Damper rectifier tubes & voltage doubler

[riccoryder]

Originally Posted by Eli Duttman

The Greinacher, AKA "full wave", doubler is (in fact) a pair of 1/2 wave rectifiers wired back to back. To get decent regulation of the resulting rail, large caps. in the doubler stack are in order. Another poster's remarks about SS diodes being the practical solution is spot on. However, I think it may be possible to safely use damper diodes in combination with good sized caps. in the stack. Install a CL-90 inrush current limiter, whose cold resistance is 120 Ω, in the line between the center of the cap. stack and the power trafo. Perhaps as much as 150 μF. per cap. will be fine. 68 μF. parts should be "a walk in the park".

Negative temperature coefficient (NTC) thermistors are common enough in SS rectified PSUs. It seems they may be exploited in combination with vacuum diodes too.

Generally the color television sets that used dampers such as the 6DW4B also had inrush current limiters. Then they were just called 120 Ohm cold thermistors. Making use of the voltage across the thermistor at start up, a degaussing coil coupled through a voltage dependent resistor (maybe a MOV?) was in parallel. That's why a sort of hum/thud surge is heard right at turn on with many c.r.t televisions and monitors. Then gently tapering down ac magnetic field takes care of tie-die color effects (beams hitting wrong phosphor segments) on magnetized c.r.t.s. Sometimes a slight wiggling of the display is seen on c.r.t monitors just as they come on while the field decays to nothing. That automatic feature is especially useful after kids have had fun with magnets near the screen. There was a comment about SPICE simulation showing a high surge for the damper tubes at turn on. While I think an inrush limiter is a good idea and it also limits the surge in cold filaments, I believe that both the specs and models don't tell the full story. Dampers are used in a mode where high peak current pulses are seen at a rate of 15.734 KHz (former U.S. NTSC system at least) and the average current is low due to the pulses being narrow (during the retrace time, the fall of the sawtooth wave for each scan line). Neither of those specs really applies to a one time surge. Plus, the current will be spread out in time some due to the cathode warming up. Hot spotting and cathode stripping issues can cause damage in some tubes seeing high currents at warm up, but neither the specs nor the models appear to be define that. I would judge dampers current capability based on two things, peak emission current based on cathode area (usually reflected by filament power, as in the different filament currents/power for the 5Y3, 5R4, 5U4, 3DG4, 6x4 6CA4 5ar4/GZ34 etc.). The second factor is thermal, estimate power loss from forward drop and current and estimate dissipation ability based on anode and bulb surface area. Thermal time constants permit large brief overloads. Longer term overloads may affect life, but few amps see as much use as 60's television sets did (there were fewer alternatives to tv then, and there were 9 minutes of ads an hour then, in contrast to the 18-20 now seen in the US) These tubes were very conservative rated in television sets and saw a very low failure rate compared to line-amp and HV rectifier tubes. Low emission was almost never seen, with most of the rare failures being arcing, perhaps from the stress of a failing line amp tube. Those often developed grid emission from contamination over time causing bias to shift and current to rise uncontrollably. I'm more of a fan of silicon diodes, saving the losses and filament power and the cost of more tube sockets. Few wasted watts leaves transformers cooler or able to deliver a bit more to something that produces output. Sometimes an intuitive understanding may tell us things the models don't.