Should filter resistors be included in DC load line calcs?

[Paul Lebow]

Maybe this is a silly question, but in the explanations I've seen for calculating DC load lines, the filter resistors in the node chain seem to be ignored. In some circuits those resistances aren't insignificant. Usually just the anode load and cathode resistors are taken into account. What am I missing?

[LOUDthud]

DC load lines drawn on the plate curves you see in a data book are ok for finding a bias point, but you may need to fine tune them in real life. You may set a target plate Voltage of 200V, but in a real life circuit, the plate Voltage may be +/- 10% or wider even if B+, load resistance and bias are right on. If you think a signal will just move the plate Voltage along the load line you drew, I'm sorry to say, you are mistaken. If you setup a circuit with a way to measure plate current and Voltage, in most circuits you won't get a straight line, you will get an ellipse or a circle due to capacitors in the circuit.

Although decoupling resistors and filter capacitors have an influence, the effect is at a frequency below the audio we a trying to amplify. If you are that worried about it, get a modeling program and build a model in software.

[Bergheim]

In regards to the AC produced by the signal through the tube, the dropping resistors in the B+ line actually are pretty insignificant because of the filter cap. The -3dB point of a filter node consisting of a 10k resistor and 22uF cap is 0.7hz, so for all frequencies above 0.7hz, the tube doesn't "see" anything upstream of that filter cap, as everything above 0.7hz is shorted to ground via the cap. A typical dropping string is a choke (100R) or 1k resistor for the screens, a 4.7k/10k PI node resistor, and one or two 10k's for the preamp, so a "maximum" of ~31k.
If we for some reason had to calculate a loadline for the sub 0.7hz frequencies in a gain stage that has a 100k plate resistor, the loadline would be 100k + 31k = 131k. A 31% gain difference is barely bigger than what you might find between two brands of a 12ax7, and in this case we're talking frequencies far below what the speaker can reproduce.

[Paul Lebow]

Thanks. My question was more focused on the DC quiescent point calc.

[TUBEDUDE]

There's no path to ground through a capacitor for D.C. so it won't be a factor for.D.C.

[TUBEDUDE]

Deleted.