Voltage dropping resistors burning up!

[Stickleback]

Hi
Burning smell from amp last night. Opened it up today and two resistors are burning up (see photo) - the 1K 2w voltage dropping resistor, and the 100K 2w bleeder resistor. Amp is a home built 6G3 lite (one channel, no vibrato). It's one thing to build an amp, but diagnosing a problem is a whole different kettle of fish. Any suggestions as to what might have caused this, or how to diagnose the problem would be greatly appreciated.
Thanks in advance, Andy, London, UK.

 

[Wally]

??? Overheating due to high current draw??? What are your voltages? Do you have a current limiter that could indicate a short causing excess current draw?

 

[gusfinley]

Remove the bleeder resistor and throw it away. This burning up doesn't make much sense. After you figure out the main problem, you can put a new one back in.

If your dropping resistor is fried, you 've got to much current going through it, so suspect everything downstream of that resistor.

How does you bias supply check out? If this fails you could definately be putting yourself in an overcurrent situation.

It could also be a bad screen on your 6V6's. Do you have an old pair you could swap in. It definately wouldn't hurt to install some screen resistors in there. It will limit screen current and provide a voltage point for debug also.

 

[dogmeat]

seems like I've had bad power tubes do that.... be sure to check the rectifier too (GZ34)

 

[RottenTheCat]

I always used 1meg ot even 2.2 meg bleeders

 

[dan40]

Is this a new build that is giving you problems on the first startup or is this an older build that has been working fine up till now?

 

[Stickleback]

Is this a new build that is giving you problems on the first startup or is this an older build that has been working fine up till now?

I built it around 4 years ago and have used it for an hour a day ever since. So that's circa 1500 hrs, so tube failure is a potential candidate.

 

[Wally]

What are the voltages? What is the current draw in each power tube? If a voltage chart was made when the build was finished, that would be a baseline to which one could compare readings....and that comparison might point to a problem.

 

[King Fan]

Hmm, I'm trying to figure sequence. How do we test the circuit while it's burning dropping resistors? Not sure, how 'bout:

1. Build or hunt up your light-bulb limiter, to find shorts and prevent burning out the next set.

2. You're right about your old tubes. Swap in new or known-good rectifier and power tubes. If you don't have some, all tube amps need spares. JJs are fine for testing and startup purposes.

3. Solder in new dropping resistors, metal oxide, 2 or 3W.

4. Fire it up on the limiter, say a 60-100W *incandescent bulb.* Bright glow? Folks here can tell you how to localize the short using a sequence of tests with the limiter.

5. No bright glow? Pull all tubes, fire up amp and measure HT and filament VACs on the rectifier socket. Record on voltage table. Just for grins, measure 6.3v filament VAC wire-to-wire at the lamp and each of the other sockets.

6. If OK, still no tubes, power up and verify correct negative bias voltage is present on the control grid pin of both power tube sockets. Record on voltage table.

7. If bias OK, plug in rectifier, power up, measure plate voltage on every empty tube socket. They'll be high.

8. If all OK, plug in tubes (and speaker), fire up on the limiter one more time for good luck, *remove limiter*, fire up and play the amp. Sound OK? Now measure B+ nodes VDC to chassis at the positive end of each filter cap. Then measure plate, grid, and cathode voltages for all tubes. Record on voltage table. Calculate bias.

I may be wrong here, but some order of steps is gonna be important.

 

[King Fan]

Also, you said you do have a bias pot? When you measure the bias voltage, also set it to max negative... You can adjust upwards later.

 

[Stickleback]

Thanks King fan, that's just what I needed. I'll report back when I've completed the tests. One thing I did notice today - I took the back off the mains plug and noticed that the earth wire had come loose. Irrespective of whether that has anything to do with the problem, it is clearly a safety issue so I'll be replacing the lead with a moulded plug one.

 

[Wally]

Low voltages would indicate an excessive current draw situation.

 

[King Fan]

Good point, Wally, if I'm thinking straight that's a reason to use a light-bulb limiter when we fire up again.

 

[Wally]

Or just a quick voltage and bias check if a limiter isn't available.

 

[muchxs]

I built it around 4 years ago and have used it for an hour a day ever since. So that's circa 1500 hrs, so tube failure is a potential candidate.

Troubleshooting 101 is to consider all the possibilities and eliminate them one at a time.

Advanced troubleshooting is to triage probable cause, in this case a shorted tube.

 

[King Fan]

Good ideas here; Wally and muchxs are hyper-experienced at amp repair. If the amp is still working (if the resistor burn is a slow burn) other options would be to make a quick voltage table and then swap tubes and see make another -- or just swap tubes and make a voltage table.

 

[Stickleback]

Thanks for all the info and suggestions. Parts now ordered including new valves. Will report back when I have test results. One question - I'm still trying to figure out if the discovery of the earth wire being disconnected in the mains plug could have caused the problem. In such a situation, where is all the power ending up? (ie in normal operation power is dumped to earth via the bleeding resistor, but with no earth connection where does it go?).

 

[Wally]

That draining is to chassis ground.

 

[King Fan]

In such a situation, where is all the power ending up?

The mains / safety earth is not part of the normal AC circuit powering the amp. Merlin says, "This earth bond is for safety only; it plays no part in circuit operation and no current flows in it except under fault conditions."

What is it for then?

"Most guitar amps are built in a metal chassis. Even if it is enclosed in a wooden box, it is still possible for the user to touch the metal somehow, via fixing screws or when replacing valves, etc. Thus for the appliance to be safe it must be completely impossible for the metal chassis (and anything else the user might touch) to become live. This is achieved by physically connecting the chassis to planet Earth via the mains earth wire. Once the chassis is earthed it will be at the same potential as the person using it, and if any live wire were to touch the chassis it would immediately be shorted to earth and cannot shock the user, whether or not a fuse blows."

By the way, you might want to read the next paragraph after *that,* which says to *solder* the safety earth to the IEC inlet: "Do not use a push-fit connector for this."

 

[Paul G.]

We're touching the chassis ground when touching the strings of a guitar or bass, we're touching another chassis ground when our lips brush the microphone. The earth connection, along with polarized plugs, is taking care of us while we're using the equipment.

Geezers like me remember non-polarized plugs, no earth connections. Bad things happened.