UHF oscillator

A single transistor minimum component oscillator generating signals in the ultra high frequency range - a few hundreds of Megahertz.

I have measured its frequency here: http://www.instructables.com/id/Frequency-measurement-by-Lecher-Line/

Step 1 Three - dee assembly

I am assembling this to find out the limit of oscillation frequency of a transistor from my junk box. After it is completed, it shall be used as a signal source for some uhf circuits.

The frequency of the oscillator depends upon a tuned circuit - a parallel connection of an inductor and a capacitor. I shall not use a separate capacitor, depending on the capacitance of the wiring and the internal capacitance of the transistor for this function. The inductor shall be made pluggable, so that I can use successively smaller inductors until the circuit stops oscillating. The smallest inductor which still allows the circuit to work should produce the highest frequency.

For supporting the inductor, I am using a small scrap of board soldered at right angles to a larger piece, and supported by struts of connector pins soldered in. The sockets shall be pins extracted from an IC socket. The components shall be placed close together to minimise inductance so that the highest frequency may be attained.

It all starts with a circuit diagram - I have alloted about two thirds of the supply voltage as Vce and around 5 ma of collector current with a supply voltage of 12V.

M/s Freescale just provided the post-it note I used to draw this circuit. They do not endorse or recommend this circuit in any way.

Comments

[Xellers]

Can you please add some better schematics? And can you specify some of the characteristics that are required by some of the components (ie. recovery time from the diode, etc.)? I can't read you current schematics, but would very much like to build this.

[fozzy13]

I'd really like to try this but the schematic is very tough to read and follow, could you maybe redraw it in paint with typed labels??

[stevend]

I've tried to make this circuit by duplicating it as closely as possibly but testing has so far failed. My electronics knowledge is limited to what I learn as a go along. Can someone have a look at http://rimstar.org/equip/oscillator_uhf_1.htm and provide any feedback?

[neelandan]

I think your diode isn't detecting the rf. Try measuring the current drain of the oscillator, while you damp the inductor - just grasp it with your hand, short it with a thick wire, anything - and see if it changes. If the current drain is sensitive to changes around the inductor it is oscillating. Another way to detect the oscillations is to use a small TV set tuned to somewhere in the UHF. Tune the oscillator by changing the voltage or putting ferrite near the coil, and / or tune the TV set and you might be able to see the carrier as a blank place with no snow. Those manually tuned B/W portables are great for this purpose. You can even find the frequency by reading off the dial, making sure you are not tuned to a harmonic, of course.

[stevend]

There was no change to the current drain when grasping the inductor versus not grasping it.

I finally managed to get a UHF diode and when I moved the diode/probe loop close to the inductor as you do in your video, I get see no voltage on the analog meter. However, if I touch the bare wire of the diode loop to the inductor then I do see a very small voltage, around 0.1V or less, on the 1V scale. I tried with my more sensitive DMM on the mV scale and see the same sort of results.

In the meantime, I've managed to make another oscillator using a Mini-circuits POS400+ UHF chip that works just fine:
http://rimstar.org/equip/oscillator_mcpos400plus.htm
It would still be fun to get yours working though.
Thanks for your help and your great instructables.

[cylver]

The traditional way to do it (for an unmarked transistor) would be with a UHF frequency generator, a frequency counter, and your choice of a low-power wattmeter, rf voltmeter (A chance to use your RF probe!), or oscilloscope (Though if you have an oscilloscope with the required bandwidth, you could measure the oscillator directly)

[neelandan]

Right! This is my uhf generator. The test circuit for transistors will be wired up separately with sockets / pads for easy connection to the test transistor.

[theprofessor]

frequency limits are much easier gotten from s or y parameters off a datasheet. You cant use a socket because the socket has some capacitance associated with it, the same reason you couldnt just build this on a protoboard. Capacitance associated with a protoboard or socket will reduce the bandwidth of the transistor.

[Nuno]

The connector for the inductor also adds capacitance. Anyways, my point is that the fact that he'll have to solder/de-solder all of his transistors to make this test seems to remove a good part of the circuit's practicality for him. Furthermore, the type of assembly and the position of the transistor will make it very hard to swap it. On the inductor you can always have a bit more of wire and then short some windings or put them closer/farther apart. The transistor is the part that should be easily swapped, in my opinion! Or at least it should be positioned in a accessible way.

[Nuno]

Interesting.... I have a question though: shouldn't the transistor be in a socket too? Otherwise, how are you going to test your transistor pile without the help of the soldering iron?....

[VIRON]

If it can move that rf meter, it can be a nice wireless remote control.

[royalestel]

I'm very interested in this. Though it was a little bit hard to follow.