Output tube cathode fusing

[Gaz]

If one wanted to add cathode fuses, how could they determine the rating? Would it be as simple measuring across a 1 ohm in the cathode with the amp cranked and using the next highest fuse value?

[pdf64]

Yes, F400 or 500mA quick blow types are usually fine for 6L6 / EL34.
A LED in series with a 10k resistor can be put across the fuse. This provides a 'blown fuse' indicator, and stops the cathode floating up to a high voltage, which may damage the heater to cathode insulation (assuming the tube is any good).
Pete.

[Gaz]

Thanks for the advice, Pete. I would like to know how to actually size the fuse by measuring the maximum current flow through the tube. Is that how you do it, just measure the cathode current with max signal and round up? I actually found an old Ampage thread where you talk about doing it in your amps, but the method of finding the fuse value is not discussed

[pdf64]

Yes, sorry, measure the cathode current when cranked.
Check all tubes in a p-p amp, in case the duty cycle isn't symetrical.
Also check with a lower load value, as the current may go up some, eg 4 ohm load on 8 ohm tap.
Use a fuse value with a comfortable margin above the highest reading, eg 20 - 50% above.
Pete.

[Gaz]

Thanks, that's exactly the info I was looking for. Man, I don't know why this is not always done on homebrews.

[Gaz]

Pete, I'm going to put these in each pair of power tubes. I'm getting about 460mA max per pair (KT88s) and 475mA when off-setting the load like you suggested. Do you think .75A Fast is okay or too high? I'd have to order 630mA fuses, which kind of an uncommon value. argh.

[Gaz]

Bumping for Pete's musing.

[Roe]

usually 1ohm/0.6w resistors are good for el34s and kt66s and 6l6s. I've personally seen these resistors go out before the tube did

[Gaz]

Thanks for the reply, but I hardly trust resistors to work properly on this application let alone resistors. It's also really easy to measure cathode current, so I don't need to go by rule of thumb. I'm really just more curious about fuse headroom, and what typical fault current might look like.

[pdf64]

From those measurements, I think 0.63A or 0.75 would work fine.
However, my reading of your post is that each fuse would serve 2 tubes?
I see a benefit in fusing each cathode individually. I don't have a clear idea of what fault currents might be from the various failure modes of the various tube types. So if a tube's fault current in a certain situation is only say twice it's normal max current, a fuse serving a pair of tubes may not blow quickly in response to a fault in one of them.
Regards, the fusing characteristics of a 1 ohm cathode resistor, I agree that a ~1/2 watt type will often blow in response to a tube fault. However, I had the unfortunate experience of a tube short which did end up blowing the cathode resistor, but not before the primary winding of the output transformer was damaged (the slow blow B+ fuse didn't blow). That's what set me off looking into more specific fusing methods.
Pete.

[Structo]

Wouldn't fusing the HT circuit serve the same purpose?

[Gaz]

Tom, from what I understand fuses in the cathodes will blow faster (if they rated properly) in the event of a tube short or bis failure. I've read that they won't blow for certain shorts, but they should be more reliable for certain failures than the HT fuse. And like Pete said, sometime the HT fuse doesn't blow...

Thanks for the advice, Pete. Yes, I currently (no pun) have them set up for a single fuse for each pair. They idea was that if one blew the amp would still run on half power so I could finish the show. It will be crappy if that ever happens, but very cool at the same time I first ran into this on a JCM900, and thought it was the coolest thing ever. The rest of the amp?... not so much.

[Structo]

Isn't there also a mod you can add with a LED that will show which side has blown?

[Gaz]

Yeah, that how I have it set up on this amp as well - pretty cool! That's how it is on the 900 too.

[Reeltarded]

LEDs light up? PIMP my PROBLEM!

this is suddenly very interesting...