View Full Version : Measure your projector fan's noise.


gsmollin
06-09-07, 08:52 PM
This problem surfaces repeatedly in forum posts. The noise from projector fans are a constant issue, and just how noisy are they really? The Radio Shack Sound Level Meter is an inexpensive tool for the hobbyist, but it has a minimum sensitivity of 50 dB, too high to measure a modern projector fan at the standard 1 meter distance, or at your viewing seat. But, the meter has an analog output, and if you connect an inexpensive true RMS DMM to that, you can extend the sensitivity down to 30 dB, or lower.

Calibrate the DMM against the meter readout, using a source of sound. I used an air conditioner fan. Set the RS meter to A weighting, 60 dB range, and slow response. Set the DMM to AC volts. My meter read 0.230 VRMS at 60 dBA. Call this reading "V60". Now set up at the projector. I put the RS meter 1 meter from the edge of the projector, in the direction of my seating area. The meter must be secured by a stand to get consistent readings. Do not stand near it while taking measurements, or reflections from your body can affect the reading. Take a background reading. You may have to go around and shut off noise sources, such as air conditioning, fish tank pumps, computer fans, and fluorescent lamp ballasts. You may also have to wait until 2 AM for traffic to die down. Record your lowest room-measurement voltage, and call it "Vr". My reading was 0.012 VRMS. Convert this to dB by using the formula:

dB = 20 * log (Vr / V60) + 60. (1)

My room was 34 dBA, not all that quiet, but it wasn't 2 AM either.

Now turn on the projector and take a second reading. Call this reading "Vp". You can convert that to dB using (1) above. My reading was 0.015 VRMS, which converted to 36 dBA.

Now we must subtract the room background from the projector noise to get the sound level of the projector by itself, as if it were in an anechoic chamber. We do this by subtracting the equivalent power levels of the two readings, using a root-sum-square method. This result is ratioed to V60, and converted to dB as before.

dB = 20 * log [ sqrt ( Vp * Vp - Vr * Vr ) / V60 ] + 60. (2)

My result was 32 dBA. If you need more fan noise to get a measurement, you can move the meter closer to the projector. At 1/2 meter, I got a result of 37 dBA. Inverse-square law theory tells us the measurement should increase 6 dB for halving the distance to the source. The 5 dB change indicates this is true, but the measurement becomes sensitive to exact placement and near-field effects that may interfere with the inverse-square-law relationship. Also, halving the distance to the edge of the projector does not halve the distance to the noise source, which is recessed inside the projector case.

I find this projector audible when I sit through quiet passages. My seating area is on the exhaust side of the projector, which makes it worse.

I hope this helps somebody measure and compare their projector fan noises. Remember that the specified levels are taken in an anechoic chamber. A normal listening room has many reflective surfaces, which raise the noise level. It is unlikely you will equal the manufacturer's specified sound level.

Kevin McCarthy
06-10-07, 05:08 AM
gsmollin: Very helpful, thanks! I had no idea the RS meter had the ability to go way below 50 dB via its analog out connector. It's odd that they set up their A to D input resistive divider scaling so as to not be able to use anything below 50 dB.

Kevin

gsmollin
06-10-07, 08:20 AM
You can also feed the analog output into the sound card of your computer. There are a number of analyzer programs that can be used with the RS sound level meter. Some of them are freeware. For instance, this website has a very nice octave band real time analyzer that is freeware: http://www.trueaudio.com/ It would be very interesting to compare the frequency spectra of different projectors. Some fans have higher frequency whines that are very noisy to the ear, even though they not any louder when measured with a sound level meter.

Just remember a couple of things: If you don't have >10 dB of signal-noise ratio, you must use equation (2) above to subtract the room background from the fan noise. Computers have fans of their own, so long cables may be necessary.