SSTRAN Voltage Measurements
Posted on May 4, 2007
The SSTRAN is a great little transmitter. I am extremely pleased with it, like just about everyone else who has built one! The finished unit has great sound, and the signal has outstanding frequency accuracy and stability. The kit itself is beautifully packaged and documented! The quality of the manual is very impressive– very reminiscent of the best of the Heathkits.
However, my unit has seemed a bit lacking in range compared to some reports. I’ve just been testing it with the indoor antenna so far. I made the following measurements using a Tektronix 475 oscilloscope with a 10x passive probe (10 Megohm +12 pF). With the transmitter operating on 1500 kHz, connected to a 10′ wire antenna, and adjusted as instructed in the manual, except for re-peaking C5 with the scope probe connected, the voltage at the antenna jack is 40V p-p (unmodulated). This seemed low to me, so I took a reading at the collector of Q5. The reading I’m getting there is about 3V p-p. Then I measured the drive at the base of Q4 and got a reading of about 0.5V p-p. The voltage at Pin 1 of U3 is a nice 5V p-p square wave, so the voltage at Q4 is being padded down by a 10:1 voltage divider consisting of R27 and R28. I think this may be slightly under driven to produce full output. I am told that the load at the collector of Q5 should be around 750 Ohms, which suggests that the power being developed there is only about 1.5 mW. This would explain the relatively short range I’m getting (around the house, but not much more).
For fun, I dragged out my Knight-Kit wireless broadcaster and measured that. Using a 10′ antenna and operating the broadcaster with an isolation transformer, but with the chassis grounded to the same ground as the SSTRAN (ac conduit), I’m getting a reading of 100V p-p with no modulation. With modulation, the waveform is a bit strange because you see the audio superimposed on the RF so that the peaks look lopsided (in reality, they aren’t). This unit isn’t capable of 100% modulation, but on positive peaks I see about 300V or so. Unsurprisingly, the range of the Knight far surpasses that of the SSTRAN, even on a very noisy frequency.
Has anyone done any voltage measurements of the SSTRAN? What did you find?
I took the SSTRAN into the lab and made some additional measurements on the scope and spectrum analyzer. The first thing I discovered is that the voltage at Q5’s collector is dependent not on only on the antenna position but also on the tuning of C5. (And, my C5 is starting to become “flaky” and difficult to set. I think I will have to replace it.) As you peak C5 for maximum radiated signal (as measured using a spectrum analyzer with a field probe antenna), the collector voltage drops off sharply on one side of resonance and very slowly on the other side. Thus it is easy to have quite different voltage readings at the collector with only minor changes in the output tuning. In any case, at that location I couldn’t duplicate the antenna position I was using at home, and all the readings were different. The voltage at the antenna as measured in the lab was 60V p-p. The collector voltage was 13V p-p. I also discovered that increasing the drive to the base of Q6 did not increase the output at all. I have no way of knowing exactly what load impedance is being presented to the PA in that situation, but assuming roughly 750 Ohms, then the output power would be about 28 mW (much better). I am not using the 820 Ohm resistor because the manual does not call for it unless you are operating on the lower portion of the band.
In response to radio8z, the comparison between the Knight and the SSTRAN was only to find out which one delivered more output voltage. I used the same antenna and ground on each unit. A 10′ wire is a very short antenna at these frequencies, and represents what we hams would call a “voltage-fed” antenna, so more voltage = more homes. And indeed, the Knight did transmit quite a bit further. Radio8z is correct that the input power of the Knight is more than 100 mW. I have not attempted to measure it, but based on the tube charts, I would estimate somewhere in the neighborhood of 2 watts. They take the output right off the tank circuit with no attempt being made to do any kind of power matching. But it is fair to say that the high impedance of the plate circuit lends itself pretty well to directly driving a short wire antenna. I have made some modifications to my Knight to reduce the hum and lower the distortion, but I haven’t modified the RF section at all. It really ought to be crystal controlled– after listening to the SSTRAN, you never want to go back to a free-running oscillator! If there is interest, I will post more information about what I’ve done to my Knight kit to improve the audio.
Input power only counts insofar as FCC compliance is concerned. It doesn’t really tell you anything about how much output power there is. I killed the drive to the output stage of the SSTRAN, and the input power didn’t change much; in fact it went up slightly from about 98 mW under normal conditions to 120 mW. But that was entirely DC power; there was no RF output at all.
I am working on the outdoor antenna, but meanwhile I am trying to optimize the indoor setup. I think there are many people who can’t put up an outdoor antenna and would like to optimize their station.
My next entry will describe my experience building and testing the Ramsey AM25C and comparing it to the SSTRAN.