Here's a mystery from my latest project: I've repaired a Blues Junior (schematic attached), and have it working and sounding fine. However, when I measure the B+ voltage, it's far lower (at 255V) than the schematic from the manual suggests I should find it (329V). Other supply voltages are similarly low by >20%.
Before the rectifier, I find 256VAC across the PT secondary as expected, and measure 130VAC at the junction of D1 and D3, which matches the schematic.
Pulling the preamp and PI tubes raises all the voltages a little bit, but doesn't meaningfully change the situation. Pulling the power tubes as well leaves all supply voltages at 350V. So, seems to me the power tubes are pulling the voltages down, and I imagined I'd find them running hot.
But no: bias voltage measures -10.6V just as the schematic suggests. I measured plate currents via OT resistance & voltage, and found 15ma (1.41V/91.6R) and 14ma, which seems quite low for EL84s — not at all what I expected! Switching to a different pair of EL84s didn't make any significant difference. Just in case something was wrong with my plate current measurements, I tried paralleling some resistors across the bias supply voltage divider (R51 and R52), to drive the bias to a very cold -16V, but even this left B+ at barely 300V.
In case it's a helpful clue: the problem I found and repaired was that the first voltage dropping resistor in the power supply, R47, had burnt out. The replacement I installed (2.2k 2w) runs warm, 48C according to my IR thermometer, but doesn't seem to be in danger of a similar fate.
Any ideas on what might be going on? Given that the amp sounds fine and the power tubes seem to have low dissipation, I'm tempted to call this a success, but I'm worried I may be overlooking something that will cause problems.
Before the rectifier, I find 256VAC across the PT secondary as expected, and measure 130VAC at the junction of D1 and D3, which matches the schematic.
Pulling the preamp and PI tubes raises all the voltages a little bit, but doesn't meaningfully change the situation. Pulling the power tubes as well leaves all supply voltages at 350V. So, seems to me the power tubes are pulling the voltages down, and I imagined I'd find them running hot.
But no: bias voltage measures -10.6V just as the schematic suggests. I measured plate currents via OT resistance & voltage, and found 15ma (1.41V/91.6R) and 14ma, which seems quite low for EL84s — not at all what I expected! Switching to a different pair of EL84s didn't make any significant difference. Just in case something was wrong with my plate current measurements, I tried paralleling some resistors across the bias supply voltage divider (R51 and R52), to drive the bias to a very cold -16V, but even this left B+ at barely 300V.
In case it's a helpful clue: the problem I found and repaired was that the first voltage dropping resistor in the power supply, R47, had burnt out. The replacement I installed (2.2k 2w) runs warm, 48C according to my IR thermometer, but doesn't seem to be in danger of a similar fate.
Any ideas on what might be going on? Given that the amp sounds fine and the power tubes seem to have low dissipation, I'm tempted to call this a success, but I'm worried I may be overlooking something that will cause problems.
You found the resistor burnt out, which means it was getting hot at one time. Any chance the solder joint on the first filter cap was bad due to the overheating, and that it got reflowed into working when the resistor was replaced? With the cap disconnected, you would measure an average voltage almost the same as the RMS secondary - if your meter is like most inexpensive ones (not true RMS). It would have produced one hell of a hum thru the speaker if that’s what was going on - but you might not notice using a dummy load and just measuring voltages. If that’s what it was, I’d re-solder the cap for sure to make sure it stays working.
Then you are measuring wrong or your meter is defective.
256VAC across secondary means rectified 362V DC
Indicated 329V DC under load is reasonable.
255V DC is NOT.
You can not have same AC and rectified DC, period.
OR, series resistance is high.
You have terrible contact quality somewhere.
You sure you are not measuring at the far end of R32 by mistake?
This makes sense. Whatever is dropping the DC voltage should be pulling down the AC on the other side of the diodes as well. But, I've confirmed it's not (well, only slightly — it drops to about 250VAC as the tubes warm up).
And yet, I have continuity between the rectifier outputs and the first filter cap on one side and ground on the other. No sign of a series resistance there. I tried jumpering the connection just to be sure, and that didn't change anything.
I'm sure I'm measuring the correct end of R47 (the first voltage-dropping resistor is labelled R47, not R32 — I should have specified that my board matches the *second* schematic in the file I attached, very similar to the first schematic but with different labels) because I've measured both ends. One end is at 255V, the other is at 240V.
Could the rectifier diodes themselves be introducing a series resistance, or failing to fully rectify? I checked them and they all show normal voltage drops on my meter. My next thought was to measure AC across the outputs of the rectifier — before the tubes start conducting my meter gives me "this is DC" beeps, but once they warm up, I'm seeing 60VAC on the rectifier outputs. So, now I'm suspecting problems with either the rectifier diodes or maybe the first filter cap.
And there it is: I paralleled a 22uF 450V cap across the first filter cap and B+ now holds at 300V. Not all the way there, but much better. The original cap looks fine and measures fine, but I guess it needs to go.
These amps are notorious for two things.
Bad solder joints, and bad filter caps.
If the caps in the main section of the power supply are Lelon, replace them all.
That's probably why the 2.2k burnt.
Bad solder joints, and bad filter caps.
If the caps in the main section of the power supply are Lelon, replace them all.
That's probably why the 2.2k burnt.
Yes, I second this... The caps Fender uses are absolute crap. They typically last 2-3 years and they go out. A telltale sign is how hot the PT is when running. The PT normally run hot in these amps but if you can't comfortably touch the PT, then it's too hot.
The brand I usually see in these amps is IC or Illinois Capacitor. They are worthless; change them all out and your amp will be much happier.
The brand I usually see in these amps is IC or Illinois Capacitor. They are worthless; change them all out and your amp will be much happier.