Bias problem on original JMP 100W (plexi 1959 circuit)

[gkaada]

Hi all,

I have a problem with an original Marshall JMP 100W, (metal front but the early 1959 plexi circuit), made in year 1969. It has not been used for at least 20 years and I am now trying to get it back to life.

The problem: The bias point is drifting (runaway) evenly on all four power tubes when the amp gets warmed up, and would eventually lead to red-plating of all four tubes if the power was not shut off. Below is a table showing measurements of plate voltage, plate current and bias voltage over 20 minutes.

_____________  V , mA , Bias V
After 1 min... 486, 27.1, -41.4
After 5 min... 486, 29.5, -41.0
After 10 min. 482, 38.4, -38.8
After 15 min. 470, 55.2, -35.1
After 20 min. 456, 80.2, -30.5


What I have done after first time of powering on the amp and discovering the problem:

1. Tested all old tubes on a digital tube tester, all tubes tested good however one power tube had slightly higher value than the others, and will probably be a little off in terms of a matched quad in the amp.

2. Replaced all tubes with a known good and perfectly matched set of JJ´s, problem still there.

3. Checked all diodes and resistors in the bias circuit, all reads normal values. Checked the bias pot (0 to 20k ohm), reads normal and seems to work fine, cleaned pot with Deoxit. Checked all screen and grid resistors and they have all normal values. Repeated these measurements after the amp has been powered on for 20 min. and cannot see any significant change in values.

4. Replaced all electrolytic capacitors in the amp, including the bias caps with F&T and Sprague atoms, problem still there.

5. Replaced all coupling caps with Mallory 150´s, problem still there.

6. Replaced the two 0.22uf caps next to the SS rectifier with Sprague orange dops, problem still there.

I have run out of ideas and I am hoping that someone here on this group may understand what is going on.

Many thanks,

[RagnarMarGunn]

Try measuring Bias Voltage (100VAC) as it comes from the PT at the junction of the first resistor in bias circuit (27K resistor). If it drifts significantly downward I would suspect the PT is the cause. I just fixed the same problem in my amp and it was bad 220K bias resistors and dodgy soldering, redo every connection in the bias circuit including ground.