Fewest component oscillators

http://everycircuit.com/circuit/5201150013079552

Even simpler; http://everycircuit.com/circuit/4806960397680640

@thebugger: Sorry mate, yours is dependant on the parasitic capacitance within the transistor. There's 5pf between each pair of terminals. I have a little rule for myself: in EveryCircuit, don't make capacitors less than 100pF if they're near a transistor. I'm not all that happy with going below 1nF. http://everycircuit.com/circuit/6364919183441920

@BillyT: Yeah, I made that change too. It's just the same in operation.

If you're ok with depending on parasitic capacitance, here's one with only 4 components :) http://everycircuit.com/circuit/5052085791096832

Just increase the values. It's operating @1.5GHz right now.

http://everycircuit.com/circuit/6385513954279424

Anyway I recommend something more lime this. First an antiparallel diode at the base emitter junction will help with the Veb voltage. Transistors usually have a very low Veb breakdown voltage (sometimes less than 5V). A negative base potential would make the emitter more positive in respect to base and you'll exceed this voltage. The diode scrubs off the extra voltage. Also, I'd suggest you use a lower hfe transistor, because it is tied up in a common emitter configuration, and as you mentioned every junction here in EC has 5pF capacitance. Millers effects multiplies this capacitance multifold - you may end up with an equivalent capacitance of 500pF or even more.

What? I did "just increase the values". The oscillation died out soon after starting. Your point about transistor reverse breakdown is noted, thank you.

If the Miller effect is in play, why are they oscillating so much faster than the components' resonance? (Left 1.23x, right 1.56x.) A larger capacitor would make it slower. If anything, I think there's some opposite effect here, but with the transistor taking an input from between the resonant components and applying it across both of them, I don't think it's so straightforward.

@thebugger: 10hFe? Ouch, that's low! lol Anyway, there are many ways to set this up so the diode isn't needed. An earlier version of these had C=10nF L=1uH, and had trouble building up resonance at all. Both oscillators worked, but the signals were much smaller and never went below 0V. The reason was the low impedance of the resonant components at the oscillation frequency. I should have picked values in between those and these.

Curiously, the 10nF/1uH versions oscillated even faster: left 4.41MHz; right 4.86MHz. The only difference between left and right was the base resistors, 9k/1k like they are here.

After years of irritating statements from @thebugger about RF and which amplifier are developed most and why experienced developer do it this way while he @thebugger prefers to do it different. Statements about noise and distortion in RF circuits. Statements about why AM is only used from 70kHz to 1.5MHz and not for mobile phone communicatios in GHz or satellite communication. After years we have to listen to all these statements it came out @thebugger found the miller effect and tuned his circuits down to 10 in forward beta. After years were he is telling us how RF works. After years he is telling us how far we can communicate with FM and AM while forgot the carrier frequency makes the difference. After years he told us which amplifier are common and mainly sold. After years years and years of statements, he found the Miller effect. Here we can see an artefact with beta 1000. @thebugger and his statements. Statements over years http://everycircuit.com/circuit/5436584664236032

it's miller time

@eekee the miller effect would be superseded by the transistors hfe and the oscillations at this frequency would rise faster - but keep in mind this is an oscillator. In reality the transistor hfe would change within given boundaries due to temperature fluctuations and with this - millers capacitance will drift too (as it is directly tied to transistor hfe). With this changing capacitance you're introducing new capacitance in the circuit with which the tank circuit may couple and drift in frequency. At 2-3MHz it might be negligible, but the higher in frequency you go the more expressed this drifting will be. I suggest you try a oscillator configuration where the tank active component ia wired as an emitter follower. This way you don't have to worry about miller's effect, because it appears in common base and common emitter topologies only.

http://everycircuit.com/circuit/6750283714789376

And he just continues, stubborn as he is.

From 1000 now down to ignorant 500 even 10 is enough, funny really funny

Remember ... AM is only used at medium wave. SSB doesn't count although QAM has been buggered.

He can read what google found for him. But its get funny if google doesn't have an proper answer he just invents its own methode which can be very funny like his stereo AM modulation and side band interpretation according to Kahn-Hazeltine.

Yes. His “Kahn Hazeltine” transceiver is actually nothing like the real thing, and apparently no matter how much we try to show him why his current transceiver doesn’t work and how the Kahn-Hazeltine system actually works he just ignores us and reasserts his idea hoping that somehow he’ll manage to convince us of his idea.

Huh, I forgot to check comments on this. Anyway, I thought I wrote somewhere that these circuits exhibit something like the opposite of the Miller effect. It's not like the capacitor is even close to the right place for it, and on top of that my observation about my circuit's performance should have been enough to show there can't possibly be any semblence of a Miller effect here: "Both of them resonate at a substantially higher frequency than the resonant components alone."

I think you have more then one question. But the main question, why both oscillators are running above 1.59MHz as the LC resonace is? Thats what you like to understand 😜

You're right, it is. I never did figure it out. Today I found a related oscillator would run over 1MHz with a 500hFe BJT or under 60kHz with a 2,500hFe Darlington: http://everycircuit.com/circuit/6711608675139584