B &K Sound Level Meter

[donquixote]

I'm using a Bruel and Kjaer 2209 sound level meter to take some measurements, but I don't have the manual. Does anyone know what the difference between 'peak' and 'impulse' is? Also, the SLM has six weighting networks, 'A', 'B', 'C', 'D', 'lin' and the external filter. I know only generally what 'A', 'B', and 'C' look like, and I'm not sure what 'D' or 'lin' look like at all (I can venture a guess that 'lin' is a flat eq). Does anyone know the exact curves for these networks? For the external filter, I have a dial that I can choose what frequency is weighted, how much does the filter boost a chosen frequency? What do some boosted frequency curves look like (ie. are they peaked or is the slope gentle)?

Does any one have a manual or know of a site that I could find out this information?

thanks in advance for your help.

Joe

 

[VibrationSpecialist]

Dear Joe,
When the sound to be measured consists of isolated impulses or contains a high proportion of impact noise, then the normal F (Fast) or S (Slow) responses are not sufficiently short to give the measurement that is representative of the subjective human response. In this case you need to use the Impulse mode to be able to detect and display this transient noise.
The D-weighting approximately follows a contour of perceived noisiness. It is used for single event such as aircraft noise measurement.
Please refer to B&K book titled Acoustic Noise Measurement.
;-)

 

[estarg]

The standard ANSI S1.42-1986 gives the poles and zeros of the A, B, and C transfer functions. I don't have any inormation on the D weighting function.

Regards.

 

[GregLocock]

Um, you'll be very lucky to keep a 2209 going, they were not repairable 15 years ago, to my disgust, since they are the best SLMs ever made. (However I see some people still rent them out, maybe we should have tried harder to get ours fixed).

I haven't got access to a D weighting curve at the moment but when I get back to work I'll post it. It is only used for airport noise and is somewhat biased in the airlines favour (imho). I've also got formulae for A, B and C.

Lin is indeed flat weighting, I think 2 Hz to 20 kHz.

The filter curves are to ANSI spec, they are flat topped third or full octaves. If its got 11 frequencies then they are bandpass octaves, centred on the indicated frequency. The nominal gain is 0 dB. The roll off rate is something like 22dB per 1/3 octave, from memory.








Cheers

Greg Locock

 

[estarg]

In my reply of 27 Dec 2002, I overloked the fact that the appendix of standard ANSI S1.42-1986 gives the D and E weighting functions also.

 

[GregLocock]

In the Australian Standard AS1259-1982 "Sound Level Meters", Appendix A gives the formulas for A, B, C and D weighting.

The formulas as used in an Excel spreadsheet are as follows, for a frequency in B11. The weightings are in dB.

A weighting

=10*LOG10(B11^8/((B11^2+20.6^2)^2*(B11^2+12200^2)^2*(B11^2+107.7^2)*(B11^2+737.9^2)))+165.45

B weighting

=10*LOG10(B11^6/((B11^2+20.6^2)^2*(B11^2+12200^2)^2*(B11^2+158.5^2)))+163.62

C weighting

=10*LOG10((B11^2/((B11^2+20.6^2)*(B11^2+12200^2)))^2)+163.51

D weighting

=10*LOG10(B11^2*(B11^4-997.53^2*B11^2+1018.78^4)/((B11^4-2819.73^2*B11^2+3136.45^4)*(B11^2+282.7^2)*(B11^2+1160^2)))+83.23


I have not checked the D weighting curve, but have checked the others against Beranek, and they are correct.
Cheers

Greg Locock