RCA Victor 8PT Selenium to Si Conversion

[Jon1967us]

RCA Victor 8PT Selenium to Si Conversion

I would like to convert this TV's 64 year old Selenium to Silicon.



It seems to differ from most Seleniums that I've seen, in that the AC from the transformer comes into the center, and branches out in 2 directions. On one side, wire 1 - returning to the transformer secondary, going through the 60uF and wire 2 going feeding the B+. On the other side of the Selenium, passes through another rectifier section to ground.



In order to convert to Silicon rectification, will I need 2 Si diodes: 1 going from the AC to the 2 yellow wires, and the other from AC to ground?

Note, I am aware that I may need to add a dropping resistor.

[maxhifi]

That is a voltage doubler circuit. Wire it just like the schematic and you are good to go. I did not add a dropping resistor to my 8PT, others may have.

[old_coot88]

Bear in mind that the selenium stack needs to be taken out of circuit electrically, though you probably want to keep it in place for aesthetics. You can still use one of its lugs for a tie point. The + sign on the stack is in fact cathode*. Banded end of Si diodes is cathode. Series resistance probably not needed since the xfmr winding contributes that.

*Seleniums conventionally show a + sign on the cathode for some obscure reason.

[Jon1967us]

Greatly appreciate this valuable info, folks. Will post results when done.
I've been distracted by another couple of problems with the set, namely picture cutting out when brightness or contrast are adjusted and a significant drop in volume.

Well...1 step forward, 2 steps back...welcome to the hobby, right?

[Electronic M]

One weird thing I sometimes do to approximate the selenium voltage drop in some sets is to replace 1 single section selenium rectifier (yours is a dual section rect) with at least as many diodes as plates connected in series. Selenium's have around 1V forward drop per plate and Si diodes have .7V per diode so connecting series So diodes can approximate the effect....If you ever do this make sure to leave as much space between new diodes as possible or mount them to a terminal strip. When you connect more than 4 in series the center ones get hot and I've had a center diode burn open once before(the rest were fine and protected the set).

[Tom9589]

If all of the Si diodes are connected in series, they would all have the same current flowing through them. Why would the center diodes get hotter and burn out?

[Yamamaya42]

most likely has something to do with thermal conductivity, thermal runaway, and resistance, the center diode being the most susceptible while under load.

[Electronic M]

Pretty much. The end diodes have leads connected to terminal strips that act as heat sink but the center didn't. The way I built mine the diodes leads were soldered together mid air with the series diodes closely spaced and their connections covered in heat shrink tubing.
I could have added a terminal strip and put each diode to diode connection on its own terminal for better heat disapation but I haven't tried that yet.

[jr_tech]

How much current were you drawing through the diode chain?

jr

[Electronic M]

How much current does a CTC2B draw?

[Tom9589]

Guess I worked too long for a company that manufactured equipment for the defense industry. I thought that the diodes in question were mounted on a terminal strip, with each diode connected to an individual terminal. Connecting the diodes together mid air would definitely put a thermal strain on the middle diode.

Heck, we even had to use a rubber strap around any part that weighed over 1 oz to anchor it to the printed circuit board.