Full Wave valve rectifier selection

[steve]

Hi All,

Anyone know of an obtainable Full Wave rectifier with a Va rating of at least 900V (PIV >= 1.8KV) that only requires a Rs of say 100R preferably with an octal base and 5V/2A heater? (Heaters can change though). I am currently using 5R4WGB's but Rs needs to be 575R at that plate supply voltage for a 4uF cap input filter and I get too much voltage drop for my application and the OP of the 5R4WGB's is right on the top limit at just over 1KV.

I have seen higher voltages used with 5R4WGB's in a bridge configuration with the internal diodes paralleled together and only the transformers Rs. Could I use 2 x 5R4WGB's with the two internal diodes operating in parallel configured for FW and get more leeway with Rs? (and therefore a higher O/P voltage). Only need 95ma @ 1100V-1200V OP from the filter. (mains Tx OP is 925-0-925 @ 150mA)

Suggestions anyone?

[John Caswell]

Try a pair 6D22s or possibly U19s

John - Dr John

[pre65]

The 6D22s is a 6.3v half wave rectifier so (as Dr John says) two would be needed for full wave.

I use two in each channel for my ECL82 amp power supply and two as part of a bridge (hybrid) rectifier in my 845 power supply to get 950v HT.

The heaters can be elevated to full HT if the supplying transformer is suitably rated.

[steve]

Thanks John, Philip.

U19/6D22 look fine - (anode caps arggh I hate those).

John can you help me confirm or correct my understanding with the theory using 2 x 5R4WGB's with the internal diodes in parallel. My thinking was as follows:

The 575R Rs is necessary to limit the current pulses during the charging cycle of the first 4uF cap of the filter when operating with Va of 900V. (Rs=100R for plate curves less than 900V). If the two internal diodes are connected in parallel this should allow me to half Rs to say 260R (120R of TX impedance and 140R external resistor). In this configuration each anode is at the same potential and as PIV of 5R4WGB is 3KV I should in theory be able to run easily with a Va of 935V.

The downside I can think of is if I have a failure of one diode inside the envelope, then the remaining diode will be operating with 50% Rs but that is just going to cause failure of remaining diode prematurely. Since the valve is already failing it's no loss.

The above config would mean that I can keep using my octal 5R4WGB's and simply add a second (I have plenty spare plus I can avoid anode caps)

Any thoughts?