[RTTY] 300hz or 500hz IF filter?

[Kok Chen]

Oh, another thing.

You might notice that the 3IMD column don't begin until the audio generator has dropped to -6 dB.

That is because the two tones for IMD testing need to each be 6 dB lower than full scale (i.e., -6dBFS), since summing them gets the same peak voltage of a single tone.

Following common audio measurements practice, I have used dBc (a single IMD component relative to a single tone) instead of dB PEP which the ARRL uses to characterize IMD for amateur transmitters.  The dB PEP vs dBc difference precisely explains the missing first 6 dB -- i.e., when measured relative to PEP, the devices appear 6.02 dB better than when measured with dBc.

The audio world tends to use dBc.  

In the professional RF world, both dBc and dB PEP as acceptable (as long as you mention which one).  But once the ARRL chose dB PEP for transmitters, all the ham manufacturers followed, since it makes their transmitters appear 6 dB better than some competition who uses the dBc number :-).  Pretty much everybody today uses dB PEP in their transmitter specs.

Anyhow, be aware of the dBc vs dB PEP difference.  They are only directly comparable when you lop off 6 dB from one of them.

The other thing in the audio world is that they tend to spec their SNR in terms of A Weighted spectrum.  This is a little different from the 2125/2295 tone pair :-).  

Interestingly, though, my Presonus measurement came in within 0.5 dB or Presonus' A-Weighted dynamic range spec (they claimed 118 dB of A-weighted dynamic range, I measured 117.5 dB for unweighed dynamic range).

73
Chen, W7AY