Sine wave resonance using square wave input

[Jdo300]

Hello All,

I have been doing a few basic experiments with coils lately and I am trying to figure out the simplest way to drive a coil at its natural resonant frequency so that I can get the Q as high as possible. My only problem is that I don't have a sine wave generator that I can use. I built a square wave pulse generator that I can vary the frequency and pulse width on and I want to use that to drive a MOSFET to pulse the coil into resonance.

My current thought is to use a MOSFET as a low-side switch and tie one end of the coil to a DC source while the other end of the coil is connected to the MOSFET drain which would connect it to ground when turned on. I'm not even sure if this is the best way to get the coil to ring on it's own but I'm also not sure what dutty cycle I should use on the square wave itself?

So let's say that the coil has a natural resonant frequency of 500 kHz, should the pulse input have a duty cycle of 25% or smaller? or 50%? I want the sine wave to come out fairly clean and with minimal distortion from the driving pulses, so I thought I would ask you all here if you have any ideas for the best way to accomplish this?

One final thing I should note is that I want to be able to feed the coil from a DC source so I can control the amplitude of the sine wave on the coil. I know there are various and sundry oscillator circuits out there that I could employ but I'm looking for the simplest solution that would allow me to use my pulse generator as the feed for the coil.

Thanks,
Jason O

 

[vk6kro]

The frequency you get out of a tuned circuit is always the same as the one you drive it with. The amplitude may be different, but the frequency will be the same.

Tuned circuits do not just oscillate at their own resonant frequency.

If you drive a tuned circuit at some low frequency with a square wave, the tuned circuit will react to harmonics in the rising or falling edge of the square wave if these fall near its resonant frequency, but this is no different to driving it at or near its resonant frequency.

If your coil has a self resonant frequency of 500 KHz you need to drive it at 500 KHz or (with a square wave) at an odd submultiple of that frequency. ie 1/3 1/5 1/7 etc. ie 166 KHz, 100 KHz, 71 KHz etc

One problem would be that putting an oscilloscope probe on the coil to observe the resonance would add capacitance and shift the resonance. You could use a times 10 probe to minimise this effect.
You could put a small winding on the large one and drive that with your square wave generator.

 

[Jdo300]

Hi vk6kro,

Thank you for the great information :-). I am interested in what you mentioned about putting a small winding on the end of the coil to kick it into resonance. Now, since I am not using a capacitor across the coil to form the tank circuit, would I just leave the coil open-ended if I was driving it in this way?

Also, to get it to work properly, should the square wave be 50% duty cycle or can I get by using a smaller duty cycle as long as I'm on a harmonic of the resonant frequency? one last thing also. Is it best to drive the coil right at its resonant frequency or on some off harmonic of it?

Thanks,
Jason O

 

[vk6kro]

Hi vk6kro,

Thank you for the great information :-). I am interested in what you mentioned about putting a small winding on the end of the coil to kick it into resonance. Now, since I am not using a capacitor across the coil to form the tank circuit, would I just leave the coil open-ended if I was driving it in this way?

Also, to get it to work properly, should the square wave be 50% duty cycle or can I get by using a smaller duty cycle as long as I'm on a harmonic of the resonant frequency? one last thing also. Is it best to drive the coil right at its resonant frequency or on some off harmonic of it?

Thanks,
Jason O

You would have to experiment a bit, but you could ground one end of both coils then drive the small one at fairly low impedance and get a step-up in impedance to the larger coil.
It is best to vary the frequency of the generator to just get best resulting voltage from the big coil.

A 1:1 square wave would be OK as it is only the fundamental that matters and it should be on the frequency of resonance of the coil under test, mainly because you can get maximum amplitude at that frequency.

You will quickly see that this is a voltage step-up device that magnifies the input voltage at the same frequency as the input. There is no sudden jumping to the resonant frequency of the tuned circuit.

Instead of the extra winding, you could have a tapping on the main coil. If there were 50 turns on the main coil, you could try putting a connection about 7 turns up from the bottom end of the coil and drive here with the square wave generator with the bottom end of the coil grounded.

Try to watch this with an oscilloscope if you can get access to one. Should be spectacular.

 

[Jdo300]

Thanks for the idea! I never thought about driving it in that way. Sounds like an auto transformer configuration. :). I'll give it a shot and let you know how it works out.

Thanks!
Jason O