mosfet leakage. It is less then peanuts and not worth looking at. If guys want to know, just put a bulb or led on the mosfet drain and source and pulse the gate. If that gate has any meaningful voltage with enough north/south potential being leaked, that led will light up, but it will not. lol

If you have time, please just try this.

1) Put your 9 volt battery back on the 555 circuit.
2) Put a small bulb on the pick up coil. Remove the pickup coil from the stand it is on so you can move it around the toroid.
3) Pulse the toroid and move the pick up coil and find both the best frequency and position where the pick up coil lights the light brightest.
4) If you have the patience, try removing the magnets to see any difference. Then when you find the brightest spot, bring down the applied voltage to the toroid to see how much you really needed to keep that pickup coil bulb lit. Then play with the duty cycle to find the same. Sometimes lowering the voltage by 40% will reduce the brightness by 5% so this becomes a good deal to not only look for the brightest but most frugal settings.

Advise the result.

Now if you add another pickup coil, you may have to find the new frequency or frequencies that will light both bulbs brightest, etc. In my videos, I had found frequencies when using solo coils, but had to find new frequencies when several coils where on the toroid. You can load that toroid to the hilt.

Oh, I tried my bucking coil with your series jump connection and it does give some nice effects. I will learn more first then give some results. I had prepared a much longer post but will keep this one short instead.

wattsup

[NextGen67]

Hi Gyula,

yes, amazing specs!... I read that post after I posted.  I'll look into it and see if I can buy a few.

It was a surprise to me to see the current increase and decease connecting and disconnecting the load to the second pickup coil. It did not do this previously when the loop was not closed.

Do you think if the pickup coils were tuned it would have less of a coupling effect with the primary coil?

Seeing this test 10 video do you think it is worth continuing?

Thanks

Luc

@Luc,

YES! it is definitely worth continuing.

A few things happen in your video #10 that makes it rather logical why you have amps draw now [tough if you do your comparison test, you will find a *very* tiny difference].

What happened:

1) although you are at a 90 degree angle with the resonance coil, your pick-up coil *actually* drains energy from that toroid coil, because it is way to close to the toroid coil... Since it has 5 layers [instead of just 1], you are 'stealing' energy from it

2) the mosfet is triggered with not enough energy to *properly* operate... Yes, the coil is in resonance, but you earlier found gain effect is very very limited now. See the data sheet of your FET and notice *how* high Ciss (and the other 2) are BELOW the 2.5 volt [that is why 5v and up would be better] .... So even with *optimal* tuning, the resonance will not be as good as you had before.

3) 5 volt and upwards is better to see for PtP [this is even true for the fet you might want to buy next].

So all with all, test #10 gives wrongly data.

What you *could* do, and will *definitely* show the differences  with your earlier scope shots you put on the board here is:

1) Just feed the 555 with external power for a while (battery), and remove your pick-up coil. [Make sure you have at least a PtP as what you had before what was 7.81 V if I recollect well].

2) Tune your circuit into best possible way again.

3) Save 2 scope shots of it (like the same as which you posted here before).

3) Now lower the PtP to the 3.81 volt [as what you had in your video#10].

4)  Tune your circuit into best possible way again.

5) Save 2 scope shots of it  (like the same as which you posted here before).

6) Post the 4 scope shots again.

What you might see your when you do these tests, would probably be that your 7.81 and 3.81 images give [firstly] quite similar pictures, but if you look closely, especially that *bump* you mentioned earlier before and the horizontal line of the OFF phase, you probably see difference there.

Secondly, when you would do the 'energy going up in Cap' again (where it wend up from 16.64 volt to 16.80 volt [as shown in video#7], you would [and *should*] have the same or better result [with the 7.81 PtP]... Try that again.. you should get a raise from 16.64 to 16.80 volt over *exactly* 20 seconds (well might be 20.5 seconds).

When you repeat the same 'energy going up in Cap'  again but now with the 3.81 PtP you will *most likely* notice that [in comparable way !], the energy going in the cap has a much lesser charge rate.... Actually I wonder if you with a PtP of 3.81 volt still can get a decent 'charge back' at all.

