A Good Old VFO (by Rick, KK7B)

Here is another really great message from Rick, KK7B, sent to the emrfd yahoo group: [emrfd] A Good Old VFO Saturday, August 22, 2009 10:29 PM From: Rick To: emrfd@yahoogroups.com For several critical receiver applications in my lab I've used old Collins PTOs converted to solid state (I just replace the triode in the classic Hartley circuit with a J310 and run the circuit from a 9 volt regulator). I have half a dozen of them in dedicated propagation study receivers, and one SSB exciter I occasionally use on UHF. The other day I was changing something else in one of my receivers and connected the solid-state PTO to the frequency counter on my bench. The PTO was set to 3.100000 MHz. From a cold start (it hadn't been turned on for years) it drifted three Hz over the first ten minutes, and then a total of 10 Hz over the next few hours. When I calibrated one of my 144 MHz propagation study receivers 25 years ago, total frequency drift was <18Hz/hour. I expect most of that was aging of the overtone crystal oscillator in the premix circuit. Old Collins PTOs are common (someone at Dayton this year had a box of unknown ones in decent shape for $10 each, and there are R390 PTOs in the current Fair Radio flyer). I've never had one fail, tuning resolution is infinite, phase noise is low, digital noise is zero, and once I build one into a receiver, that part of the project is done--no improvements, software upgrades, needed. My research receivers are connected to a baseband Fourier analyzer (yes...even 25 years ago). The finest resolution I've used for serious experiments is 10 milliHertz, but more often I use 1 Hz resolution, with 1024 channels in the output spectrum. I often average spectra for more than a minute, so frequency drift needs to be less than 1 Hz per minute. The solid-state Collins PTO is much more stable than needed even for those critical experiments. This is not a fluke. Every Collins PTO I've converted to solid state using a U310 or J310 has had similar performance. Sometimes it is useful to remember that the major benefit of digital frequency synthesis is that it is quick, cheap, and frequency agile. No commercial manufacturer could afford to build a transceiver with a Collins Mil-Spec PTO in it these days. But for an amateur with mechanical skills or access to surplus hardware who needs just one good oscillator, the venerable Hartley with a temperature compensated tuned circuit and a JFET can provide outstanding performance. In music, art, architecture, automobiles, motorcycles. .. there are recognized "golden eras" where some combination of factors resulted in technical hardware that is widely recognized as being superior to what is being produced today. Often the difference is directly related to the amount of skilled labor needed during production. As technical hobbyists, we automatically assume that new is better, but as experimenters, we should be open to the idea that sometimes the technology, ideas, and block diagrams of an earlier era are superior to the cost-driven disposable technology coming off fully automatic assembly lines in some out-of-the-way place with inexpensive labor and attractive business tax codes. The idea that old technology designed decades ago by retired guys might be better than new technology is a radical concept in electronics. But NASA is using a brand new, hand built, Traveling Wave vacuum tube in the current Moon exploration mission. After 100 years of radio experiments- -it is fun to look back and find old technology that might actually work better than some of the new things we've been inventing recently. Best Regards, Rick KK7B