Fw: Re: Fw: Re: [Test-Equipment] Spectrum Analyzer BW measurement

[WolfBob]

True RMS is most difficult to measure without using a bolometer or other 
thermal detection device. Envelope distortion in the time domain leads 
to all kinds of assumptions in how long an integrator (average reading) 
or how many peak readings need to be processed to get to RMS. Fun 
problem in instrumentation. In the frequency domain, finding all the 
harmonics and other crap that make up the signal and them combining them 
back together as sine waves whose average to RMS relationship is known 
is another nightmare.

WBob

John Miles wrote:

> If there were any justice in the world, this book
> (http://www.amazon.com/exec/obidos/isbn=1884932169) would be featured on
> Oprah.  Robert Witte is an HP engineer from their analyzer group, and he's
> great at explaining topics like this.  Highly recommended.
> 
> -- john KE5FX
> 
> 
> 
>>	For most spectrum analyzers what you are saying is roughly true.
>>What is typically measured is the low pass filtered output of an
>>envelope  (or sometimes quasi RMS) detector looking at the output of a
>>log IF amplifier.   Some very modern instruments sample the IF signal
>>and use DSP processing and can measure true RMS in some situations,
>>however.
>>
> 
> 
> 
> 
>