Musicman HD-130 red plating issue

[mdeetz]

Hey Y’all!

Long time lurker, first time poster here. Just wanted to first say thanks for all the great knowledge on this site; it’s been endlessly helpful for my learning process and inspiration!

I’m currently repairing my friend’s Musicman HD-130 head and am running into trouble with some red plating tubes. I don’t see tube number designations in the schematic, so I will refer to them like this (from left to right, looking at the back of the amp):

v1: 12ax7 driver/phase inverter
v2/v3: first el34 pair
v4/v5: second el34 pair

Here’s a link to the schematic:  http://el34world.com/charts/Schematics/files/Musicman/Musicman_2475_130_2275_130.pdf

It’s the last page in the pdf, with the tube PI.

I was told that the amp was plugged into the wrong cab and it blew one of the EL-34s while they were playing it. I tested all the output tubes on my old Seco tube tester (apparently from a local grocery store), which isn’t very accurate but can tell me if things are completely not good. Three of the tubes tested fine and one (v5) was indeed dead. The dead tube had some burns on its socket (only on the bottom side where the tube gets put in) as well, between the plate and the screen grid pins.

First thing I did was replace all the electrolytics. They were all original, and a complete recap has helped me before with these amps. I also replaced the burned tube socket and the 10 ohm cathode resistors just in case.

I’ve seen these musicman amps blow output tubes before and have heard that they can sometimes take the output transformer with them. So:

- I measured resistance between the two primary leads going to the plates and found that it was open.
- I then measured resistance between the blue lead going to the plates of the first pair of EL-34s (v2/v3) and the center tap and it was open.
- Then I measured resistance between the brown lead going to the second pair (v4/v5) and the center tap and I got 113 ohms.

I opened up the transformer and found the break in the wire connecting to the blue lead. There was enough slack that I carefully wrapped the two ends together and quickly soldered them. I measured for resistance and got 105 ohms from blue lead to center tap and 216 ohms from blue to brown. Occasionally, my meter would have a hard time reading and I would need to switch into diode mode and then back to resistance for the number to stop jumping around. Does this indicate anything in itself? My meter is a fluke 77/an.

I used some silicone to cover up the bare transformer winding wires and put it back together with the paper covering that was inside the bell cover. I tested it again before installing into the amp and got the same resistance readings. Installed it back into the amp, and still got the same readings.

Powered it up without tubes to make sure I was getting correct voltages. Plates were all showing around 725, screens at 375, and the grids were all getting negative bias voltage. I turned the bias pot to max negative bias position, powered down, and installed a new set of matched JJ E34L tubes.

I biased them with the recommended method (measuring .5v across one of the 10ohm cathode resistors), but kept them at .45 volts to hopefully be a little bit safer. So, 725v on the plates, .45 volts across the cathode resistor which measured about 5.6ohms with the amp off (it’s two 10 ohms in parallel). Assuming they’re sharing that load equally, that’s about 14w Plate dissipation per tube, no?

Figured I was in a good safe zone. I let the amp sit and watched the voltage across the cathode resistor. Everything was stable after 30 mins so I decided to play some guitar through it. Performed well and sounded great at low volumes for another 30 minutes. Master volume all the way up, everything worked well until about 4 on the vibrato channel. I hit a big chord, the voltage across the cathode grew to 3v, and (just) the second pair of E34Ls started red plating. I turned the amp off quickly.

The amp still idles just fine and all voltages look good at idle. I’m not exactly sure how to proceed at this point. I’m concerned that I may have fixed the connection in the output transformer but perhaps it’s not strong enough to handle the amp when turned up. I’d love to not have to replace the output transformer if possible!

Any advice on where to go from here?

[PRR]

Welcome.

My first thought is: use rugged 6550/KT88 instead of economical EL34. MusicMan really pushed those finals hard. I'm not sure I would trust just any new-made EL34 to take the abuse. I have no idea if anybody actually does that, or if "tone" would change, or if you could be sure when the amp was LOUD.

The OT repair seems reasonable. I assume you can solder old wire. The difference of DCR is typical. This may be easier to read if you *short* the secondary; the high inductance confuses a digi-meter's auto-ranging and it won't find a right range.

> Assuming they’re sharing that load equally

I sense doubt, which I too wonder about. At the risk of "changing" a Classic Amp, I'd be real tempted to replace the common resistor with four separate resistors for individual monitoring, and higher value to enforce better balance. Even 50 Ohms 2 Watts per cathode. This will take a small amount of gain off, but hardly-any (and much less than another blow-up). Your math is good and you can compute the new bias goal. My guess is that at full-wail each tube should only pass 100mA max average, so you should never have over 5V on this resistor.

Hmmm... I might also wonder at the 330K grid resistors on EL34s. Max spec  is 100K per tube, so 50K per side. The cathodyne won't pull 50K, but I might tack 100K-120K across each 330K and see if it is less prone to trouble in hard driving.

Golly, it is usually the later opamp-NPN-driven MMs which are impossible to troubleshoot. This is "old school" design so should not be hard. I do wonder if MM was buying much higher quality EL34s than we get today.

[mdeetz]

Thanks for the insight!

Hm, 6550/KT88 is a fun thought! I have some Sovtek KT88s and some JJ 6550s sitting around that I might be able to try out. Seems like they'll fit width wise, but I'll need to get the bear clamps out of the way. They'd probably like a lower grid leak resistor value too, no? Looking at the datasheet for the JJ 6550s, is has max plate voltage at 600v. Is that accurate or should I just ignore it?

I like the idea of making it four separate cathode resistors. It wouldn't be difficult on this amp and would be very useful. Folks working on it in the future wouldn't be able to do musicman's stock bias instructions, but hopefully they should be able to figure it out or they shouldn't be working on it! I have more of the 10ohm 3 watts I installed. Maybe I'll start there to see what's going on and then up them to 50 when I can get those?

I have time today to work on the amp, so I'll try lowering the grid resistor value and report back. I've read some vague posts around the internet that say the PI can somehow cause the tubes to runaway and blow, but I haven't really found an actual electrical explanation of it. Though, having this thing cranked up, it certainly sounds like the PI is slamming the output tubes!

> The cathodyne won't pull 50K

Can you elaborate more on this? Do you mean that 50k would be too big of a load for it?

> This is "old school" design so should not be hard

That's what I was hoping for! Haven't had much fun working on these in the past, they always just seem full of trouble.

I might try biasing them quite cold today and see how they perform. The post below mentions 12-13ma per tube.

http://www.pacair.com/discus/messages/6/213.html?1031310930

[mdeetz]

Disconnected the cathodes and set up each socket to have an individual 10ohm going to ground. Tubes are pretty close; they're drawing within 1 or 2ma of each other.

Biased them to 12/13ma each with nothing else changed besides the individual cathode resistors and this time the first pair (v2/v3) red plated when the volume was turned to 2 (master volume on 10).

Then I soldered in some 120ks in parallel with the 330k grid resistors. Biased again to 13ma per tube and it held in there! I played it for about 5 minutes with the volume at 3 and nothing ran away.

I'm going to try it again later when I have more time to play loud, perhaps with it biased hotter, and see how it performs. I'd like to make sure it can hang at extreme volumes as that's most likely how it will be used!