RMS and decibels, was Re: RMS

On 12/26/2012 08:38 PM, Chris Tofu wrote: Yes. In fact, the most accurate RMS to DC converters (as components in RMS voltmeters) are "thermal converters", which convert the input energy to heat, then measure how much DC energy is required to heat another heating element to the same temperature. Measuring the area under a complex RF waveform in real time is tough to do. I have an HP 434A Calorimetric Power Meter, introduced in 1958. It uses a recirculating oil system (it's as much plumbing as electronics) to do just that. See a photo and block diagram about midway down this page: http://www.hpmemory.org/wa_pages/wall_a_page_10.htm Later the same basic setup was implemented by Linear Technology in a DIP package, as the LT1088. Some metrology-grade "true RMS" volmeters, like the Fluke 8506A, use a heating element and a thermocouple in a sealed, evacuated glass bulb. Voltages? Not terribly often, but sometimes. dB, however, is a relative measurement...it is not possible to directly represent a voltage in dB. To do so, you could say "dBV", which would be "decibels relative to one volt". It is common (like about 100% common) in electronics to represent RF power in dBm, decibels in relation to one milliwatt". For example, a common doubly-balanced mixer has an input drive level of +7dBm, which is 7dB higher than one milliwatt, or roughly 5mW. 20dBm, or 20dB over 1mW, is 100mW. dB for ratios, like gain or attenuation figures, and dBm for absolute power, is a lot more convenient to work with than (say) mW. -Dave -- Dave McGuire, AK4HZ New Kensington, PA