AES17 notch vs AP

moto

Registered
Thread Starter
Joined
Aug 13, 2020
Messages
7
Does the AES17 notch for calculating the strength of a fundamental tone match AP software?
I read the description in REW of the 2 notch types but don’t understand how the compare to other software.
I see how a 1octave wide notch might make sense at say 1kHz but a 10k wide notch when the fundamental is 10kHz seems odd.
 

moto

Registered
Thread Starter
Joined
Aug 13, 2020
Messages
7
@John Mulcahy does it really make sense with these thd +n calculations for all audio situations. For instance when you use a notch to filter out say 1kHz then run it through REW with a cal correction?
Is say a 30db reduced fundamental after the notch filter close enough to a dac’s noise floor(call it -100db) to warrant aes17 option in REW?
Using the AES17 option can artificially increase the fundamental strength and therefore the snr, etc right?
 

John Mulcahy

REW Author
Joined
Apr 3, 2017
Messages
5,554
Yes, hence the passage in the help, though if the cal file restores the pre-notch levels it may still be appropriate:

If the Use AES17-2015 standard notch option is selected the fundamental power for THD and THD+N calculations will be the power within a one octave span around the fundamental frequency. If that option is not selected it will be the power in the main lobe of the fundamental. When the fundamental level approaches the noise floor using the standard notch will produce a much higher figure than the fundamental main lobe, in those cases it is better not to use the standard notch.
 

moto

Registered
Thread Starter
Joined
Aug 13, 2020
Messages
7
Oh I stupidly forgot that the REW notch options are applied after the cal file adjustments. Thx
 

John Mulcahy

REW Author
Joined
Apr 3, 2017
Messages
5,554
The main lobe is the portion of the response that is monotonically decreasing from the peak. I have a todo item to implement a way of highlighting what contributes to the fundamental, I'll see if I can get that done soon.
 

moto

Registered
Thread Starter
Joined
Aug 13, 2020
Messages
7
Appreciate the response. It would be interesting to characterize the effect of one notch vs the other on thd+n. No way for any closed form solution to do that though.
 

John Mulcahy

REW Author
Joined
Apr 3, 2017
Messages
5,554
The effect on the fundamental power is negligible unless the fundamental is very low. The main concern is to make sure that any skirt of the fundamental is not included in the N+D calculation, since that would artificially raise their contribution. Window choice is a factor there, if the stimulus is not chosen to be centered on an FFT bin.
 

moto

Registered
Thread Starter
Joined
Aug 13, 2020
Messages
7
@John Mulcahy sorry but I just wanted to make sure I understood something above. Is the aes17 or standard notch applied after or before the .cal corrections are made to the data?
 

John Mulcahy

REW Author
Joined
Apr 3, 2017
Messages
5,554
All calculations are done on data after applying cal file corrections, though I'm not sure what difference that would make. The selection for the fundamental defines the frequency span for the data that is considered part of the fundamental.
 
Top Bottom