[stevedee]

I'm very sceptical about meter loading affecting voltage measurements in typical valve circuits, especially for commonly quoted test points like supply rails, anode and cathode Voltages.

The reason old service sheets specified a 20,000 Ohm per Volt meter back in the day, was because some people had very low resistance volt meters back then. So this would have made a difference to readings, but modern digital meters are high impedance.

If you are reading an anode voltage of (say) 210V with a 20,000 Ohms per volt meter (i.e. one with a 50uA movement), you must be using a Volt range of 250V or higher. 20,000 x 250 = 5M Ohms. I doubt that putting a 5M Ohms resistor between anode and chassis would make any difference in most wireless circuits. Or to put it another way; anode & cathode currents are usually several mA, while an AVO 8 generally draws less than 50uA.

However, you may need to be careful with grid voltages (I think only screen grid voltages are generally quoted) as a few uA of additional load could make a big difference.

So whether you are using an AVO 8 or a high impedance digital meter should generally not make much difference (ignoring calibration error), and the difference is likely to be less than the natural spread of readings when comparing one good valve with another.