10-4 Good Buddy! - A Slider for Your Novice Transmitter
Executive Summary
In my on-going effort to write more and more about things of interest to fewer and fewer people, consider this.
Good VFOs for vintage-crystal controlled transmitters are popular and scarce.
I'm putting together a vintage station for a winter QTH, and after a summer hamfest shopping spree, I found and fixed a Drake R4C receiver and 2NT transmitter. So the search was on for a suitable VFO for the 2NT.
A chance garage sale encounter with an old Siltronix CB "slider" suggested there might be some easily converted vintage VFOs out there.
Turns out the Siltronix 90 has a lot of great features, and is easily converted to a 7 mHz VFO. It drives the 2NT fine on 40 through 10 meters, despite having lower output than Drake suggests is needed.
I expect the modded Model 90 will work well with other vacuum tube transmitters, but may well require another buffer stage or an unun for increasing its output voltage.
What's Good About a Siltronix Model 90?
Well, first it's available. $7 at a garage sale. It would be easy enough to scratch build a VFO, especially a modern DDS VFO. But what's the fun in that?
The Siltronix 90 is from a family of Siltronics CB "sliders" that cover a number of different frequency ranges. As found, mine oscillated from about 10.4 to 10.8 mHz. Conversion to 7 mHz seemed worth a shot.
If you look at the schematic, you can see why conversion might be a good idea.
For example:
1) The slider is solid state.
2) It has a built-in 115 VAC power supply, and an auxiliary 12V power tap that can be used for an additional buffer amp, if needed;
3) The measured output at 10 mHz was about +10 dBm, which is about 2 volts peak to peak, and maybe enough to drive a tube transmitter like the 2NT; and
4) The best part - all the frequency determining elements are mounted off of the circuit board. This made it easy for Siltronics to manufacture units for different frequency ranges, and should make it easy to convert my unit to 7 mHz without fiddling with the oscillator/buffer circuit board.
And if you take a look at the outside, there are still more condiments on this slider.
1) The cabinet and front panel are about the same height and style as the 2NT;
2) The unit has strong physical construction - critical for a VFO;
3) OMG, it has an exquisite two speed concentric tuning dial. This style tuning drive was used on several Swan and Siltronics radios and VFOs, and uses not one but two Jackson Bros 6:1 reduction ball bearing drives.
Take that, you cheesy encoders, this feels like a quality vintage VFO, and really, has a better feel than almost any other VFO out there; and
4) The knobs are Drake style, in this case the tuning dial is from the MSR/DSR series receivers, and the on-off switch from the 7-line. OK, a bit later than the 2NT and the 4-line, but the genes are there.
How to Start?
The inside of the Model 90 looks like this.
The major frequency determining elements, other than the tuning cap, reside on the white ceramic coil form in the upper middle of the photo. The only thing holding the form in place is a single easily accessed screw.
A closeup of the form suggests a couple possible approaches. I picked the wrong one first, of course.
Looking from left to right you can see both approaches in process. The red coil has a lead that extends from the cold end of the coil to a ground lug at the left end of the form. The yellow toroid was an early experiment in increasing the inductance to lower the frequency.
To the right, at the top end of the coil form, you can see a couple ceramic disc caps paralleled across the coil (see the earlier schematic) as well as a silver mica I used as an attempt to lower the frequency of the tuning range.
At the very top of the coil is a miniature air variable capacitor that makes it easy to shift the resonant frequency of the circuit several hundred kilohertz to dial in the calibration of the VFO.
By measurement and observation I learned that the coil had a value of about 2.7 uH, and that the two caps paralleled across the top of the coil totaled 47 pF.
Being lazy, I calculated how much capacitance it would take to shift the frequency down to 40 meter, about 120 pF including the unknown value of the calibrating variable cap. I got close, but of course failed to consider that all that additional fixed capacitance would limit the tuning range of the VFO. DUH.
Back at the computer I decided about 2 uH additional inductance might be a better way to go. And that's why you see the T50-6 toroid mounted in series with the original coil in this picture.
After removing the silver mica cap I had paralleled with the two disc ceramics, I tweaked the calibration air variable a bit, and I had a tuning range of 7.0 to 7.28 mHz. Close enough for jazz.
The reinstalled modified coil looked like this.
But, is Bob your uncle?
Initial Test Results
I anxiously hooked the beast up to the spectrum analyzer to see how the output looked after the initial dialing in of the tuning range. To my amazement, the output remained about +10 dBm!
But we're talking about volts here. Drake, in October of 1969, published an addendum to the 2NT manual that explained how much VFO voltage is required by the transmitter.
This is it.
Measured at the xtal socket into the unknown (but higher than 50 ohm) transmitter load, you see about 3.8 volts peak-to-peak, which is a bit more than the 1 volt RMS Drake spec'd for 40.
This is actually a better result than I expected, but I do want to run 20, 15 and 10 meters on the 2NT, so it looks like extra hardware will be needed.
Two options here, another buffer stage, or an unun to boost the voltage. I opted for the later.
I decided to try a 9:1 unun for maximum voltage boost. One schematic of such an unun looks like this.
Not sure it was worth the effort trying to build the unun into a crystal holder. But there is a great AA8V video from the Antique Wireless Association that strongly advocates putting the transformer at the load, not at the VFO. AA8V is a smart guy whose work I've followed for several years and he's never led me wrong, so there it is.
I tested the unun with a signal generator and a scope. The unun put out almost twice the voltage present at the input. I was a bit puzzled by this (I thought it should be more) but the voltage did increase significantly so I was hopeful.
Time to test the VFO
The VFO Test
Figured the best way to test the VFO was to set everything up on a dummy load and then compare the output with a crystal to the output with the VFO, with and without the unun.
This is what I got.
Band Xtal Output w/VFO Output (no unun) w/VFO Output (w/unun)
40 65 60 60
20 60 60 60
15 55 55 55
10 45 45 45
I scratched my head here quite a bit. And I'm still scratching.
Drake's addendum made sense on 40 meters - I had about 1 volt RMS, like they recommended, and I had full power.
But that Drake addendum claims that 20 and 10 meters require 4 volts rms, and that 15 meters requires 12 volts rms.
I reran the transmitter test with a sniffer antenna on the spectrum analyzer to make sure nothing hinky was going on. On every band I saw the fundamental where it should be, peaking as it should be, and the second harmonic about 40 dB below it, where it should be.
I could only conclude the unun made no difference because the transmitter had all the VFO voltage it needed.
On The Air Testing
I've done a bit of on the air testing. Things seem to work as they should. Power output remains good, there is little drift, the note sounds ok, and that two speed vernier dial drive is sweet.
At this point, I'm going to declare victory and get on the air a bit more to see what goes wrong. I still need to add a relay so that the VFO switches off when the 2NT goes into receive, but this should be a fairly easy proposition.
Next Steps
If you look at the picture of the inside of the VFO, you can see that it would be easy to add a band switch to the front panel opposite the on/off switch.
Hopefully the board will continue to oscillate well as frequency drops. I've had it working as low as 4.9 mHz during my experiments with good output, so 80 may be in reach.
If that works, then I'm thinking a 5 mHz range to double to 30 meters and an 8 mHz range to double to 17 meters might be neat tricks. Don't know if the 2NT can work either place without serious mods though.
The Fine Print
As always I write about what I've done.
These aren't recommendations, just reports. You must always determine what is safe for you, your equipment and the people around you. The Siltronix Model 90 has 120 volts inside, and the 2NT much more.
Be careful, have fun.
And please send me notice of my errors, ignorance, omissions and mistakes, and your improvements!
73 Scott ka9p/zf2sc
Comments