RM100 Tube failure light and randomly blowing tube fuses

[mThomasDutch]

Hi guys. This one is for the tech heads. I got some weird stuff going on with my RM100 all of a sudden. Was working fine, then it wasn't - as is always the case.

So I took it out of it's head cabinet to try a little experiment I was curious about. I flipped it over with the tranny's resting on wood to ensure that the tubes were safe, powered it up and started playing just a little bit at low to moderate volume. In a minute or so I saw a tube failure light go on at tube 4. I assumed it was a bad tube, and I wanted to get on with my experiment, so I simply removed the fuse (it was blown) for that tube and tube 1, halved the impedance at the switch and powered back up.

A few minutes later I see a failure light at tube 2. That fuse was also blown. So, I tried a known good set of tubes in all 4 slots - biased them up properly, played a few minutes - at a much more cautious volume now - and another fuse blew.

Now, the tube failure light on tube 2 stays lit, even though the fuse is OK, and I can no longer bias the tube all the way up. In it's maxed position on the bias pot I can only get to about 29mA where before it was at 34 long before the max.

Any ideas?

 

[Jaded Faith]

What rating of fuse are you using? A 400mA right?

 

[tschrama]

Bias voltage is failing, or becoming unstable?

 

[mThomasDutch]

Yes, I am using 400mA fuses. I measured the bias voltage on pin 5 of all the power tubes and at the max I see about -70 vdc and at minimum about -20 vdc.

I found that as long as a tube is present in the socket the tube failure light comes on. If I leave the fuse in but tube out no light comes on.

 

[mThomasDutch]

I also checked all the resistors immediately around all 4 tube tube sockets. I didn't have the schematic handy, but I was able to at least confirm that the resistors on the visible side of the board all showed the same resistance when measured in circuit. I expected if I had a bad resistor something would have shown up different between them.

 

[tschrama]

If the tube failure LED is on.. that means the cathode current is flowing through the LED, not through the fuse. Try a different fuse, 250mA might be saver.

The Fuse can be burned open if the tube was defect, or maybe if the bias voltage has a defect. Sometimes it can be hard to see. Did you check the fuse with a multimeter for continuity?

 

[mThomasDutch]

Also, to add to the mystery - I tried yet another set of tubes, this time a set of 6L6 from my RM100 combo (previous sets had been EL-34). I only put them in positions 1 and 4 and no fuses have blown playing briefly at moderate volumes.

I suspected a bad cathode resistor, but those (the big 5 watt ones I think) all measured good.

Baffled at the moment...

 

[mThomasDutch]

I did check the fuse with a meter, and also tried swapping it with a known good from another socket. Still no joy.

 

[ibenhad]

Pull the amp out of the cabinet and check for fried traces around the Power tube sockets. Early models had lots of arching issues on the tube boards.

 

[mThomasDutch]

Updates.
I measured all resistances and voltages around the tube sockets. All measure good according to the schematic I got from somewhere on this site.

I found that the reason the failure light was on was that the resistor R19A (1 ohm 1 watt) was open. I found that the voltage on screen 2 was about the same as plate (510 to 513), but other than that the heater voltages looked good, bias voltage was there and steady, etc.

So, I put a set of 6L6's in and biased them up. I biased one of them in a different socket, measured the bias voltage at the pin on the tube, and then put that tube in V3 and made the bias voltage the same until I can replace the blown resistor.

I plugged in and played a few minutes at low volume and all was well until the 4th of July happened.

Now this is weird. I saw light above and below thee board, so I made a thorough inspection. Here's what I now know:

1. The 6L6 that lit up came out of my RM100C. It now has a crack in the glass near the base and of course is bad. Unfortunately I don't know if it had the crack and lost vacuum before the event, or if the event caused the crack.
2. The resistors R9A and R10A on the A board both have holes blown in them. The schematic identifies them as R9G and R10G, but they are definitely on the A board and connect to the orange wires coming from the PT. I believe page 1 of the schematic identifies this as the 6.3 Volt filament supply.
3. The orange wires coming from the PT have 70+ volts AC on them, as does pin . That seems like a lot more than 6.3.

Am I misreading this schematic, or is my PT all kinds of jacked up?

 

[tschrama]

R9G and R10G are 100Ohm, right? Just replace them with 220Ohm.

They reference the heater voltage to ground. Take measurements again after you replaced the faulty resistors.

 

[mThomasDutch]

Just to not leave this hanging - I replaced all the failed components and carefully checked all the components again before firing back up with a pair of known good 6l6's. I brought it up slowly on a variac while keeping a close eye on things, then fed it some audio from a looper while continuing to monitor. So far so good. I have the tubes biased at about 23ma right now, but I'll gradually increase as the amp earns my trust.

Thanks to all who offered help her, and thanks Rob for talking with me and looking into the replacement board in case it was needed.

At this point, as unlikely as it seems, the evidence is pointing to one or more bad tubes in two batches. I for one am done with EL-34's as I have had failures in every amp I own.

 

[tschrama]

Thanks for getting back to us, that's how we learn! And good to hear you got it working again.

Grtz,
Thijs

PS
I am a low-bias guy.. I would be totaly fine with 23mA. That's about how Mesa Boogie biases 6L6 in their amps.