3W resistor in distress?

[mcamelo]

Dear amp heads,

twice now, my Plexi 6v6 crapped out after several minutes of playing: one of the 6v6s lit up like a bulb in what I can only assume was some kind of short. The first time was in the hands of a beast of player friend of mine while playing it fully dimmed for several minutes. The second was after about 30 minutes with both gain and volume at 75%.

A bit of smoke billowed through one of the chassis' holes before I was able to switch the amp off. While looking for the source of the smoke, I noticed the two 3W (screen grid?) resistors both look a bit sick (see attached).

Could it be that they are getting fried and ending up shorted when too hot? If so, what could be causing it? I can obviously replace them, but if the root cause is somewhere else, they will soon need replacing once again.

Disclosure 1: I play the amp through a reactive load that I've built. Being a much simpler circuit, I trust it more than I trust my amp build. Having said that, I am open to it being the source my problem.

Disclosure 2: I have pushed this amp hard in the past during band rehearsal without any problems (though maybe not dimmed for as long as this last time).

Disclosure 3: I have banged the amp around a bit when transporting it (still haven't got around building a proper cab for it) and have had to replace damaged power tubes in the past for that reason. I have since physically inspected the integrity of the wiring (probing with a wooden chopstick with the amp on) and it still looked pretty solid.

Disclosure 4: The current pair of tubes was borrowed from another amp. In that amp, I have experienced a once-off small change in output volume. I had noticed a similar change in volume a few minutes before the last event. Might this be pointing to a potential defective tube?

What should I do? Any help would be much appreciated.

-Marcelo

[pdf64]

Wow, I wouldn't have thought 6V6 could draw that much screen grid current

So by 'plexi', you mean a typical Marshall JMP era 1987? eg https://drtube.com/schematics/marshall/1987pljp.gif
What the HT VDC on the screen grid node; at idle and at full power output?
What's the nominal OT primary impedance?
Are you sure that the reactive load's impedance-frequency characteristics are reasonable?

[tubeswell]

^What pdf64 said^
Information needed
Schematic and Voltages?
Tube current?
OT spec?

[sluckey]

If you gonna play that hard you may consider replacing those 470Ω screen resistors with 1K/5W resistors. Lot of people do that with Deluxe Reverbs. And pretty standard in many 18W EL84 builds.

[sluckey]

So by 'plexi', you mean a typical Marshall JMP era 1987?

I believe he means this...

     https://el34world.com/Hoffman/files/Hoffman_Plexi_6V6-V2.pdf

[tubeswell]

Okay - so it's a Marshally amp with 6V6s. The screen voltages are higher than the plates in that Hoffman schematic. The LTP is also warm-biased.  When you play it hard, you're combining the tendency of warm-biased LTP inverters to muck up the output stage duty cycle, with high g2 current = Recipe for red-plating.

[mresistor]

What brand 6V6 are you using? If you're going to push this amp hard I would suggest using JJ 6V6 power tubes.


Even then maybe think about a 6L6 EL34 5881 based amp for more power.

[PRR]

.....I play the amp through a reactive load that I've built......

An under-loaded amp will suck high screen current. Show this "reactive load".

[mcamelo]

I believe he means this...
     https://el34world.com/Hoffman/files/Hoffman_Plexi_6V6-V2.pdf

Yes. Thank you.

6V6s are a pair of Mullards.
OT impedance is 8400 OHMs. I will measure voltages, resistance (and current) next time I get a chance to fire it up.

The reactive load circuit is this one: https://www.marshallforum.com/threads/simple-attenuators-design-and-testing.98285/.

I may need to find a farm or warehouse district to stress test the amp without an attenuator before the speaker. 18W is still pretty loud for my quiet residential suburd :-)

[mcamelo]

An under-loaded amp will suck high screen current. Show this "reactive load".

For what is worth:

My multimeter measures the speaker at ~10 OMHS (+/- 2).
If I put the load box in front of it, it reads ~12 OMHS (+/- 2).

[mcamelo]

Okay - so it's a Marshally amp with 6V6s. The screen voltages are higher than the plates in that Hoffman schematic. The LTP is also warm-biased.  When you play it hard, you're combining the tendency of warm-biased LTP inverters to muck up the output stage duty cycle, with high g2 current = Recipe for red-plating.

Alright. Could that be addressed with Sluckey's suggestion of replacing the 470Ω screen resistors with 1K/5W ones?

That are the trade-offs of such a move?

[sluckey]

Alright. Could that be addressed with Sluckey's suggestion of replacing the 470Ω screen resistors with 1K/5W ones?

That are the trade-offs of such a move?

Tubes will probably last longer. Same for the resistors. Probably sound the same.

[dude]

I use 750 ohm, 5 watt screens, for the reasons mentioned above, no issues. I use USA made 6V6, GE, RCA, etc., matched by current draw,  never any red plating or signs of heat on the grids but I don't use an attenuator. Why don't you try a PPIMV, LAMAR, with dual 250kB, Lin  pot. With both masters (assuming you have the pre), work well together, IMO. 

[mcamelo]

I've bough a few 5W resistors ranging from 850 to 1500 OMH. I will try them on and pay attention to how the amp behaves and report back.

Somewhere else I've read that one wants to choose the smallest resistor value needed to keep the tubes from glowing brighter (indicating that the screen is over dissipating) as the intensity of the input signal increases. Can someone here confirm that? My amp, as current set, is definitely doing exactly that.

Ref: https://ampgarage.com/forum/viewtopic.php?t=20807

PS: does the resistor construction matters? I can't find 5W metal film resistors locally, only wire wounds.

[PRR]

> does the resistor construction matters? I can't find 5W metal film resistors locally, only wire wounds.

Not much.

[mcamelo]

For closure and future reference: I did replace the resistors for a set of 1K 5W. The severe problem I was experiencing before did went away, but the offending tube was still red plating somewhat. I swapped them around on their sockets and the offending tube was still red platting (now plugged to the other socket). I replaced the tube and, now, all seems to be working as intended. I did not notice any difference to the tone, so I'm keeping it like this for now. Thank you all for your input and help, folks.

[sluckey]

For closure and future reference: I did replace the resistors for a set of 1M 5W.

Surely you mean 1K 5W.

[mcamelo]

Surely you mean 1K 5W.

Yes. Fixed in the post. Thank you for the correction.