Question about Arithmetic Function of All SPL

[eepty]

I want to measure the sound spectrum of a hair dryer. So I did two RTA measurement: With (A) and without (B) switching on the hair dryer.
And then in the "All SPL" graph I do the (A) minus (B) and generated the third graph. I hope this more or less represent the spectrum of the hair dryer. However I got a very weird graph. Please see the attached saved measurement.

Why there seems nothing at the higher frequency part in the A-B graph?

Thank you very much.

 

[John Mulcahy]

Try using the current beta version. Above the region in your measurements which is just background noise (about 250 Hz) A-B is pretty much identical to A though - don't forget the level scale is logarithmic so you'll only start to see differences where the background gets within a few dB of the measurement.