Power tube failure modes

[BackwardsBoB]

What's happening when screen resistors fry? What is the screen shorting to? and when the heater circuit cooks from a power tube failure, what is the heater shorting to? What are the top power tube failure modes?

[Wilder Amplification]

What's happening when screen resistors fry? What is the screen shorting to? and when the heater circuit cooks from a power tube failure, what is the heater shorting to? What are the top power tube failure modes?

In the case of screen shorts, I would think either the beam forming plates/suppressor grid. Since they're so close together, I suppose it's very possible for a screen to short to a control grid which would have a much greater voltage differential relative to the screen voltage, but you'd think you'd also possibly take out the swamper and the grid load resistor in that event.

For heater shorts, more than likely a socket arc between pins 2 and 3 (heater and plate).

[stokes]

When a screen resistor fries it is doing its job.When the screen draws too much current,the resistor is there to keep the screen from cooking and will fry,acting like a fuse.I would assume the screen itself most likely would short to the suppressor grid,since as was already stated its close to the screen and connected to ground via the cathode.I would also assume that that is not always the case,tho.The screen,just like the plate,draws current and increases as the tube is pushed.Age,and running the screen close to its max tolerances would also cause failure,so it isnt necessarilly shorting to anything,it just "gives it up" sometimes.

[Merlinb]

In the case of tetrodes it will usually short to the control grid (it won't short to the beam-plates since they are quite far away, around the 'side' of the support rods). For example, the photo below you can see how the support rod has sagged and touched the control grid support rod.


With pentodes it might short to the suppressor, but the support rods are being pulled inwards rather than outwards, so this seems generally unlikely.
However, this sort of failure should never happen if the screen resistors are large enough for the job; screen failure is an indicator of undersized screen resistors.