Getting By with Less than Ideal Test Equipment

or, how to calibrate radios without spending a fortune


What follows are tricks that I have learned and developed while trying to build a 6m CW rig (and subsequent projects).

My Oscilloscope doesn't quite go fast enough to see the waveform.

This is my favorite. I just LOVE this trick. There I was, having borrowed an old but nice oscilloscope (Sencore PS 163, 1 Mhz maximum sweep, can make the trace 5x wider than the screen, but suffer distortion accordingly...) from my father, KC9ASJ. He wished me luck trying to watch anything fast, and said that I'd probably just see a band of waves running across the screen, and wouldn't be able to discern a waveform. And, he was right. However, I realized that I could set up Channel A to look at the waveform, and Channel B to look at the waveform after it had been run through a 4020 14 stage Binary Ripple Counter. Sure enough, looking at that showed me a beautiful square wave - not what I wanted to see. But then I started randomly pushing buttons on the scope, and the waveform jumped out. I had just set it to trigger on a slower waveform that happened to be in sync with the waveform I was interested in (just maybe 32 or 64 times slower). A little more twiddling on the trigger settings, and I was happy. There was a sine wave, changing in frequency as I twiddled the pot on the circuit! The downside at the time was that it was only going about 22 Mhz... Still, that's not bad at all for a 1 Mhz scope to show me.

More experimenting and higher grade equipment (A Tektronix 549, which has some pretty spiffy features) has shown me what people with "real" equipment do: they turn on the storage display and set the scope for single sweep. The Sencore, while significantly better than nothing, doesn't have these features. And, the Tektronix, while significantly more expensive, doesn't have a faster sweep, which is what would REALLY be helpful.

Oscilloscope? All I've got is a computer!

oscope is a great little program. I'm running it on a Pentium 133 with 8 megs of ram that somebody gave me. Configured the Linux kernel to recognize the sound card, and voila - a 44100 samples per second digital oscilloscope. Of course, to have any real idea of what a wave looks like, it has to be less than around 11khz... BUT when you combine this scope program (which happens to include a frequency counter....) with the previously mentioned ripple counter, you can discover that no matter how emphatic the datasheets are that what you're doing SHOULD work, you're still only getting 22 Mhz out of the VCO.

Now you just said "huh?", right? Well, the waveform's frequency gets divided by 16384 in my case (and converted to a square wave), so about the only thing you can tell about the original wave is it's frequency. Take the readout of your oscope's frequency counter, multiply by 16384, and that's how fast the oscillator is running. (KC9ASJ says that this is called a "Prescaler", and it may come up on one of the tests to increase the operating privileges. I probably would have guessed that.)

Just as a note to myself, for a 50 Mhz oscillator, one should be getting 3051 - 3052 Hz from Oscope.

Ibiblio look here for oscope.

But I run Windows!

Then check out one of these fine software packages.

I have not run any of these, I don't use Windows (except when I have to at work), so I'm not endorsing any of them.

Further questions regarding Windows, Windows software... will be unpleasantly ignored, and possibly flamed.

I've got a DMM that measures frequency up to 2 Mhz... but how does that help?

See above about dividing the frequency. Measure the frequency through a divider, then multiply.

More things will be added as I figure them out.

Everyone should be able to do this without spending thousands of dollars on gear... right?

Send your helpful suggestions to: KC9ASI