Tesla's "COIL FOR ELECTRO-MAGNETS".

[TinselKoala]

TK:

It's amazing how you can remember stuff from 30 years ago but you can't remember stuff from last year or last week.

I can't explain your observations because of the following:  If you start with a sine wave as stimulation, then you have no where else to go.  Your device under test has to respond with a sine wave at the same frequency (but not necessarily the same amplitude and phase.)  This presumes that the device under test consists of linear components.

The only way generate new frequencies, i.e.; "You don't see the pure sinusoidal output" is if your device under test has non-linear components in it, like a nasty diode.  I am assuming that the LC tank circuit is in general pretty damn linear.  So it's a mystery to me.

MileHigh

You can get "beat note" rippling superimposed on the main sinusoidal output that happens at too long a time frame to show up on the average display of only a few cycles, and also higher frequencies that show up as a thickening of the trace line. But you are by and large correct, it's just a quibbly point.
However it's certain that using the square pulse produces the same eventual resonant frequency, and it's also true that the square pulse will often produce a measurable output when a sine wave at the same amplitude won't.

[MileHigh]

Farmhand:

FYI square waves, triangle waves, saw tooth waves, can all be broken down into their fundamental frequency and associcated harmonics.  Some waves have only odd harmonics, some even harmonics, some both.

http://en.wikipedia.org/wiki/Square_wave

MileHigh

P.S.:  The attached gif image is animated but I can't see it here, but I can see it on the Wiki entry.

[Farmhand]

Milehigh, Thanks for the tips with the Function generator, I was thinking of having a look inside, but it's handy in the I don't care too much if I damage it kinda way. I might have a look in there, I've got some temperature probes and non contact thermometers ect.

I do get what you're saying on the square waves and harmonics ect. and I agree you're right, if say you take the square and sine signals you see in the top diagram,
and fed a transformer or such like a Tesla coil with each one at the same amplitude the only difference will be the square wave input will give a higher amplitude resonant sine wave output than with the sine wave input and similarly a triangle wave will give less than a sine, just from memory. With some transformers the resulting sine wave from a square wave input can be a bit "off" shape but uniform when showing the amplitude peak, usually this happens when a core is used other than air. We can get very low frequency "beats" with two transformers almost identical but just a bit different in frequency.

So we can excite the coil with very sharp pulses and if the Q is good enough with no loading we might see only slight damping between the input events of a lower harmonic input frequency, the third harmonic is strong with a pulsed DC input, say for example my transformer is resonant at 750 kHz then if I excite it at somewhere around 250 kHz I can get a
strong and only slightly damped response at 750 kHz, that's basic stuff. I'm not a ham radio man or a full on "coiler", I'm a boilermaker experimenting as a hobby, I can't get too complicated, I don't have time.

I do very much appreciate everybody's input and I will try to respond to posts that are to me. I'll have to work back when I have time later.

Thanks all
Cheers

[Farmhand]

The optical sensor circuit I have can be used for snipping the center out of a wave ect. as well, "like coil shorting".
ie. I could run a regular pulse motor and use the sensor on an arc adjustment to load the generator coil at just the right moment.
A second circuit can work in burst mode also so it can short and open a coil several times for one trigger event.

Sensor module circuit is linked in the video description.
http://www.youtube.com/watch?v=GfkEv8oUUls

Here I test the generator as a motor, (the rotor is actually driving a universal motor shaft which it was mounted on) that is a fair bit of load,
plus as can be seen by the wave form the magnets seem to be inducing a lot in the motor coils or something. Anyway a much better and free spinning
pulse motor can be made with the optical sensor. The output of the optical sensor circuit goes to the switching circuit for the motor coils or to the loading switch.
EDIT: (this video is glitching for me at 9 seconds in, just skip to 12 seconds or something if it does that.)
http://www.youtube.com/watch?v=4mRVjbXNLBs

Or it can be used to load the resonant rise out of a gen coil maybe, which is what I would like to try some time. 
Like make the gen coils resonant then load them just enough to clip the resonant rise off the top. For an experiment.

Forgot to mention the two circuits together can output a variable pulse width for the switches even though the input PW from the sensor is constant.

I'll try to find the drawing for the switching control section.

P.S. An RPR 220 optical sensor is worth about 75 cents and a CD4001 or 4011 chip is worth about 45 cents.
The most expensive parts are usually the mosfet driver chips which are not really necessary anyway I also used a CD 4047 chip to make the bursts.
That's the drawing I'm looking for. 

...

[Farmhand]

Just to add the idea I had before, see attachment pic.
The idea is that when harmonics are present in the motor or generator core the steel inserts might be attracted to them. Maybe a red herring.

Also the pic shows what I mean by using the serial connected bifilar coil as a charging inductor, when the current is interrupted in the charging inductor it
discharges it's energy into the second capacitor which builds a higher charge to discharge through the primary on the next cycle. The built up charge in the second capacitor
depends on the energy released from the charging inductor.

Cheers