 |
 | Home | Contact Us | FAQ | Search & Site Map | Link to Us |
 | Sign In | Join | Other 45 Sites in Network |
Natural Science Forum / Physics / Acoustics / April 2006Semi-Accurate Measurement of <20dB Levels
I'm looking for a good way to measure the noise levels of quiet computers in dB(A). Since it is very expensive to get a decibel meter to measure levels that low (and even with no budget limitation, I can't even find one!), I'm looking into alternatives. I'm thinking about just using a parabolic amplifier for our microphone, and then calibrating to known levels. Am I just missing something, or is that a solid route to getting the measurements I need? I don't need absolute precision, but need it close enough that I can at least have confidence about the numbers I get! Thank you, Jon Bach Puget Custom Computers ------------------------------------ http://www.pugetsystems.com ------------------------------------ I no expert, so maybe I should not comment, but it would seem to me that you could move the decibel meter closer to the source and then used a formula to calculate the decibels at a farther distance. DaveL > I'm looking for a good way to measure the noise levels of quiet > computers in dB(A). Since it is very expensive to get a decibel meter [quoted text clipped - 15 lines] > http://www.pugetsystems.com > ------------------------------------ Very true....but measurements are usually taken from 1 meter...with that plan I worry about getting a measurement of just a component of the computer instead of the computer as a whole. Though I suppose I run the same risk with the parabolic microphone, if it has a well defined "target" area. Maybe only testing it will tell me for sure! Kind-of an expensive test though, if it doesn't work well. Which frequency are you hoping to measure less than 20dB? It should not be difficult to find a SLM with this kind of sensitivity, although it would have to be a proffessional quality one. You could hire the meter if it was only a one off measurement. Tom Whatever frequencies you can hear :) It is intended to be a measurement of the quietness of the computer...pretty subjective. > I'm looking for a good way to measure the noise levels of quiet > computers in dB(A). Since it is very expensive to get a decibel meter > to measure levels that low (and even with no budget limitation, I can't > even find one!), I'm looking into alternatives. In addition to finding an SLM/mic combo that can measure that low (you should be able to find 1-inch mics with a self-noise lower than 10 dBA) you also need to have an *environment* where the ambient noise levels are lower than 20 dBA. Do you have access to an anechoic chamber or some other area where the ambient noise levels are that low? Herb Several products are the market that will measure to that level. The big problem I am detecting is that you are not familair with what you hoping for entails. The usual way is to determine total power level using a REVERBERATION chamber, and then measuring the directivity of the sound using an ANECHOIC chamber. Aside from their name descriptors, the chambers themselves are very quiet. All three things (live acoustics, dead acoustics, and quiet acoustics) are expensive. Good luck! Andre "avare" <avare@hotmail.com> wrote in news:1144175728.069525.273960 @u72g2000cwu.googlegroups.com: > Several products are the market that will measure to that level. The > big problem I am detecting is that you are not familair with what you [quoted text clipped - 6 lines] > Good luck! > Andre Definately go with a sound intensity map of the whole computer -in 3D of course! > Definately go with a sound intensity map of the whole computer -in 3D of > course! Just in case you were tempted to take that seriously, that is a waste of time. Cheers Greg Locock >>Definately go with a sound intensity map of the whole computer -in 3D of >>course! > Just in case you were tempted to take that seriously, that is a waste of > time. i.e. bigger is not always better. Sound Intensity lines pockets of instrument makers and professors. Ang. C. > I'm looking for a good way to measure the noise levels of quiet > computers in dB(A). Since it is very expensive to get a decibel meter > to measure levels that low (and even with no budget limitation, I can't > even find one!), I'm looking into alternatives. The suggestion by DaveL is practical. First of all, you need to be aware of the normal threshold of hearing, which, in octave bands, is a few dB positive at 1000-4000Hz and rises above and below that frequency range (takes more level to be heard). Figuring that you want to have your measurement apparatus to have a noise floor below that by as much as 10 dB, then you need to have the sound level meter system to have a threshold around 0 dB in that 1000-0000 Hz range. The only microphones I have ever had tat comes close to that is a 1" condenser microphone. The 1/2" mics have a noise floor about 10 dB and more above that. But getting closer to the computer equipment is the most reliable way to get good data. Since you are willing to settle for "Approximate" values, extrapolation by inverse square law will give you he answers you want. For instance, If you measure various peripherals or the CPU box or a lap top, at a distance of one foot from what you believe to be the noise center (fan discharge opening, etc., then if you want the vale that should be at one meter, that's about 10 dB less. > I'm thinking about just using a parabolic amplifier for our microphone, > and then calibrating to known levels. Am I just missing something, or > is that a solid route to getting the measurements I need? The problem with a parabolic "amplifier" (I think you mean a "parabolic reflector") is that it has gain proportional with frequency; another variable you do not want to fuss with at this stage. Just find a good quiet room (I can get about 27 dBA at night in my home office), and set your existing equipment microphone closer and closer to a test unit. When the measured level rises 6 dB when you halve the distance, you know you are getting proper data (the test sound is well above the room background.) > I don't need absolute precision, but need it close enough that I can at > least have confidence about the numbers I get! You might tell us what the "noise floor" of you unit is (just cover the microphone with an inactive calibrator, and read what's indicated). Also get few close-up readings and tell us the result. You can measure right up to any surface (I have often measured about 1/2" from walls that I hear sound coming from). Just keep the microphone out of any air currents (if you can feel an air current on you cheek, it is too much). Place the microphone an inch or two aside from that blast. (I can feel such a mini blast coming out of this T43 Think Pad, side discharge. Angelo Campanella |
Reply to this Message |
 Sign In
 Join
 My Latest Posts My Monitored Threads My Blog My Photo Gallery My Profile My Homepage Start New Thread Enable EMail Alerts Rate this Thread
|
©2009 Advenet LLC Privacy Policy - Terms of Use
This website includes both content owned or controlled by Advenet as well as content owned or controlled by third parties.