R17 cooked after 1.5 years of use

[duckbert]

So... I built my SSE about a year and a half ago, and have used it (with great delight) almost daily. A few days ago, my house slowly started to have the (very distinctive) smell of the magic smoke. I opened it up, and R17 (but not R27) has quite obviously heated up significantly. Its a 560 ohm 5W resistor as spec'd out in the chart for my 6L6 tubes.

Now - my electronics education is failing me since our coverage of tubes was very minimal. Where do I begin to troubleshoot this? Should I make some voltage measurements with the power tubes out?

 

[mike567]

Replace the resistor, check coupling capacitor(s). (C11 or C21) Make sure your tube doesn't have any shorts that isn't supposed to have. (heater to cathode, screen grid shorted to something) You may have other problems but a basic place to start.




See Sudden SSE Capacitor Failure

 

[duckbert]

Thanks. It seems like everything is fine, but that maybe I should have deviated from the default bias resistor. The tubes are fine, and the voltage across the resistors is 35V. When I checked these resistors, one is at the low end of the tolerance and one is at the high end, so maybe it just got warm, and it was just the cumulative effect of slowly cooking the board in the same place over a year and a bit.

My B+ is 360, and output impedence is 5000 ohms, and am using a 6V6GT tube - should I be using a different resistor maybe? Like 820 or so? Worth trying?

Thanks

 

[VictoriaGuy]

Its a 560 ohm 5W resistor as spec'd out in the chart for my 6L6 tubes.

My B+ is 360, and output impedence is 5000 ohms, and am using a 6V6GT tube - should I be using a different resistor maybe? Like 820 or so? Worth trying?

Which power tubes are you using?

35V drop across a 560R is 62.5 mA.

 

[Tubelab_com]

using a 6V6GT tube - should I be using a different resistor maybe? Like 820 or so? Worth trying?

Yes, you are cooking those poor 6V6's. 35 volts across a 560 ohm resistor means that 62.5 mA are flowing through your tubes. With 360 volts of B+, minus the 35 lost in the resistor, the tubes see about 325 volts.......That's about 20 watts per tube......I'm guessing that the tube which was connected to the toasted resistor has gone into a red plate runaway, maybe more than once.

I had a 6V6 based SSE several years ago. Mine only ran about 330 volts of B+. I know I had something bigger than 560 ohms in it, but I don't remember what it was. 820 ohms would be a good place to start. You need to get the plate dissipation down to the 10 to 12 watt area, which would be 33 to 40 mA assuming that there is about 300 volts actually across the tube (B+ minus the cathode resistor and OPT drop).

 

[w5jag]

Seems like I was using 910 ohms when I ran 6V6's in the SSE ... I agree 820 is a good place to start, and I would probably use a 5Y3GT, or even 5W4GT, for the rectifier tube to drop that B+ a bit.

Win W5JAG

edit: looked at my notes for the Elekit TU-879S that I have at our lake house, and which I modified for 6V6GT - my notes show that B+ is 337 VDC, V Screen is 308 VDC, V cathode is 21 VDC. Rk is 750 ohms. Ik is 28 ma, I screen is 1 ma. A bit less than 9 watts dissipation, including the current in the screens. Those measurements are with Bendix 5992 tubes, pretty similar to regular 6V6 ...

 

[Tubelab_com]

I had built an SSE just for 6V6's since I had a box full of old used tubes. It used one of the Allied 6K56 VG power transformers that I usually put in 45 amps. It will make 320 to 330 volts max on a good day with a 5AR4, high line voltage, and light load.

I wound up putting a pair of those old smoky grey glass RCA 6V6GT's in it wired as triodes. The sounded almost as good as my 45 based TSE. Someone offered me far too much money for it and it went bye bye......

6V6GT's and 6V6GTA's are known to be capable of eating a lot more voltage than they are rated for. Some old Fender amps feed them 400+ volts on modern high line voltage. High voltage AND too much dissipation is probably not a good combination, especially with some modern new production tubes with less than perfect vacuum. They will eventually turn gassy and runaway.

New tubes may fix the problem, but they will go down the same path eventually at 20 watts.

 

[w5jag]

I don't have any notes handy from when I was running 6V6's in the SSE, but, IIRC, I was running vanilla GE and Sylvania up around 365 volts, maybe a little more, less the cathode voltage, because the data chart / models on your site indicate that power output with 6V6 about doubled at the higher voltages. Hence the 910 ohm Rk .... I didn't have any failures of tubes or parts.

The B+ voltage in the Elekit is from an R-core xfmr to a solid state rectifier, MOSFET ripple filter, so non negotiable B+ voltage. 5992 has max 300 plate / 275 screen ratings, so that may be why I set the screen lower than the plate when I modded it; I was probably aiming for 275 volts on the screens and did not have the right parts in stock at the time ... edit: it is a single ended pentode amp, and I left it in single ended pentode ...

It's got Chinese 6V6's in it now ... my wife has been known to leave the house with the amp turned on, and it can be months before someone comes back to turn it off, so it's cheap tubes and everything conservative ...

I have run the JJ 6V6S at 18 watts or so without seeing red plate, but I don't know what that would do to tube life, and I eventually gave all my JJ's away to my nephew.

Win W5JAG

 

[Ritchie]

Another option could be a choke input or very small cap input to trim B+.
I ran a 6v6 in the sse with a choke input and it worked very well. My favorite sound from that amp.