I checked the manual for my HP 334A Distortion analyzer. The notch filter in that unit seems to be capable of 80dB, and the bandwidth is quite narrow, under 0.01%. This unit just removes the fundamental and measures whatever is left.
I found several good articles and explanations of SINAD measurements and equipment on the web. The simplest description shows an RMS meter at the input, a 1kHz notch filter, and another RMS meter at the filter's output. You adjust the input RF signal until the two meters show 12dB difference, and you've got the 12dB SINAD sensitivity. Some use an AGC amplifier in the input and do away with the first meter. I would think that the distortion analyzer does essentially the same thing. In the "SET" mode, you measure the 1kHz tone amplitude and adjust the input level for a specific value. In the "DISTORTION" mode, that tone is notched out and the residual signal is measured. This could probably be done with an audio spectrum analyzer if you could measure and sum (RMS) all the signals except the fundamental 1kHz tone. If the SA has that function, you should be good to go. I don't know if this would give the same reading, but one time I tried feeding in a 1kHz tone at 3kHz deviation with a full-quieting signal, and noted the amplitude of that signal on an RMS voltmeter. I then removed the modulating tone and reduced the RF level until the meter read 12dB below the level with the tone. I didn't notch the 1kHz tone, I just turned it off. This would leave just the noise, so I suppose any distortion produced by the IF and detector stages was eliminated from the measurement. It would be an interesting experiment to compare this measurement to what a real SINAD meter shows. Bob M. ====== --- "Bob M." <[EMAIL PROTECTED]> wrote: > I'd think it might, but the AC voltmeter is > measuring > the average (or RMS-calibrated) signal that remains > after the 1kHz tone has been removed. I suppose it > only has to be attenuated by at least 12dB (probably > more) so it doesn't contribute to the rest of the > signal. But if the 1kHz tone level is measured at > the > peak, then the noise can probably be measured the > same > way. > > Someone with both setups should do an experiment and > see what the correlation is. > > Another test would be to go for 20dB quieting, which > is easily measured, and see what the audio SA shows > for the noise amplitude compared to the unsquelched > noise. > > Bob M. > ====== > --- DCFluX <[EMAIL PROTECTED]> wrote: > > > Wouldn't the SINAD be the distance between the > 1kHz > > tone "Spike" and > > the noise "Grass"? > > > > On 3/6/06, Bob Dengler > > <[EMAIL PROTECTED]> wrote: > > > At 3/6/2006 09:38 AM, you wrote: > > > > > > >Ok, I wasn't thinking in this direction before, > > but I do have an audio > > > >band spectrum analyzer handy. > > > > > > > >Question is, how can I translate this to a > Sinad > > measurement? > > > > > > I've been trying to figure this out too, as I've > > been working on an audio > > > spectrum analyzer program & wanted to throw in a > > SINAD meter. > > > > > > Subsequent post from Bob M.: > > > > > > >With that audio spectrum analyzer, adjust it > for > > full > > > >scale on the fundamental, and look at the noise > > and > > > >any harmonics. Increase the RF signal level > until > > this > > > >drops to 25% (1/4) of the level of the 1kHz > tone. > > This > > > >will be the 12dB SINAD point, assuming that the > > > >wideband noise etc doesn't add up to more than > > the > > > >-12dB signal amplitude. Once you calibrate your > > > >equipment with a real SINAD setup, you should > be > > able > > > >to continue using the audio SA. > > > > > > So I guess I could remove the 1 kHz data from > the > > FFT'd spectrum, do an > > > inverse FFT & ratio that result with the > original > > 1 kHz signal to get > > > SINAD. Or maybe just add the values of all the > > remaining frequency bins, > > > but somehow I think that may not yield the same > > result. > > > > > > ? > > > > > > Bob NO6B __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/Repeater-Builder/ <*> To unsubscribe from this group, send an email to: [EMAIL PROTECTED] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