Don't for a second think of giving up yet :-)

When you put up the 4 scope shots again [as i mentioned above], at least we all can see what is the differences between them, and the difference with the one you posted here already.

EDIT: What I try to establish here is to show that [by using the 555 now], we are *still* able to get the same scope shots as you posted before here. Because if those scope shots [especially the 7.81 PtP one] shows *less* performance, that should be adressed firstly.

--
NextGen67

[gyulasun]

Are we talking about using an audio transformer to isolate the sig in to the fet gate?

I see the transformer as a pump of sorts and power on the input actually transfers power to the other side. Its a power in power out kinda thing. ;]

The optical is the best way. Imagine, we have optical sending internet, so freq is not an issue, and quality of signal conversion should be good also.
I have to say that all that is needed is a mosfet driver that is triggered by the optical device. And if it is 50% duty and a square wave, it should work perfectly.

I also imagine the use of a zener to keep the voltage on the charge cap leveled out at some point.

Mags

Hi Mags,

Sorry but MarkSCoffman did not mean using the audio transformer to isolate the sig to the FET gate.   He proposed a possible impedance matching with the step down audio transformer to utilize the output power from the pick up coil in the best way for feeding the CMOS 555 with supply voltage.

Basically I agree with his proposal, it is a kind of impedance matching anyway.  The pickup A is made parallel resonant with a capacitor at the output frequency and the audio transformer steps the (hopefully) 30-40V peak to peak resonant voltage down to a 8-10V at its secondary output that would be full wave rectified for feeding the 555 pulse gen that drives the MOSFET gate.

Of course if the pick up coil would have some taps in its winding, say at 1/4 and at 1/2 of its full number of turns, then the transformation could be done by the pickup coil itself, at the tuned resonance freq.

rgds, Gyula

[gotoluc]

Gyula,

Would it not be possible to simply attach only the signal generator to the mosfet and then connect the Cap to the source and drain, and just observe what goes into the cap ?

Actually a [known] voltage/amperage could be supplied to get it at the correct working point, but this can with a calculation be subtracted from the cap value later again?

EDIT: No need to attach battery I guess, since the signal generator can easily deliver enough voltage ?  Anyhow, after say 5 minutes of charge, one could see how much the cap collected.

--
NextGen67

Hi NextGen67,

I have done the test you asked for in the above post to determine how much is leaking through the gate. More tests are also coming

Supply Battery connected to capacitor bank to first tune the circuit is 12.88vdc

Setup is using signal generator and feeding 3 IRF640 connected in parallel

Tuned for maximum returned current -.000224 @ 38KHz  (with no pickup coil)

Once signal generator is set there is no changes to it as you can see in video.

Link to video demo: http://www.youtube.com/watch?v=MX2nZSOhxfc

1st Shot is Scope probe across Generator input, the other probe across a 100 Ohm resistor in series with Generator ground and coil tuned to send back most energy.

2nd Shot is Scope probe across Generator input, the other probe across a 100 Ohm resistor in series with Generator ground with source & drain to cap through coil

3rd Shot is Scope probe across Generator input, the other probe across a 100 Ohm resistor in series with Generator ground with source & drain to cap no coil

[skywatcher]

I think i can give an explanation for the raising voltage in the cap. 

The tuning with the magnet has to bring the core material *close to* saturation. That's the reason why the magnet must not be too strong, or too close to the core. If the magnet already saturates or over-saturates the core it doesn't work. The magnet drives the core into a non-linear state so that it acts like a rectifier.

Here are more details: http://www.linux-host.org/energy/srect.htm

The pulses going into the gate of the MOSFET are leaving the MOSFET on the other side as AC voltage, because of the gate capacitance. This AC voltage may be rectified by the 'magnetic rectifier' and will increase the cap voltage, if all the other losses are low enough.

Of course, this is no 'free energy'.
It's a well-known effect which was discovered more than 120 years ago...   

(maybe it would be a good idea to read some old books... some things have already been forgotten)