Tube failure -- what would happen in worst case?

Self biasing or not, the problem comes when a cathode to anode short happens in a power tube. The least damage is blowing a fuse in the amp. But for me, the cathode resistor gets fried immedately and needs replacement. This also happened on a current source tube (CARY Rocket 88) and I had to replace that resistor too - I knew it was bad and asked mfr for the value, but they gave me the wrong resistor and I had to send it in ($60 bucks each way).

So it is a pain. But the moral is this NEVER EVER leave your house with tube equipment powered on.

Cathode to anode shorts are very very rare, and the sort of thing that should blow a fuse without damaging anything else. If not, the manufacturer has not thought things through.

I leave tube equipment on all the time. However, its tube equipment that I know is designed to not get in trouble if a tube fails. No blown cathode resistors, maybe a blown fuse. But in truth I have yet to have that happen despite using such amplifiers for over 35 years.

That's not to say that all amps are safe like that. The Fourier amplifiers mentioned earlier were prone to spectacular failures as were certain ARCs (when they were sold a bad batch of Cornell Dublier filter caps back in the 1980s).

I friend had flames coming out of the top of his power amp do to a tube failure. I'd be more worried about the house burning down than damage to a component.

When I was a kid we had a Philco color tv that my dad bought in 1968. (I loved seeing star trek in color on that tv back then.) It had lots of tubes and the repairman came to visit us often that first year it was under warranty. I remember one time we turned on the tv and we heard crackling sounds and the wall lit up behind the tv. That was exciting. The repairman came out and fixed it once again. I can remember my dad sending me down to the drug store on my bicycle with a basket of tubes from that tv to test them and buy replacements. Every drug store and hardware store had tube testers back then.

Worst case? That so much depends on the design of the circuit in which the tube is being used. In my case I'm burning tubes in extreme DIY amps and preamps. All the tubes, except the output (that's on the drawing board) have current source fed,shunt regulated B+. Should a tube fail, the shunt device would be required to shunt all the bias current. It is not heat sinked for this and will fail. I had a driver tube in my beta Venice (a 3 stage fully differential phono pre from K&K Audio) arc. The driver stage is direct coupled to the input stage which uses MOSFET based current sources as plate loads for the tube half of a tube/FET cascode. That arc caused one of the MOSFETs in the input stage to fail. I had an output power tubes arc on turn on that caused the grid to short to the plate. That smoked the grid-stopper resistor on that tube. Better the resistor than the secondary winding of an expensive IT transformer. For me, trying to design for all failures is impractical. It's much easier for me to just keep on hand the necessary parts to repair an amp in the rare instance a tube does fail.