Pulling tubes for half power, and running 5Y3

[rickdog]

I got the overwhelming urge to try running one of my 36W TMBs with the outside pair pulled and replacing the solidstate octal with a 5Y3GT thinking that the lowered voltage may help keep things cool with the two power tubes being pulled. I could hear or see no adverse effects at all, and I ran it dimed for several minutes with no redplating or ill effects at all. Is there any possibility that this can damage the amp running it with 2 pwr tubes and the 5Y3 recto? Thanks for any comments!

EDIT: Well I went and played this into my isocab wide open for several minutes, then shut it down, unplugged the amp, and did a little temperature probing on the board. Surprisingly, or maybe not, it was all pretty much cool to the touch. In fact, everything was noticeably cooler than it gets running in 36W mode. This leads me to believe that this mode of reduced power operation is probably not very harmful to the amp. It runs all around cooler, the PT and OT do not even warm up to the touch and the board components remain room temp essentially after several minutes of being dimed, I'm confident this is not dangerous. Maybe I'm missing something, but it seems safe.

[Alexo]

Does this amp use one shared cathode resistor for all 4 power tubes, or are there 2 cathode resistors (one per pair)?

The cathode resistor is there to create the bias voltage for the tubes as current flows across it.

Voltage equals current times resistance.

So if you wanted 10 volts and had 100 milliamps running, 10 (volts) = .1 (amps) x resistance, so you would use a 100 ohm resistor in this case.

But if you only had 50ma running across that R (half power) you would only develop a bias voltage of 5 volts.

Sooooo if your amp has a single shared cathode resistor on all 4 power tubes, I would rewire it to have 2 separate resistors, so you don't end up biased too hot in half power mode.

My .02.

[rickdog]

Thanks Alexo, that makes perfect sense to me, I guess I should have just measured the currents and plate amd screen voltages too and calculated what it is actually doing as far as bias and plate disapation. This would take into account the actual voltage the 5Y3 is supplying and give the actual bias. I could then go from there with my cathode resistor value determination, as it is a one shared cathode resistor arrangement currently and I should probably take your suggestion and go to one resistor for each pair and possibly one resistor for each tube!

I thought about doing the half power switch per Nik's Ceriatone 36W'er wiring, but I wasn't fired up about doing the rewiring and thought that this might work (tube pulling and rectifier switching). Thanks much, I'll take some measurements and see what is actually going on with this configuration and go from there.

[Alexo]

No problem. I don't know the Ceriatone method you mentioned, but a simple and effective way to do it on a switch is to just send the grids of 2 of your power tubes to ground.

[brownnote]

Remember you're effectively doubling your plate-to-plate load when you pull 2 tubes so your reflected impedance is going from around 4k with 4 el84's to around 8K with 2 el84's.

If the quad shares one common cathode resistor it would be something like 75 Ohms which would be quite low for a pair.

Another simple way of putting this option on a switch is to give each pair it's own R/C bias, something like 150Ohm/50uF per pair, and use a switch to break the cathode connection from one pair.

[rickdog]

I dug up this blast from the past to mention that I finally modded one 36 W'er per the Ceriatone 36W half power switch specs and it works. Well, as well as it can be expected to work I guess. Many might be surprised to find that switching to half power only reduces perceived volume what I guess is about 3 dB, maybe 4 dB at the most all other settings being the same, so it is not a real big deal. Merely rotating the master volume20-30 degrees does basically the same thing. Just a FYI thing,
Rick