Max,
Q1) Does anyone happen to know how to calculate or estimate the audible noise level of products (A-weighted) before they are built? One assumption that I have made is that continuous acoustical noise from numerous identical devices add together in the same way that power does. IOW, two identical fans, with a 50 dBA rating, would measure 53 dB. Is this correct? Does this represent a worst-case estimate? Or, IOW, in an actual system with 10 fans, for example, would some of the noise cancel out due to phase angle differences in the emissions from the fans? A1) My experience with noise has been that the total noise from multiple sources is going to be the R.S.S. (Root-Sum-Square) value of all the individual sources together. This assumes free-air operation or a substantially similar construction. If you are talking about stacking several systems together then this works well as a 1st cut estimate. Never the less a followup measurement should be performed on the completed system. This method seems to account for all the possible combinations of additive and subtractive noise. However with fans I have had the additional problem of beat frequencies being a very strong irritant with the user of the equipment (no two fans are exactly the same). Even though the noise level is measured and is within acceptable levels the user may complain due to the tonal quality of the fans. Q2) Another question that I have is if emissions at different frequencies add together like power. For example, would 50 dBA "noise" emissions, at 15 Khz, from a violin add together with 50 dBA emissions, at 20 Hz, from an organ to make a total measured emission of 53 dBA? Or would a meter measure only 50 dBA in this situation? Or more specifically, has anyone had any experience with power supply fans and system cooling fans? Does the noise from these two different types of fans typically add together? A2) Typically my experience has been that Product Safety people do not get too involved with the spectrum. Mostly it's broadband levels that are of concern. For mixing 2 types of fans, once again I use the R.S.S method. The products I deal with are power supplies and the fans typically need to remove from 500W to 3kW in power supply losses. My own opinions and experience.... Doug Powell Advanced Energy Industries, Inc.
