Self running coil?

[gotoluc]

Hi Luc,

Would you measure the inductance of the small 1:1 transformator that is connected to the gate-source? 
You mentioned in the latest video the other side of the transformator is short circuited with that piece of wire.  This bothers me if really so because it greatly reduces the inductance of the other coil.  You could test it too with the L meter: measure on one side, the other side is open,  then short this latter other side, and read the L meter.

Thanks,  Gyula

Yes, you're right!... glad you are all here helping.

The inductance when not shorted is higher then my meter can measure (20H)

When shorted it's 10mH

I'll add a 5K pot on the normally shorted side of the modem transformer and vary it to see the effect on the self oscillation in hope to find the ideal inductance. I'll then match it with a coil of that value and see if it still works.

Thanks for bringing this to my attention

Luc

[NextGen67]

Hi NextGen67,

thanks for your above details and also the new test details you asked for which I will do this weekend.

First, the toroids inductance in test 13 is 25.3mH and the pickup coil is 28mH and 24.8 Ohm DC resistance.

I tried to add more capacitance to the 2nf tuning capacitor to further drop the current draw to which results in frequency drop but it looks like it's at its maximum since adding more the circuit stops to self oscillate

I also found that I can have the pickup coil further 2" or 5cm and I can pickup the resonance effect If I inset and tune a ferrite in the pickup coil. I also moved it around the toroid with the ferrite inside and there maybe a point close to the magnet that the pickup coil give a good output without current increasing on the toroid. I did this by hand so this still needs to be confirmed with a fixed setup.

I will build a setup that I can adjust the distance between the magnets and toroid to better test all the effects in order to find idea settings.

Luc

Ha, that explains my numbers 10.85 or 14.01 [10.85+14.01 = 24.86 !]...
About the resistance of your pickup coil. Good that I could predict it.

The toroid inductance of 25.3 mH... I guess then the C's of the mosfet were
3049pF - 2000pF (2nF) = 1049 Pf.... Which indicates that probably your mosfet is hardly
switching on... That would then be also why the mosfet shuts off when adding more Capacitance.

Yeah, when you have the build done, you will be able to have more control.

--
NextGen67

[gyulasun]

Yes, you're right!... A you are all here helping.

The inductance when not shorted is higher then my meter can measure (20H)

When shorted it's 10mH

I'll add a 5K pot on the normally shorted side of the modem transformer and vary it to see the effect on the self oscillation in hope to find the ideal inductance. I'll then match it with a coil of that value and see if it still works.

A for bringing this to my attention

Luc

Luc,  the 5K pot is also good but maybe you could test the inductance change with a separate magnet attached to very near the core of the modem transformer,  just like at the toroidal core.   Good thing is you can use the L meter for both tests separately and see the effect. 

Gyula

[gyulasun]

... I guess then the C's of the mosfet were
3049pF - 2000pF (2nF) = 1049 Pf.... Which indicates that probably your mosfet is hardly
switching on...

It cannot be that the 2SK MOSFET is not switching on,  you may wish to consider it switches on for every positive half wave and off for the negative ones. So the drain source capacitance must be some switched average value... with a value you calculated. 

Gyula

[mscoffman]

@gotoluc;

(1)
Excellent and innovative solutions. I am glad we got that out of the way as
I am convinced we were seeing groundloop problems, with no apologies to
the user with that screenname.    So it would be best to keep the signal
generator ground disconnected…We have a very sensitive analog energy
balance. And a groundloop can pump energy into the circuit derived from
the utility lines! There may be a small metal shunt on the back of the signal
generator that can be opened (for a quick AB test) the isolates power line
ground from the signal ground in the SigGen. I wouldn’t run The SigGen with
that shunt opened forever. Alternatively you can one of those small grey
plastic 2 prong plug to 3 prong outlet converters with the little green
ground pigtail left open, to do the same thing. These things are power line
ground isolation techniques.


(2)
What you may want to do is to totally stop now and re-design an oscillator
circuit patterned on the JT Joule Thief circuit with the official toroid coil
totally removed from the circuit. This would use an audio transformer *in
place of a JT toroid*. (a one to one transformer is often not a best choice
for use in an oscillator though). Once the (power) oscillator circuit was
working *then* remove a load resistor and insert the official large toroid
Drain to Vcc. Your circuit will then be self starting with frequency
determined by a cap across the transformer. I can tell this 1:1 transformer
is trying to pull the circuit frequency down to its self resonance design
frequency of 18KHz which isn’t that good. You could also wind a small JT
osc. toroid transformer for this app. I’d stick with mosfet transistors.

Alternatively the transistor in the drive oscillator could be replaced by a
small signal mosfet and use that output of that => to drive main mosfet +
coil. Either way would be a more operationally stable situation.

(3)
I think you can boost the efficiency of the pickup coil by making one like
the J. Naudin’s coil (one with the leds) around the magnet stack. That
alone may do it in terms of ou.

(4)
Use a capacitive voltage multiplier to charge the bulk capacitor from pickup
coil energy. The circuit as from a disposable flash camera driven by a three
volt battery comes to mind. This could use the main oscillator frequency to
switch a small signal mosfet to pump a capacitor and diode linked backed to
the main capacitor. Be real with this information;
http://en.wikipedia.org/wiki/Voltage_multiplier

PS: Connect the bridge rectifier, across the pickup coil unless a diode is
blown out, it will boost 59.uf capacitor charging efficiency!

Thanks and good work so far… Now let's get cleanly beyond a unity gain
situation.

:S:MarkSCoffman