St70 runaway bias

I have a stock st70. I replaced the power supply, bias supply, coupling caps and installed new tubes a few months ago. Its been ok since until today. It red plated one tube in one socket.

So far I have done the following.

-Switched tubes in same channel that redplated. The other tube then also redplated in that socket.  
-Switched tubes from good channel to suspect channel and those started to runaway before I quickly turned off.
-I cleaned and retension both sockets in that channel. I also remade the connections to pins 5, 6 and eyelet 23 on the socket that redplated. The 1K resistors measure ok.
-With no tubes installed in that channel, both pins 4/5 measure approx -40vdc. The current does not run away without the tubes installed.  
-At the rectifier diode I am getting approx -75vdc  and then I get -44 approx at the 100uf caps in the bias supply.
- without tubes in channel, I measured higher voltage on can cap. half circle showing approx 466vdc and blank section 458vdc.

The current behaves without tubes installed in the suspect channel. Install the tubes (which both have redplated now in that same socket) and the current runs away and I have to turn it off before it approaches the fuse rating.

I am unclear what my next step should be, unfortunately I dont have another set of tubes and need a little guidance from here. Many thanks.

Well, I would look at :
https://dynacotubeaudio.forumotion.com/t274-quick-troubleshooting-guide-for-dynaco-tube-amps
or http://www.curcioaudio.com/Troubleshooting%20Overview.pdf . Or lhttps://www.audioasylum.com/audio/tubes/messages/25/253369.html
Best to look at all three.

Thank you, I will certainly look at those links.

rmk wrote:I have a stock st70. I replaced the power supply, bias supply, coupling caps and installed new tubes a few months ago. Its been ok since until today. It red plated one tube in one socket.

So far I have done the following.

-Switched tubes in same channel that redplated. The other tube then also redplated in that socket.  
-Switched tubes from good channel to suspect channel and those started to runaway before I quickly turned off.
-I cleaned and retension both sockets in that channel. I also remade the connections to pins 5, 6 and eyelet 23 on the socket that redplated. The 1K resistors measure ok.
-With no tubes installed in that channel, both pins 4/5 measure approx -40vdc. The current does not run away without the tubes installed.  
-At the rectifier diode I am getting approx -75vdc  and then I get -44 approx at the 100uf caps in the bias supply.
- without tubes in channel, I measured higher voltage on can cap. half circle showing approx 466vdc and blank section 458vdc.

The current behaves without tubes installed in the suspect channel. Install the tubes (which both have redplated now in that same socket) and the current runs away and I have to turn it off before it approaches the fuse rating.

I am unclear what my next step should be, unfortunately I dont have another set of tubes and need a little guidance from here. Many thanks.

Start with a set of new EL34.
Tubes that have redplated should not be used they are prone to current avalanche ( what you have observed )

I posted in haste. Its my bonehead mistake $$$ and has costed me a pair of tubes.

I replaced the 270K resistors in that channel a few weeks ago. And because I try not to keep my iron on those phenolic boards too long, I made a cold joint on one of the leads. I had to look real close to see it was bad. After reflowing solder to that eyelet, it is holding bias after a half an hour.


Im not sure why I was getting proper voltage on the grids with the tubes removed despite the bad joint.

Any idea?


p.s.- In answer to my own question, the can voltages without tubes installed were higher but with tubes in the circuit they fell in line to the proper voltages... voltage drop I assume.

rmk wrote:I posted in haste. Its my bonehead mistake $$$ and has costed me a pair of tubes.

I replaced the 270K resistors in that channel a few weeks ago. And because I try not to keep my iron on those phenolic boards too long, I made a cold joint on one of the leads. I had to look real close to see it was bad. After reflowing solder to that eyelet, it is holding bias after a half an hour.


Im not sure why I was getting proper voltage on the grids with the tubes removed despite the bad joint.

Any idea?


p.s.- In answer to my own question, the can voltages without tubes installed were higher but with tubes in the circuit they fell in line to the proper voltages... voltage drop I assume.

Good that you discovered the problem. But you should replace those tubes to prevent future problems.

Yes an expensive mistake thats for sure.

Why do you think proper negative voltage was showing on the grid without tubes installed despite that grid resistor having a bad connection?

rmk wrote:Yes an expensive mistake thats for sure.

Why do you think  proper negative voltage was showing on the grid without tubes installed despite that grid resistor having a bad connection?

Even an unsoldered joint can forward voltage, but unreliable.