Why is my 1k 2w resistor burning up?

[James_Wildart]

Anyone have any ideas? It's happened twice now. It's a plexi clone, I attached a pic.

[shooter]

to much current

[AlNewman]

What does it do?

[passaloutre]

What does it do?

It burns up. Look at the photo

[passaloutre]

Sorry, bad joke. Looks like a screen resistor, which is supposed to be 5W in a plexi.

OP, do you run this amp cranked to power tube distortion frequently? Care to share more details or a schematic?

[James_Wildart]

No, I actually rarely turn it on. I was having problems with the normal channel not having any output; the bright channel was fine. But, I have two different layouts, and they both call for the same resistor. I attached a screenshot. The resistor in question is shown in light blue.

[stratomaster]

That's an odd arrangement for the screens. Do you have a schematic for a more complete picture?

Also, the workmanship is pretty poor.  Looks like there's an added 10k or so resistor between pins 1 and 8 (bad move). The wire insulation is both burned and chewed up. Solder joints that are visible need help.

There's a lot more going on here than just a burning screen resistor.

[AlNewman]

There was no schematic included with the kit?
A 2W screen supply resistor seems underpowered.  On the other hand, seeing as there's 2 screen resistors following it, you should be able to find out if one of your screens is drawing more current (IE, enough to detonate the supply resistor) than the other.

[James_Wildart]

It was based on the metropolis JTM45 schematic, I attached a pic and circled the 1k2w resistor

[passaloutre]

The Marshall schematics I can find specify a 5W resistor on each screen, where you’re feeding both screens with a single 2W resistor. That seems suspect to me, but I’ve seen plenty of amps get away with no screen resistors, and most of my amps have 3W, though mine tend to use 6V6 tubes.

[AlNewman]

I suppose it depends on the amp.  EL34 max screen dissipation is 8W/side.  If the screens are at 400V, that's 20 mA x 2.
.04 x 1000 is 40V x .04 = 1.6W

If the screens are at 300V at max wattage, then the 1k resistor would be pulling closer to 3W.

But, If the amp was designed with a 2w resistor, and has worked till now with a 2w resistor, I would suspect another component has caused the resistor to fail.

[Merlin]

If the anode becomes disconnected it will cause the screen to pull excessive current and burn out the screen resistors. Could you have a broken anode wire or dirty valve pin? Some bias probes can also create this failure mode if you have your meter on the wrong setting.

[glass54]

.....and that's the reason I prefer separate Screen Resistors. 
For 6L6 and EL34 I also prefer 5W Metal Oxide or WW.
Makes fault isolation easier in O/P tubes 
Regards
Mirek

[James_Wildart]

After replacing that resistor, I turned the amp on, and it worked perfectly for a few minutes, then the Normal channel started to fade out and flubby sounding, then totally quit. The High Treble channel works perfectly. I can't figure it out. I'm guessing a resistor slowly blew, but why only on the Normal input?

[passaloutre]

I think that’s a different issue entirely.

Probing pin voltages is usually where I start

[James_Wildart]

You're probably right. I thought I would put the symptoms in ChatGPT just for the heck of it, and it said it is very likely the Cathode Bypass Capacitor.

[passaloutre]

Probably not. That bypass cap is shared between the channels. We have the power, unlike ChatGPT, to figure it out. Measure your pin voltages.

[James_Wildart]

OkayI'll measure them I promise, but see if this makes sense. I went ahead and changed out the bypass tube (by the way this is an old JTM45 circuit, not a plexi)
So I changed it, plugged everything back in, still no sound, I started wiggling the tubes just to see if there might be a loose connection I wiggled V1 nothing, then I wiggled V2 and boom I had instant sound. Isn't the second preamp tube for the lead channel? The Lead channel was working fine it was the normal channel that had no sound.

[passaloutre]

I dunno what does ChatGPT say?

[James_Wildart]

Why ONLY your normal channel seemed dead

This is the subtle part:

The connection at V2 is flaky
The high treble channel has more gain / brighter signal
So it can still “push through” a weak connection

But:

The normal channel = lower gain → disappears first

👉 So it looked like only one channel died…
…but really the issue is downstream at V2

🔥 Your actual problem (very likely)

🎯 Bad connection at the V2 tube socket

Specifically:

Dirty socket contacts
Loose pin tension
Cold/cracked solder joints under the socket
⚡ What to do (this should fix it)
1. 🧼 Clean the socket
Spray contact cleaner (DeoxIT if you’ve got it)
Insert/remove the tube a few times
2. 🔧 Retension the socket pins (VERY effective)
Power OFF, caps discharged ⚠️
Use a small pick or needle
Gently bend the pin contacts inward

👉 Old sockets loosen over time—this is super common

[Rontone]

Have you checked the heater wiring?

Maybe moving V2 is disturbing the heater wiring that goes across to V1, could be a bad joint or wire connection and V1 is not actually heating up

The pins can move quite a bit when tubes are inserted

[James_Wildart]

This is what someone said on the Marshall forum.

"Those small tubes (ECC83) are actually two tubes in one. The two channels uses one half each of V1. After volume controls the channels mixes to one and goes to V2.

If one channel sounds good and not the other, the problem is likely before the mixing point."

[AlNewman]

Maybe you should take it to a tech.

[apeontheweb]

I think you found the source of the problem -- V2? It's one of these things:
1. the tube itself
2. the tube pin sockets
3. the solder connection between a wire and the pin socket
4. something else
So maybe try this:
1. try a different tube in that socket
2. clean the sockets and tube pins
3. closely inspect and wiggle each individual tube pin sockets
4. or go back to the drawing board