Super simple 220 MHz Oscillator

Wow, that’s cool. I bet the breadboard wires are acting as little antennas since the wavelength is only 1.36 meters. A quarter wave monopole antenna would only need to be 34 cm long (about a foot) to resonate and therefore emit efficiently. So even your oscilloscope probe wires could be acting as antennas thereby radiating away much of the power causing it to drop to only 400mV. In fact you could even be transmitting illegal quantities of VHF radio interference every time you run this circuit, although probably no one will notice except your local hams and even then they probably won’t be able to trace it to you through all the reflections from buildings and such, assuming the signal even propagates far enough to reach anyone (VHF signals are line of sight so they’re probably absorbed pretty well by household walls and such). If you want to keep this from happening you can just wrap the circuit in a faraday cage made from wire mesh (just make sure the holes aren’t larger than a few cm). Aluminum foil might also work but it’s pretty thin so I don’t know how well it’ll work.

Yeah it’s definitely really cool, but sometimes, believe it or not, everything gets so difficult at anything in the MHz range let alone GHz, even upper hundreds of KHz becomes a little problematic; but 1 or 10, let alone 30-300MHz and beyond is so incredibly difficult to amplify or even buffer. To put this into better perspective, even at around only 1MHz, you practically can’t use any resistors above 10K ohm. I’ve read up and experimented with it as much as possible and it’s still a bit of a mystery. Personally though, I think it’s mostly because higher resistance is so easily pulled strait to ground by even small amounts of parasitic capacitance (because at high frequency’s, capacitors act like shorts, kind of) ….But, I have also read up about the construction of resistors, and found that quite a bit of them are all made from a certain length of metal of some sort, (and it probably gets condensed into coils, or similar I wound imagine), so resistors (the higher the value worse it gets) often have pretty significant values of parasitic inductance, easily adding high amounts of reactance to the initial resistance and making it appear more like an open circuit rather than behaving like a 10k or 20-50k resistor. (Which by the way, the higher you go in frequency, the more you pretty much are forced to use lower and lower values of resistance, it gets scary difficult and confusing sometimes)! I really have no idea exactly which is the real reason to this resistor problem, but it might even be a combination of both reasons I mentioned above, or maybe even something else entirely, who knows. (Maybe ask an RF engineer I don’t know). So that’s when I started think just a little…. What if I use a transistor to replace the resistors, which is impossible to completely mimic, but transistors aren’t made with coils of metal also. It won’t be linear, but if I’m using a current source to drive a common emitters collector, then the current source should decrease its output voltage level when the amplifying transistor try’s to conduct more and more, thus is kinda like the current source is acting like a resistor that’s going up in value towards an open circuit, theoretically making even easier for the amplifying transistor to pull to ground quicker… and the revers should also be true when the transistor is conducting less, towards an open, because at that point the impedance will be pretty high, and the current source will have to increase its voltage in order to maintain the current.

As for the Frequency laws and regulations, I don’t know too much about it, because im more concerned with learning how all this RF magic works, but I did find this wonderful article written very nicely by Spark Fun if you wanna check it out: https://www.sparkfun.com/tutorials/398

Yeah high-speed circuits use special non-inductive resistors which coil the wire in alternating directions so that no net magnetic field is produced thereby eliminating inductance. Also it makes sense that there will be some significant series capacitance since the individual windings of wire are likely very close to each other making it like a bunch of little capacitors in series. This makes me wonder if regular (inductive) resistors have a resonant frequency since there are both inductive and capacitive effects.

Oh Man I’ve tried searching for such resistors so many times today, and all I can find are surface mount components for such frequencies. I did find a few that weren’t, but the datasheet doesn’t mention any frequency characteristics. I will say however, i found two surface mount resistors that mentioned a cutoff or max frequency of literally as high as 40 to 50 GHz!

Maybe I just need to purchase a large book of assorted SMD components and start going crazy with soldering a bunch of experimental prototypes to finally answer all my questions, or confirm a few suspicions at least. Only thing that makes that difficult though, is they don’t seem to have many blank SMD prototype pcb’s on the market, and I still have to worry about how the scope affects everything later on.

You could just solder a couple leads to the surface mounts and stick the leads in the breadboard like a normal resistor. Also you could try to a spectrum analysis thing by doing a frequency sweep and observing how the voltage and current and relative phase between voltage and current change as the frequency changes, but you’re scope may not have the ability to do that. If it can make non-sweeping sine waves then you can perform a sort of manual sweep by just generating a bunch of sine waves of progressively higher frequency and put all the data in a spreadsheet. Also if you do do a sweep then you should also measure what happens without any resistor at all and with a short instead of a resistor to see how much the wires themselves are doing funny things with the high frequency (and I bet the shape of the wires will matter too, so try not to disturb them much during and between the frequency sweeps).

If my theory about why resistors have inductance and capacitance is right then I think it can be modeled as a resistor in series with an inductor and then that whole thing in parallel with a capacitor. The only thing to do then is to find out the inductance and capacitance. The capacitance should be easy to find if you can provide a frequency high enough that the capacitance is much more significant than the resistance and inductance (which might even be well into the GHz range, so not very practical). The inductance can then be found by finding which frequency it resonates at. According to my model it’s just a tank resonator with a parallel inductor and capacitor where the inductor has a very high resistance, so making it actually resonate will be difficult or even impossible, but there will be a frequency where the impedance is the greatest and that’ll be at or near the resonance point.

Yeah my O-scope definitely doesn’t have any of those features, but I could probably design one, which I’d honestly probably find that to be a fun project. I’ve always wanted to do frequency sweeps to analyze interesting properties of components and certain designs, but I’ve just never had the test equipment to be able to do it. Also, I have thought about soldering component leads to SMDs, but at the same time I think that might be a waist, and maybe It would be better to finally transfer some of my ideas to a PCB anyway, I’ve been curious how much of a difference it actually makes compared to just what people say about it all the time.

I see. Well, it’ll certainly be interesting to see how much frequency you can get on a PCB. Maybe even 1GHz?

Yeah I sure hope I’ll be able to do that in real life one day! I honestly have never tried building any of these circuits on a PCB before though, so it very well could oscillate even faster, but I have long suspected that most of the magic is due to specialized transistors with ultra low parasitics. You know, I think after talking about it quite a bit, I will finally try that tomorrow or at least sometime this week and let you know what the results look like if your interested 👍

I can’t wait to hear what the results are.

Cool cool, I’ll let you know on here then

resistive losses? I dunno

Man I have been trying everything I can, and I can’t even get it to oscillate on the pcb, literally my oscilloscope just shows nothing but ridiculous noise, and I have no idea whether it’s just my oscope being weird or if my soldering job was terrible or if its something else. I hate to say it but all this randomness in my measurements just makes me wanna give up. I don’t know what to do at this point

Hmm, that’s weird. Maybe just go back to the breadboard but try soldering leads to the surface mounts so it’s sorta like PCB but minus the soldering.

Also I replied to your comment on your VHF amp

I don’t have surface mounts, yet, unfortunately

Oh, crap I’m sorry…. I just don’t go on here in EC as often as I used to, I’ll go look at that and send you a response