Since restoring my old hand wired D201, I've been using it on SSB for the first time since channel 16 was the place for SSB. As soon as you have to get off the crystal synthesizer board for a frequency above 23, the drift is bad enough to be an inconvenience. It takes like a half hour for the VFO to warm up to the point where it's somewhat usable on SSB. For a long time I just figured Tram VFO drift was temperature related with 15 tubes in the box.
Yesterday while listening to a QSO on SSB, the 220 volt air compressor just happened to kick on. As soon as it did, the sudden small voltage drop, yanked that VFO right off frequency, to the low side and came right back after the motor came up to speed. That made me measure the voltage at the anode of the VFO tube. That's V302B, pin 6. The schemo says it's supposed to be 110 VDC. It was in 1974, when the line voltage was 110. Today pin 6 is 121 VDC and, it moves around just as much as it did back then.
As I watched this voltage during the warmup period, it started over 130 VDC and worked its way down to 121 after a half hour. After this point any noticeable voltage drift is a result of the line fluctuations and it does move around. The warmup voltage drift is mostly the result of all of the tubes coming up to full temperature and the slight increase in current they draw at this point. That seems to be what is what is pulling the plate voltage a few volts lower after warmup. Just 500mv of voltage drift, causes a noticeable shift in frequency on SSB.
There is zero voltage regulation on this VFO and that is far more critical in terms of frequency drift, than anything temperature related. To fix this, solder the cathode side of a 100 volt, 5 watt zener diode to pin 6 of V302, with the anode tied to ground and more than 90% of your VFO drift evaporates! With everything from the VFO, to the balanced modulator aligned, this thing is a pleasure to operate on SSB after the VFO voltage regulator is added.
This additional one part modification works awesome but, there is no such thing as a free lunch. When the crystal selector is on, the VFO tube is cutoff and the two 22K plate dropping resistors dissipate no heat. In this mod, the tube still stays cutoff until you turn the VFO on but, the two dropping resistors will still dissipate essentially the same heat through the Zener diodes conduction path, even when the VFO is off. They still run cooler than the plate dropping resistors used on the 6BA6 tubes, however, if they are stock, now is a good time to upgrade them with metal film resistors.
The part that does have more stress on it when the VFO is off, is the new Zener since it will also have to dissipate the heat associated with the missing tube current when that tube is cutoff. In the hand wired D201, I just soldered the anode of the diode very close to the nearest chassis ground lug to dissipate those few watts. I do notice a few hundred cycles of drift for about 15 seconds after switching the VFO on, as the Zener current shifts mostly to the tube and the diode cools. It takes about 15 seconds to stabilize and settle 700mv lower.
In hindsight, that first 15 seconds of thermal drift could be eliminated with a 10 watt, chassis mounted Zener diode. If you remote mount the Zener, you should RF bypass the longer wire where it connects to pin 6 of the tube. The needed bypass cap is already in place so all you would need to do is add something like a 10uh choke between pin 6 (at the tube socket) and the longer wire to the remote mounted Zener diode.
Yesterday while listening to a QSO on SSB, the 220 volt air compressor just happened to kick on. As soon as it did, the sudden small voltage drop, yanked that VFO right off frequency, to the low side and came right back after the motor came up to speed. That made me measure the voltage at the anode of the VFO tube. That's V302B, pin 6. The schemo says it's supposed to be 110 VDC. It was in 1974, when the line voltage was 110. Today pin 6 is 121 VDC and, it moves around just as much as it did back then.
As I watched this voltage during the warmup period, it started over 130 VDC and worked its way down to 121 after a half hour. After this point any noticeable voltage drift is a result of the line fluctuations and it does move around. The warmup voltage drift is mostly the result of all of the tubes coming up to full temperature and the slight increase in current they draw at this point. That seems to be what is what is pulling the plate voltage a few volts lower after warmup. Just 500mv of voltage drift, causes a noticeable shift in frequency on SSB.
There is zero voltage regulation on this VFO and that is far more critical in terms of frequency drift, than anything temperature related. To fix this, solder the cathode side of a 100 volt, 5 watt zener diode to pin 6 of V302, with the anode tied to ground and more than 90% of your VFO drift evaporates! With everything from the VFO, to the balanced modulator aligned, this thing is a pleasure to operate on SSB after the VFO voltage regulator is added.
This additional one part modification works awesome but, there is no such thing as a free lunch. When the crystal selector is on, the VFO tube is cutoff and the two 22K plate dropping resistors dissipate no heat. In this mod, the tube still stays cutoff until you turn the VFO on but, the two dropping resistors will still dissipate essentially the same heat through the Zener diodes conduction path, even when the VFO is off. They still run cooler than the plate dropping resistors used on the 6BA6 tubes, however, if they are stock, now is a good time to upgrade them with metal film resistors.
The part that does have more stress on it when the VFO is off, is the new Zener since it will also have to dissipate the heat associated with the missing tube current when that tube is cutoff. In the hand wired D201, I just soldered the anode of the diode very close to the nearest chassis ground lug to dissipate those few watts. I do notice a few hundred cycles of drift for about 15 seconds after switching the VFO on, as the Zener current shifts mostly to the tube and the diode cools. It takes about 15 seconds to stabilize and settle 700mv lower.
In hindsight, that first 15 seconds of thermal drift could be eliminated with a 10 watt, chassis mounted Zener diode. If you remote mount the Zener, you should RF bypass the longer wire where it connects to pin 6 of the tube. The needed bypass cap is already in place so all you would need to do is add something like a 10uh choke between pin 6 (at the tube socket) and the longer wire to the remote mounted Zener diode.
