2
\$\begingroup\$

I am working in a large office building with a lot of people. Pretty frequently there is a very weak and low frequency noise that is very irritating. Unfortunately, I have found just one other person that can perceive it, so it must be at the utmost borders of what a human can hear. I have started searching for equipment that can detect the origin of this sound, but after some hours worth of effort I have not come up with much except some postings on this site about array microphones.

My questions are:

  • is an array microphone the correct way to solve the problem?
  • which types of microphone / sensors can detect weak, low-frequency sounds? I figure that a high amplification will also amplify noise.
  • are there other ways to locate the source of the sound? Maybe I'm overlooking ways to solve the problem.

EDIT: A sensor or other electronic part would help me as well.

\$\endgroup\$
  • \$\begingroup\$ Product recommendations are off-topic here but if you don't mind dropping $8 for on-line access to an issue the following related article in a local electronics magazine might be worth a look. I didn't have much personal interest in the subject so can't remember the solution but most of their designs are at a hobbyist / low-cost level: siliconchip.com.au/Issue/2013/March/… \$\endgroup\$ – PeterJ Dec 4 '13 at 11:59
  • \$\begingroup\$ Use your own ears: at a quiet time, take a floor plan of the building and walk through the corridors. Using coloured markers or numbers, grade the loudness of the noise at each location. That should give you at least a vague idea of where to look. I'm placing my bets on an HVAC system being the origin. \$\endgroup\$ – Mels Dec 4 '13 at 12:30
  • \$\begingroup\$ That was the first thing I tried. The vibration is there only occasionally ... \$\endgroup\$ – hochl Dec 4 '13 at 12:33
  • \$\begingroup\$ As @Mels said, my first suspect would be a large fan somewhere that is part of the air handling system of the building. Something may have happened to the fan so that it is slightly imballanced, which causes it to vibrate at the rotation speed. For a large fan, this could easily be just below the hearing range but still high enough for it to be annoying or irritating, even if not perceived directly. Ufortunately, such vibrations are hard to localize and can use the hard structure of the building as sound conduits. \$\endgroup\$ – Olin Lathrop Dec 4 '13 at 14:56
  • \$\begingroup\$ From my observation I think it is a machine being switched on or some door. The sound is not there continuously; but then, it is very faint -- quite possibly you have a point there... if only I could locate it, at low frequencies it seems to be near impossible. I'm fairly certain it is on a different floor of the building because I've walked over the whole place several times in the last months. If the building structure is transporting it (for example some pipe) then I'm not sure how to find it since they most probably are inside the walls (we have double floors and ceilings). \$\endgroup\$ – hochl Dec 4 '13 at 14:59
2
\$\begingroup\$

A good quick and dirty sensor for low frequency sound is a large speaker used in reverse. Basically, you use it as a microphone with low impedance output. You can amplify such low frequencies with just about any opamp. Low pass filter the signal with a rolloff in the 100-200 Hz range at each step, with the high pass roloff at 1 Hz or a little below. After a gain of probably around 1000 or so, you should be able to push on the speaker cone with a finger and watch a scope trace go up and down.

Be sure to shield everything carefully. You won't be able to easily reject the power line hum frequency, so the better strategy is to keep it from getting into your circuit in the first place. Power at least the first stage of the amplifier from a battery, all mounted right at the back of the speaker. After 100x voltage gain or so, you can send the amplified signal elsewhere, like to a scope you carry around or even one that is plugged into the wall.

\$\endgroup\$
  • \$\begingroup\$ Using the speakers as microphones sounds promising, esp. since I have a good selection at home. At those low frequencies, would the signal not include a whole lot of vibrations from the streets around the building and people walking (leaving aside the electrical aspects)? \$\endgroup\$ – hochl Dec 4 '13 at 15:06
  • \$\begingroup\$ @hochl: Yes, sound in the 1-20 Hz range can come from a lot of sources we don't normally consider sounds sources. \$\endgroup\$ – Olin Lathrop Dec 4 '13 at 15:32
4
\$\begingroup\$

Another quick and dirty one is your smartphone with accelerometer in it. Put your phone on the floor and start recording accelerometer recordings. Run a FFT to the recordings and you will be able to start locating the source.

\$\endgroup\$
3
\$\begingroup\$

I had a similar problem in a university I worked at. We used a seismic accelerator, a preamplifier, and a laptop. We tracked the noise down to the chillers on the roof, when both compressors were running they sent a 100hz signal with low amplitude straight through the building like a huge tuning fork. Changing the dampening springs and adding other dampening media removed the symptoms in the end

\$\endgroup\$
3
\$\begingroup\$

A similar infrasound problem happened in my appartment building. I used a stethoscope on the walls to track the source. It was a centifugal fan in a neighbor's ventilation system. It had not been cleaned for years, and dust had accumulated inside in an asymetric pattern, causing an unbalance in the center of gravity of the rotating part. The infrasound corresponded to its rotational speed, and cleaning it solved the problem.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.