nordmende: Replacing selenium rectifiers and Faulty coupling capacitor

I replaced the selenium rectifier in my Nordmende Tannhauser  V330 ch=4/633 radio.I had noticed that the selenium rectifier was getting very hot.I checked on internet.Here  it is indicated that  selenium rectifiers have a life span of 20 years max.Over time the internal resistance goes up,causing    the HT voltage to drop,the Rectifier  gets more hotter as the internal resistance goes up etc.Eventually the  selenium rectifier  simply blows up.It is no point looking on Ebay for selenium rectifiers as these are mostly old,usually from India.

I replaced the selenium rectifier with  4 1N5404- 3 A 1000 volt units.As the HT voltage would now be around 20 volt too high,I used a small transformer,Primary 230v,secondary 20 volt.Connected the secondary winding  in series with the   the Tannhauser   HT secondary 230 volt AC in such away to buck the  230volts  from the Tannhauser HT 230 volt  winding,Thus 230 volt minus 20 volt gave me 210 volts AC fed to the rectifiers to get the original  + voltage.As I did not added more resistors to the HT  circuit the  regulation did not deterioate.

While working on the radio, a strong  oschillation like cracking noise3 appeared ,removing one of the ECLL800  valves stopped it. I checked the  capacitors near the end penthodes,could not see anything wrong  .I knew it has something to do with the ECLL800 valve.Now  just checking a capacitor with a multimeter is no prove the capacitor is good,you really have to check it with  high voltage.I  thought  the 10nF  coupling capacitor c207  connected to the anode of the  triode of the ECLL800 must be faulty.I replaced it with 10nF 630 volt.Switched on and beholt everything worked.While at it ,I replaced the other 10nF  capacitors.

Maurice 

To thank the Author because you find the post helpful or well done.

I use a resistor or a resistor and thermistor to replicate the selenium drop. You are right that any selenium replacement would soon fail.

I test grid voltage with tubes out on DMM. Also I have a tester made from single use camera's built in flash. About €4.50. I replaced the 300uF with 1uF and put 2 x 1M ohm in series. Then a neon with capacitor (I forget if 0.1 or 0.01uF) across on the earthy lead. A very low leak is slow flash. A high leakage is solid on.

To thank the Author because you find the post helpful or well done.