I have no argument with the concept of a shunt resistor, but someone at NEMA had their head up their hind end when recommending 500-1000 ohms.
From
http://www.nema.org/engineering/papers/PhanBull88.html - NEMA recommends placing a resistor of approximately 500 to 1000 ohms across high-impedance multimeters.
OK, so 500 ohms across 120V results in a power dissipation of ~29 watts.
And, 1000 ohms across 240V results in power dissipation of ~58 watts.
In either case, it’s very likely that a ½-watt resistor from a Radio Shack blister pack will deflagrate in the process.
Increase the resistor by a factor of 100. Even 50,000 ohms across 240V is a power dissipation of ~1.2 watts, which will cook a ½-watt resistor if left connected for any significant amount of time.
This makes a mockery of kilovolt-impulse withstands in IEC-1010 overvoltage categories, for use with multimeters on all but the smallest of power circuits.
[This message has been edited by Bjarney (edited 03-30-2004).]