Odd voltages in P1 Extreme

[simonallaway]

I turned on my amp yesterday to find it producing no sound at all. I checked the speaker and all the tubes with another amp, and they are functioning. So I thought I'd first gather all voltages and post them here as there seems to be something odd going on with V4A (the second preamp stage). Here's all of them:

V4B
Pin 6 210v
Pin 8 1.5v

V4A
Pin 1 310v
Pin 3 10v

V1
Pin 3 362v
Pin 4 363v
Pin 8 24v

B+ 374
B+ 1 369v
B+ 2 367v
B+ 3 317v

AC 124v

So V4A has seemingly way too much on the plate (310v rather than 136v) and voltage at the cathode is way too high. I swapped the new JJ 12ax7 for an old (but good) Amperex from my Marshall and there was little difference. I also checked the resistors R3 and R4 out of circuit, and they read within 5% of their values. I also checked R24 and that one's ok too.

So do I have a short somewhere between B+2 and B+3? Or maybe it's the C4 filter cap? I'm not sure how to check capacitors much further than a resistance check (C12 seemed ok).

Schematic: http://www.ax84.com/static/p1x/AX84_P1x_101004.pdf

I appreciate any direction you guys can give me. Thanks!

[sluckey]

Measure the resistance from V4-3 to ground. Put one lead directly on pin 3 and the other on the chassis where your HT center tap is connected.

Most likely causes for your symptoms are R19 has increased in value or you have a poor ground on R19.

[simonallaway]

Thank you. I will check that.

[simonallaway]

Is it true to say that with 10v at the cathode, the tube is biased incorrectly and is therefore likely to be in full cut-off? Or in other words, the grid is -10v in relation to the cathode so no current can flow to the plate? (and the guitar's tiny voltage isn't enough to get anything to move).

[sluckey]

Is it true to say that with 10v at the cathode, the tube is biased incorrectly and is therefore likely to be in full cut-off?

No. The grid 'may' be +20V and the tube would be hard on. IOW, you must also know the grid voltage to determine how the tube is biased. Bias voltage is really the difference of voltage between the grid and cathode. You must know both. That's why it's important to post grid voltages too, even if they are zero.

Or in other words, the grid is -10v in relation to the cathode so no current can flow to the plate? (and the guitar's tiny voltage isn't enough to get anything to move).

That's a better statement since you're now talking about the voltage between the grid and cathode.

Specifically, for your situation, 10v on the cathode indicates that there is current flowing thru the tube. For your circuit, there is no other way for 10v to be on the cathode. (Unless something is wrong!) But the high voltage on the plate indicates the tube is not conducting. So, which to believe? It really can't be both ways, can it? Well, it can if something is wrong. You gotta look for something that can cause both symptoms.

Think about this... 10v across that 600Ω cathode resistor means 16ma is flowing. That's way too much for that 12AX7, but don't think about that right now. That 16ma is also flowing thru that 100KΩ plate resistor and that means there should be 1600v dropped across the plate resistor! Can't be. But what if the cathode resistor had increased to 100K. Then, 10 volts would mean that only 0.1ma was flowing. There would only be 10 volts dropped across the 100K plate resistor too, so the plate voltage would be very high.

So, an increase in cathode resistance could cause both the cathode voltage and plate voltage to be high. That's why I wanted you to check that cathode resistance.

[EL34]

Simon, re post 1:

Looks like a great place to test the new voltage program.

The seed catalog is getting boring.

[PRR]

> test the new voltage program.

Here's the link to the Valve Data program
http://www.el34world.com/charts/valve/ValveData.php

[simonallaway]

Looks like a great place to test the new voltage program.

It did cross my mind to use that, but I thought you were still tweaking it.

Here's the voltages from my first post.
http://www.el34world.com/charts/valve/ValveData.php?e=view&f=10130.csv

That's VERY cool! I'm still not used to the visual representation of the pins as my mind holds an image of the schematic view of a tube. But I ought to get used to memorising the pins, so this is a great learning tool for me.

[sluckey]

Did you fix it yet?

[simonallaway]

Think about this... 10v across that 600Ω cathode resistor means 16ma is flowing. That's way too much for that 12AX7, but don't think about that right now. That 16ma is also flowing thru that 100KΩ plate resistor and that means there should be 1600v dropped across the plate resistor! Can't be. But what if the cathode resistor had increased to 100K. Then, 10 volts would mean that only 0.1ma was flowing. There would only be 10 volts dropped across the 100K plate resistor too, so the plate voltage would be very high.

