Jim: Measuring absolute true RMS filament V is not needed. What is needed is: 1. something that will indicate a 2% increase in filament V above the point where relative P out just begins to decrease. 2. a means of adjusting filament V. . . . However, amplifier mfgs have a negative incentive to have rheostat-adjustable filament-V because allowing the user to optimize emissive life is going to reduce replacement tube sales. . The most common cause of 3-400Z and 3-500Z tube failure that I have seen in the last 40-yrs is filament to grid shorts from a sudden pulse of EMF. QST Magazine Photograph, September, 1990, p.15 This condition usually comes with a flash of light / big-bang - the energy for which comes from that stored in the filter caps. .. In my opinion, intermittent VHF parasitic oscillation is the cause. Avoiding this problem is not difficult: 1. Retrofit a glitch R in the HV+ to limit peak discharge-I from the filter caps to <200A and 2. decrease VHF amplification by lowering the Q of the VHF parasite suppressor(s). The worst quality 3-500Zs and 8877s I have ever tested/autopsied were made by Eimac in Salt Lake City in the 1990s. When Louis Pasteur suggested that surgeons in Paris hospitals should scrub before cutting he was ridiculed. . Amperex made pretty good, pretty high Mu 3-400Zs and 3-500Zs. One of the Amperex tubes in my SB-220 was made in 1967. However, Amperex 3-500Zs are no less subject to bent filament damage from VHF oscillation than any other brand of tubes so you still have to retrofit a glitch-R and lower the Q of the VHF suppressors unless you have a free supply of new tubes. It is my opinion that the people who design amplifiers are not infallible.