Good questions.
I did check the resistors (all 1% precision wirewound) from a lot of 20 purchased from the same batch. They all tested 1.1 ohms at the resistor bodies using the same meter and probes, which means the probes and their wires became part of the measurement.
At such a low resistance (1 ohm), the only way to select exact matches is to string 10 or more together in series, connect them to a regulated power supply and measure the voltage drop across each resistor. Using Ohm's law, the exact resistance of each resistor can be calculated from the individual voltage drops measured.
But given the inherent resistance in the probe wires, the test lead wires and the meter itself, I felt that laboratory precision was unnecessary for this test (I bought more than 4 sets of sockets just in case). because I will be measuring the voltage drop at the amplifier's working voltage-not the operating voltage of any meter. And keep in mind, these are all 1% precision resisters from the same batch, as long as they are all within tolerance and remain there, I think we're fine.
So to keep the above noted factors equal between test socket sets, I used the same lengths of wire from socket to plug to minimize these issues.
Now..the real reason I wasn't going for "laboratory precision" while measuring and calculating the current passing through each tube's cathode was because I only wanted to see if the tubes in each pair were more or less equal and passing a current that was in the same area code. When I use four meters, I'll know if I have a matched quad or if it's necessary to split the bias supply into left and right channels.
Keep in mind-The Fisher instructions are to adjust bias to 7.8 vdc at pin 8 of each tube. That's pretty general, especially with an allowable range of 7.3vdc to 8.3vdc. I'd rather do this with a 1 ohm resistor in series than 10ohms. Because I used stackable plugs, I can check each tube individually, which might indicate an internal issue if the voltage measured was higher on one channel or tube. Another reason I was waiting to see if splitting the bias supply was necessary.
Now, for DC Balance-
I can place a third meter from positive to positive on each meter's positive plug (another reason why I used stackable plugs) so that I can accurately adjust the DC balance to a minimum value while keeping an eye on the total bias currents passing through each tube in that pair. If I'm interested in DC balance alone I can simply connect the common (blue) wires together (the great thing about stackable plugs) and use the two red wires with one meter as I adjust the DC balance to zero voltage. In this (preferred) manner, the cathodes from each tube in the tested channel are tied together through 2 ohms (1 ohm within each socket) then back to the common cathode resistor with capacitor bypass to ground. Again, I thought it was better to add 2 ohms instead of 20 to this circuit.
I see your point, but I thought even if I used 10% resistors, adding 2.2 ohms to the circuit was better than adding 22 ohms. Not that I think any of it would really matter anyway.
To answer another point you raised, I have not yet swapped sockets and meters at the same tube socket positions to verify accuracy....but I will. If I note anything significant-I'll add a +/- notation to each test socket.
I want to make one thing clear-the reason I thought using external test sockets was the better way to go was the flexibility they afford and that I can test and adjust this or another amp's DC balance after it has warmed up without having to remove the bottom cover, etc.. I can run the amp for an hour with the tubes in their test sockets, each socket connected to it's own meter and jumper the plugs at each meter thus removing the meter from the circuit (final reason for using stackable plugs). After the amp has warmed up, I can pull the meter bypass jumpers and set the bias voltage, quiescent current and DC balance. And I don't have to move around probes, keep my eyes on more than one place, or otherwise risk an explosion.
How simple is that?
One more thing-I decided to put the 1 ohm resistors in the test sockets to measure voltage drops rather than breaking the circuit to measure the current through a meter after I read about the possibility of causing an amp to oscillate using the latter method.
Last point--some people adjust bias and balance very successfully by ear. What does that say?
