Dr Ronald Stiffler SEC technology

[Slider2732]

Very informative Itsu, thanks.
I've found that the several builds now of this sort of circuit have worked with 18pF to 30pF caps across the Emitter/Collector. Am not using a toroid such as you have, which I can only think of that would cause that difference.
Is there any chance it could be the permeability values of the toroid itself ?


Have done some adjusting and now am more familiar with the FFT function of the scope.
It's not to the point yet of measuring the actual frequency harmonics, but they are much more visible and their positions can be better assessed.
In this example shot, am looking at the output of a 10MHz crystal. The peaks of 10MHz,  20MHz, 40MHz and 50MHz can be seen, with 30MHz not as prominent.

Gyula - 4 pin crystal, but meaning to draw comparison to a frequency locked in by a crystal, rather than a free running circuit.

[gyulasun]

Hi Itsu,

Thanks for the video and all your efforts.

From your earlier posts I understood (maybe I misunderstood) the 40 MHz appears only when the 5 kOhm trimmer pot reduces the 12V supply voltage to the oscillator, you did not write that you set the variable cap towards its minimum value. Now you showed you detuned the toroidal tank with the 6 to 60 pF variable capacitor and I assume the 40 MHz operation occured when you detuned the tank also yesterday, right?

The explanation for the 40 MHz operation is this: When there is no resonant collector impedance at 13.5 MHz because the tank is detuned from 13.5 MHz towards the higher frequencies, then the transistor simply cannot amplify high enough at the base frequency any more but it can amplify with enough gain near the 3rd harmonic because the tank impedance has become higher somewhere in the 25-40 MHz range the 6-8 pF variable cap set its resonance, so the crystal jumped to overtone mode. This is all. We can step over this 'issue'.

I think the many harmonics appear because probably the tap is not so good for the transistor operational point when reduced supply voltage feeds the collector. For the BC337-40 probably a higher than 100 kOhm base resistor would be better, think of the base current calculation. But no need to deal with this because:

I think now if you wish to continue tinkering with this circuit, a better transistor type should be obtained.  Then it could be mounted on a heat sink if needed and will not blow from the higher peak collector voltages either. 

Thanks,  Gyula

[TinselKoala]

Very informative Itsu, thanks.
I've found that the several builds now of this sort of circuit have worked with 18pF to 30pF caps across the Emitter/Collector. Am not using a toroid such as you have, which I can only think of that would cause that difference.
Is there any chance it could be the permeability values of the toroid itself ?


Have done some adjusting and now am more familiar with the FFT function of the scope.
It's not to the point yet of measuring the actual frequency harmonics, but they are much more visible and their positions can be better assessed.
In this example shot, am looking at the output of a 10MHz crystal. The peaks of 10MHz,  20MHz, 40MHz and 50MHz can be seen, with 30MHz not as prominent.

Gyula - 4 pin crystal, but meaning to draw comparison to a frequency locked in by a crystal, rather than a free running circuit.

Excellent. Now you can use the scope's Cursors in manual mode to get the precise frequencies of the peaks in the FFT.

I'm glad people are going beyond the basics in their use of this nice little oscilloscope. The FFT function isn't particularly easy to use, it takes some fiddling to get an interpretable display, and of course it's nowhere near as capable as a stand-alone Spectrum Analyzer, but it's not too bad for a basic FFT.  And it's amazing that this function is available at all in these low-cost scopes.


Demo 1054z screen showing cursors measuring frequency and attenuation of harmonics on FFT of a 1 kHz square wave input:

[Slider2732]

TK - Am glad to see your input on this.
Yes indeed, that's the next step, to use the cursors.
One thing i've been able to do, is to notice that the better the winding of a coil, the better the main output peak.
A 13.5225MHz crystal will show the main peak at twice the amplitude of any others, with coils intentionally wound for that frequency area. Put a 10MHz coil on and it may work ok, especially with a piece of ferrite, but the harmonics away from the crystal frequency will be much stronger. 
Also, the better the winding and closer to the frequency of a crystal, the less need there is for any ferrite in that coil. Looking at the screen and adjusting turns for less and less harmonics content, I just got a coil to work the best of any on a Dollar Tree bulb, with no ferrite.
It's very useful


Update: Using the peaks reduction method for tuning on the scope, i've got a coil now that doesn't need ferrite and, in fact, ferrite diminishes output brightness of the attached Dollar Tree bulb.
Best results so far are by connecting the L3 to an AV plug, cliplead to the aluminium back.
The brightness might not be anything special, but it's a lot better than previous versions or coils on this bench here. Also no heatsink required, just a cliplead.
Am hoping to further improve with this FFT malarky....the coil is 74 turns of 30AWG on a solder tube.
Crystal is 13.5225 MHz 

[AlienGrey]

Simple Tesla coil L1 Pri 1mhry with L2 3 turns 0.4uh, with NE555 driving with very narrow pulse drive from current driver and FET at 60khz
Tesla coil is not in any kind of resonance at all. Led display is verry bright with pencil earth to garden earth.

AG