Can anyone please show me how to calculate the A-weighted noise power of a noise source.
Consider the case of a resistor having a spectral density of 4kTR.
What is the A-weighted noise power of this resistor?
\$\begingroup\$ \$\endgroup\$
3 
- 2\$\begingroup\$ Can you define what does A weighting mean in your context? \$\endgroup\$Justme– Justme2024-05-12 14:48:34 +00:00Commented May 12, 2024 at 14:48
- 1\$\begingroup\$ Audibility weighting curve perhaps? Regardless, it's just 4kT times the weighting spectral curve. Note that the noise power spectral density of a resistor does not depend on R, just 4kT \$\endgroup\$tobalt– tobalt2024-05-12 16:01:03 +00:00Commented May 12, 2024 at 16:01
- \$\begingroup\$ @Justme I mean the A-Weighting curve commonly applied in audio measurement. \$\endgroup\$Kenlucius– Kenlucius2024-05-13 02:11:53 +00:00Commented May 13, 2024 at 2:11
Add a comment |
1 Answer
\$\begingroup\$ \$\endgroup\$
1 A weighting accounts for the frequency dependence of the human ear. There are numerous useful links on google, including wikipedia which has a transfer function suitable for incorporation into software.
That's all you need, except the scaling factor that accounts for the system between the resistor and the sound-creating device.
- \$\begingroup\$ Thanks for your answer. Is it correct to say the A-Weighted noise spectrum of the resistor is "4kTR x (A-Weighting Curve) x (A-Weighting Curve)"? Here I multiply the noise spectrum of the resistor by the transfer function squared (i.e. A-weighting curve squared) to obtain A-weighted noise spectrum for this resistor, which can then be integrated from 20~20k to get the A-weighted noise power in the 20~20k bandwidth. \$\endgroup\$Kenlucius– Kenlucius2024-05-13 02:21:58 +00:00Commented May 13, 2024 at 2:21