So, an increase in cathode resistance could cause both the cathode voltage and plate voltage to be high. That's why I wanted you to check that cathode resistance.

That's a great explanation, thank you. I didn't consider the whole circuit, so I just tested the components in isolation.

I checked the resistance from the cathode pin to ground (which was OK) but discovered a dodgy solder joint to ground. So I removed old solder and re-did it. Now it's fine. Voltages are back to within 5% of the schematic and I have a functioning amp again, and I like to think I learned something.

Thank you for your patient assistance. I really appreciate it.

[simonallaway]

  A light-bulb just went off in my head. I think this may be the resolution to this previous thread where I was getting random silence in the same amp, and could never really reproduce it. The thought at the time was oscillation, so I ended up moving the grid resistor on the output tube.

Is it reasonable to blame the dry solder joint I found/fixed in this thread?

[sluckey]

Is it reasonable to blame the dry solder joint I found/fixed in this thread?

Maybe. Exactly where was the bad ground?

[simonallaway]

Is it reasonable to blame the dry solder joint I found/fixed in this thread?

Maybe. Exactly where was the bad ground?

Right where the cathode cap/resistor connects to ground. I re-soldered the areas inside that box.

[EL34]

Thanks for the voltage chart test Simon.
Check it out now, I combined the graphic and table data on the same page

Here's the voltages from my first post.
http://www.el34world.com/charts/valve/ValveData.php?e=view&f=10130.csv

[simonallaway]

That is really cool :)

What happens if:

1) I want to modify the values I originally typed?

2) I just want to delete it?

[EL34]

I would just create a new file.

There's no way to edit or delete right now.

Don't think I can add a delete feature or some a-hole could just delete all the files.
May be able to add an edit feature later on.

[simonallaway]

Don't think I can add a delete feature or some a-hole could just delete all the files.
May be able to add an edit feature later on.

That's a good point. And the only way to fix that is to force people to log in, which then makes it less "available", which isn't desirable.

[HotBluePlates]

You might want to add what the B+ is in your amp feeding the plate load resistor of that stage (maybe that should be added to the program).

The reason is twofold:
If the B+ was available, then we would have seen the very small voltage drop across the plate load, and therefore the very small current.

The second reason is that folks who aren't familiar with this troubleshooting technique might find it instructive. What Sluckey demonstrated was that if the cathode resistor value was right, the very high drop across it means very big current through the cathode resistor, tube and plate resistor. But then Sluckey took the apparent indicated tube current and multipled by the value of the plate resistor, and found that the voltage drop implied by the current and a correct plate resistor value greatly exceeded the value of the B+.

So the end result is the resistance from cathode to ground had to be much too high, or the plate to B+ much too low. A follow-on check with the amp off found the cathode circuit was the culprit, with the poor solder joint being neither a short circuit or an infinite resistance.

[simonallaway]

You just have to set the field "a-hole" to "false" everytime 

[simonallaway]

You might want to add what the B+ is in your amp feeding the plate load resistor of that stage (maybe that should be added to the program).

If you mean "add B+ to the cool new pin voltage web app" then I agree. It crossed my mind, but I thought that was just because I am so new to this that they might not be necessary, but I put them in my first post though just in case.

The reason is twofold:
If the B+ was available, then we would have seen the very small voltage drop across the plate load, and therefore the very small current.

The second reason is that folks who aren't familiar with this troubleshooting technique might find it instructive. What Sluckey demonstrated was that if the cathode resistor value was right, the very high drop across it means very big current through the cathode resistor, tube and plate resistor. But then Sluckey took the apparent indicated tube current and multipled by the value of the plate resistor, and found that the voltage drop implied by the current and a correct plate resistor value greatly exceeded the value of the B+.

So the end result is the resistance from cathode to ground had to be much too high, or the plate to B+ much too low. A follow-on check with the amp off found the cathode circuit was the culprit, with the poor solder joint being neither a short circuit or an infinite resistance.

I'm going to have to work through that diagnostic part again with Ohm's-Law so I get it. The fact that my amp broke and I got to attempt a fix was a golden opportunity. It's kind of boring when it works. This has put a few pieces of the puzzle into place 

[EL34]

Delete is not going to work on it's own.

There is no way to log someone in with user names and passwords without making this whole thing a really complicated program.

Here's what I think will work.

You view/load the existing file.
You edit the existing values
You check a box to save the new file and delete the old file
Then the new edited file is saved and the old file is deleted.

This keeps someone from a rampant delete session because you have to put some effort in to edit and save the new file.