Self running coil?

[gotoluc]

Hi all,

thanks for all your input as to why the differences between the 2 coils. I will make another video when I have time that will have a coil of exact resistance and inductance and we can continue from there.

Here is an update on the self pulsing coil at this time.

Using a IRF640 it is self pulsing at 20KHz with 3vdc input at 14uA + or - 0.5uA

Channel 1 (green) is across the drain and source and Channel 2 (yellow) is across the gate and source. Both probe grounds are connected to source.

Take note of the beautiful Sine Wave the pulse coil is now making to trigger the mosfet gate. I think this is a new and ideal way of triggering a MOSFET.

@Gyula, could you please look up the specs of the IRF640 and do a calculation of an approximate wattage needed to keep its gate triggered at 20KHz. If anyone else feels up to the task please feel free to do so.

Thanks all for sharing

Luc

[gyulasun]

Hi all,

thanks for all your input as to why the differences between the 2 coils. I will make another video when I have time that will have a coil of exact resistance and inductance and we can continue from there.

Here is an update on the self pulsing coil at this time.

Using a IRF640 it is self pulsing at 20KHz with 3vdc input at 14uA + or - 0.5uA

Channel 1 (green) is across the drain and source and Channel 2 (yellow) is across the gate and source. Both probe grounds are connected to source.

Take note of the beautiful Sine Wave the pulse coil is now making to trigger the mosfet gate. I think this is a new and ideal way of triggering a MOSFET.

@Gyula, could you please look up the specs of the IRF640 and do a calculation of an approximate wattage needed to keep its gate triggered at 20KHz. If anyone else feels up to the task please feel free to do so.

Thanks all for A.

Luc

Hi Luc,

You can approach input power to the gate-source by considering the charge in the gate source + gate drain interelectrode capacitances first, then using  the data sheet for the IRF 640 and this Application guide
http://focus.ti.com/lit/ml/slup169/slup169.pdf 

I received about 3.8 mW going into the gate source at 20kHz.  (This depends on how I estimate the capacitances of the MOSFET from the data sheet, worst case I got 4.22mW and best case 3.67mW.)

I assume you used the toroidal coil, not the big air core coil, would you measure its inductance as it was when you took the scope shots, just in case later might be useful. Also wonder if you used capacitor in parallel with it? 
The pulse coil is at the gate source? and also tuned? maybe with the gate source cap?

One more thing: Does the table contain metal surfaces or is there any metal structure under it where the big air coil was laid?  Because such things may ruin the quality factor too much when close by.  Also, try to back it up at least 20-30cm in the air, far from anything to see if the circuit parameters change.

rgds,  Gyula

[gotoluc]

Hi Luc,

You can approach input power to the gate-source by considering the charge in the gate source + gate drain interelectrode capacitances first, then using  the data sheet for the IRF 640 and this Application guide
http://focus.ti.com/lit/ml/slup169/slup169.pdf 

I received about 3.8 mW going into the gate source at 20kHz.  (This depends on how I estimate the capacitances of the MOSFET from the data sheet, worst case I got 4.22mW and best case 3.67mW.)

I assume you used the toroidal coil, not the big air core coil, would you measure its inductance as it was when you took the scope shots, just in case later might be useful. Also wonder if you used capacitor in parallel with it? 
The pulse coil is at the gate source? and also tuned? maybe with the gate source cap?

One more thing: Does the table contain metal surfaces or is there any metal structure under it where the big air coil was laid?  Because such things may ruin the quality factor too much when close by.  Also, try to back it up at least 20-30cm in the air, far from anything to see if the circuit parameters change.

rgds,  Gyula

Hi Gyula,

thanks for taking the time to calculate the approximate wattage the IRF640 would needs to switch the gate at 20KHz

So in my setup of the scope shot above the IRF640 is using no external energy since it's self pulsing by using some of the energy from the coil.

It is doing this at 20KHz with a 3vdc input and using 14uA = 42uW

From your calculations you come up with 3.67mW at best

My question would be now, where does the 3.6mW needed to switch the mosfet come from

No capacitors used or needed. Pulse coil is between gate and source. It appears to me that the MOSFET's capacitance maybe what makes this work. I believe (not fact) the other part would be the coil geometry and maybe the magnet.

No, my table has no metal under it. It's made of particle board (wood sawdust). I know for a fact that adding 2 Ohms to the Brooks coil will not make it resonate. If you want to see a video demonstrating this I can do it for you no problem my friend

Luc

[wattsup]

@gotoluc

Always good stuff. I don't have the answer to your question but it must have something to do with the greater copper mass of the larger coil. You may have close to the same inductance as the small toroid but the pulse still has to saturate the mass of the bigger one. I don't really know enough about it.

Regarding the magnet on your coil, here is a video that could help explain some of the effect. I put this on Thanes' thread but it has some relation to what you are doing also. Basically, when he adds the magnet to his toroid, you can see on the scope that the pulses go faster probably because the magnet is pre-saturating the core so the impulse wastes less time saturating the core per cycle.

http://www.youtube.com/watch?v=g42hzVKIa0I&feature=channel

[mscoffman]

@All

What I think I would do is try to duplicate this set-up
with a 2" toroid. But with maybe some secondary windings
to step-up the required gate voltage to 8 volts making
the oscillator self starting. Then try to get both toriod
coils into resonance tuned with magnets rather than
capacitors.

If one could then build an identical set-up using p_channel
mosfets. One could consider doing bipolar coil driving from an
H-bridge final. Diodes would direct the resonating gate voltage
into N-channel circuit or P-channel circuit depending on it's
polarity. The N-channel circuit would drive the coil polarity
one way then the P_channel drive it the other. Clean!

I don't like the idea of open air core coil designs because
powerful oscillating magnetic fields go all over.

:S:MarkSCoffman