Self running coil?

[gyulasun]

Hi Gyula,

The inductance of the main coil depends on the frequency specified for the resonance to happen... So at for example 8Khz, the main coil mH is much higher then when the circuit is operating at say 20Khz.  We might say it the other way around.. The inductance mainly dictates at which frequency the circuit would work [this is in combination with the given mosfet due to the mosfets Capacitance].

The pulse coils inductance is again depending on the inductance of the main coil... I *think* the resistance of the pulse coil was rather high... some 50 Ohm or so... not sure. Maybe Luc can tell.

Hi NextGen67,

I understand and agree with what you said here. The MOSFET at such low frequencies mainly have capacitances what do influence the oscillating frequency. And these capacitances change by the pulse voltage amplitudes between its electrodes, what we have to consider in this respect is the change from zero to the peak pulse voltages of cca 10-11V. The effect of these capacitance changes may cause but a few kHz change, this rate of change also depends on the frequency whether we are at 5kHz or 20kHz.

And then there remains the change of the inductances by you to tune the frequency in a much wider range if desired, and you always have to adjust both inductances (sometimes only the one in the gate-source) for the 'correct' values to insure oscillation at a new frequency, this is quite normal and accepted.  The problem of self-starting remains to be solved at certain frequency ranges, this surely depends on the amplitude and phase conditions ruling in the circuit (just like in any oscillator circuit) when you just apply the 3V supply to start.

Thanks for the possible pulse coil DC resistance estimation, it is possibly correct if Luc uses that coil taken from a shaded pole motor: my good news is that if this coil has indeed 10-20 Ohm or higher DC resistance then the reactive power circulating in it could be increased freely by using a coil with similar inductance but wound with thicker wire so that the copper loss should reduce, the peak to peak gate-source voltage could go up to 30-35V or so peak to peak, the limit is the max +/-20V from the data sheet.  (+/-20V=40Vpp)    However if you increase the Q factor of this coil by using thicker wire, you have to consider the adjustment for self oscillation becomes more cumbersome (gate-source voltage amplitude changes quicker with the higher Q if you adjust the position of the ferrite rod).

I do not agree that the IRF640 does not use external energy for its operation, sorry. The 3.8 mW or so power must be the reactive power inside the pulse coil which must be parallel resonant with the gate-source and gate-drain capacitors.

Not sure how to interpret this? I think Luc refers to the fact that the switching energy needed for the mosfet itself, *might* be greater then what the circuit as whole is consuming.... Say at 8Khz, 3Vdc the circuit takes 0.000014A, which would be 42uW of consumption. Then *if* the mosfet would use some 3+mW or so for its *own* switching, then where is that coming from, because the circuit only uses 42uW in the first place. In other words, does the [in this case] IRF640 switched at 8Khz uses more or less then this 42uW of energy normally for it's switching. I agree however that this 42uW is coming from a battery or capacitor.

I think the key point in here is determine *exactly* what the IRF640 would theoretically consume in energy when it is being switched on and off at the 8Khz rate, with the given info, as we know the circuit energy usage as a whole.

--
NextGen67

Well, first of all, I have to correct myself on the 3+mW (I wrote 3.8mW minimum) because I used data sheet numbers for the gate to sorce and gate to drain 'Miller' charge quantity but it was connected to conditions higly different from the 3V and microamper drain current range (what is given is charges needed at 160V drain voltage and 18A drain current).  Sorry for this and I would like to correct my mistake as follows:

From Luc scope shot (Reply #345, 2 pages back) we see the gate source voltage is 7.75V peak to peak.  This is its maximum value and means all the energy swinging in that LC parallel resonant circuit is stored in the input capacitance, C_iss of the FET.  From data sheet, C_iss consists of about 1600pF gate source and about 500pF gate drain capacitances at the 3V drain voltage, together it is 2.1nF. 
Now if you calculate the stored energy in this cap (.5CV²) you get .063uWs  (microWatt second) energy. If you relate this energy to 50us (one time period for 20kHz) you get 1.26mW power, this swings in the resonant circuit in every 50us. This is called reactive power, it changes at every fraction of moment, if you try to extract it, then the effect of any load manifests as a loss in the LC circuit, consuming from this swinging energy, hence you have to take it from the battery for making up for the loss.

So eventually 1.2mW peak reactive power circulates in the gate source circuit (my earlier 3.8mW came from not relevant data sheet calcs), this controls the gate and the loss in it is supplied from the 3V by the 14 - 20uA current taken from it.

rgds, Gyula

[gotoluc]

@Luc:

Thanks for posting the data.  Isn't it based on the standard formula as shown below.  I'll work on a resonance calculator.

Here's the formula:

F_r = .159 / square root of LC

F_r = the resonant frequency in hertz
L = the inductance in henries
C = the capacitance in farads

I really do appreciate your work,  I apologize for being hard on you (it was uncalled for).

GB

That's great GB if you can figure this out and a bigger bonus if you can create the calculator program.

I must let you know that each one of those scope shots the coils are tuned to the point where the circuit uses the less amount of current. So one little change like a drop in inductance on the pulse coil and the circuit stops as at this point it is just using the tops of the resonating sine waves to switch. So if one wants to start the circuit by just taping the positive feed connection it would work if you would raise the inductance of the pulse coil by a little and then when the circuit is started it can be dropped to the most efficient point. But keep in mind that if it goes too low there is nothing that can start it as not enough energy is at the pulse coil to trigger the gate.

I think using mosfets with a smaller gate voltage would have much benefit for this and should show more energy gains since so much less energy from the mail coil will need to go to the pulse coil which will have a drop in inductance from the test data I did.

Hope this helps.

Luc

[gravityblock]

Here's a resonance calculator I found on-line, http://www3.telus.net/chemelec/Calculators/LC-Calculator.htm

Enter the total inductance of both coils, then enter the capacitance, somewhere between 900 -1000 picofarads.  The capacitance appears to be varying slightly.  It's coming really close to the actual resonance frequency.

I'll see if I can find a better solution.

GB

[gyulasun]

@Luc:

Thanks for posting the data.  Isn't it based on the standard formula as shown below.  I'll work on a resonance calculator.

Here's the formula:

F_r = .159 / square root of LC

F_r = the resonant frequency in hertz
L = the inductance in henries
C = the capacitance in farads

I really do appreciate your work,  I apologize for being hard on you (it was uncalled for).

GB

Hi GB,

It is the Thomson formula for LC resonance calculations, there are several online calculators for this, here is one I already included several of my answers to Luc or else:

http://www.whatcircuits.com/lc-resonance-frequency-calculator/

Sorry for not able to answer to you on this, hope it is not too late for you.

rgds, Gyula

[void109]

I hope this doesn't seem unrelated, but I'm working with Luc's setup and trying variations to which this applies.

I worked out a circuit to get an AC square wave pulse to drive my parallel resonant circuit for the coil, however, it ends up with:

555 timer x 1
p-mos x 1
n-mos x 5

I spent quite some time foraging through google before I came up with this solution - however it seems ungainly (a lot of components!  I'm new to electronics in general) - does anyone know of an easier way to turn a 12V source into an AC square wave?

You can view the circuit at

http://www.falstad.com/circuit/

And then import the schematic I'm attaching to this post.

Also, I appreciate all of the work that's being shared here, I know I haven't posted much, but I try to refrain unless I have something new and or original to say.  Most of what I'm seeing others are seeing and posting in more than sufficient detail to warrant a post from myself   Thanks all