Transistor Cutoff Frequency Checker

I bought it for the meter. It was a nice little absorption wavemeter, complete with coils, and everyone at the swap was ignoring it. A couple of bucks passed hands and it followed me home. When I removed the meter I saw that this was built by a craftsman. Lots of thought, good engineering, and careful construction went into it. The three tapped coils labeled as covering 3.5 to 106 MHz were beautiful. This had to be re-purposed.

One of my faults is that I seem to collect assortments of unlabeled, house numbered, and other “mystery” transistors. They are cheap. They are unused. They look nice, with long leads, they promise me that they will make nifty projects. They just need to be identified. Toward that end I have some transistor testers including Heathkit and Eico curve tracers, and a couple of utilitarian test units to ascertain DC characteristics. But, since many of my projects involve RF, I really need to know if the device will function at an intended frequency.

The tapped coils suggested a Hartley oscillator circuit. I could substitute mystery transistors in the oscillator and see just how high they would go. Certainly not a measurement instrument, just a comparative git-er-done kind of test. Right up my alley.

I cut a piece of single sided printed circuit board to fill the hole from the missing meter. Into this I drilled one hole for an LED and one hole for a SPST spring loaded toggle switch that I had on hand.

One of the screws holding the PC board to the case also mounts the clip for the 9 volt battery. I mounted a transistor socket (remember those?) on the top of the case near the coil socket and mounted a small DPDT slide switch on the other side of that socket. That slide switch allows me to reverse supply voltages to the oscillator so that I can test PNP as well as NPN devices.The spring loaded toggle power switch applies operating current to the two-transistor LED driver and to the input to the DPDT polarity switch. No detector diodes needed, the driver and LED work just as well or better without one. I just use the indicator brightness as a comparison to a known transistor type but any brightness indicates activity at the tuned frequency. No change in operation occurs with prolonged testing so I just key the switch on for a quick test and I am done. The battery lasts for years at that rate.

Parts values are not critical. No engineering time was spent. I just grabbed parts, soldered them in, and they worked. Well, except for C3. I started out with a .01 uF but found that it loaded the oscillator above 50 MHz too much. 10 pF gave me a false peak at 100 MHz so I used a 20 pF as the final choice.

I must admit that I didn't so much as pick up a calculator for once. The hard stuff was all done for me by my talented benefactor.

For those that may wish to build something like this from scratch, the coil forms were 3⁄4 ” in diameter. Coil #1 was wound tightly with #30 AWG magnet wire 36 turns, tapped at 12 turns, sealed with coil dope. Coil #2 was a piece of 1⁄2 ” coil stock, 14 turns and tapped at 5 turns. Coil #3 was 4 turns of the same 1⁄2 ” stock tapped at 1 1⁄2 turns. These last two coils were mounted within the pluggable coil forms.

You could use just about any inductor arrangement or, for that matter, any sort of single transistor wide range tunable oscillator. There is no need for stability just as long as it oscillates reliably. If you have a hankering to test FETs then you might want to try a SPDT switch to move the end of R2 off the collector and on to chassis ground. I haven't tried it but it looks like it might work.

de ND6T

Return to Home