1. Register NOW and become part of this fantastic knowledge base forum! This message will go away once you have registered.

low frequency recording

Discussion in 'Pro Audio Equipment' started by stjohnson, Nov 26, 2007.

  1. stjohnson

    stjohnson Guest

    Hello all,

    I have an issue I'm not sure how to tackle, so I'm searching for some advice. I need to record for legal documentation a very low-frequency low-amplitude drone that is present in some households. Apparently it is caused by some fans in a warehouse that is located about .25miles away from a neighborhood. But, I'm not sure of the best mic (omni, amorphic, cardio) to buy to record this. Is there a good mic someone knows that is particularly sensitive in very low frequencies?

    My task will be to go into people's homes into different rooms, make the house as quiet as possible (turn off all appliances and A/Cs) and start recording. Then I will need to pull up the drone sound and record for a few minutes.

    The audio is so low that many people don't hear it. I think some people are just less sensitive to it than others, but that's the type of noise I'm trying to pick up. So I need to be able to document the noise-floor of a mic to prove that the noise recorded is not actually mic-hum, but rather the low-freq from the warehouse.

    Also, it would be absolutely wonderful if I had a mic (shotgun type) that I could bring to the warehouse, target one of these external fans, and pull a sample that I could filter out all other frequencies and prove the sound print was identical to what was sampled in people's homes.

    The high frequency sounds are lost in space from the warehouse to the homes, but the bass travels through the walls and insulation into homes. So that low sound I have to record.


    I would like to go through my mixer and record to disk on my macbook's analog input. But if recording to HDV or DVCAM tape is better, I could bring a camera along soley for it's audio inputs. Not really sure what provides the best signal to noise ratio.

    I would like to do this w/o breaking the bank, and have a mic that would be good for other purposes as well, but I'll take what I can get.

    Any advice would be greatly appreciated!
     
  2. rockstardave

    rockstardave Active Member

    wow great question. i'm anxious to see what everyone thinks.

    i'd assume some sort of large diaphram condensor microphone, but i wouldnt know which ones to go to...
     
  3. i would think if you recorded something you cant hear....your not going to be able to hear it. there are tools you can make to show the existance of the frequency being there but what you hear is what you get. if its possible id say an ultra sensitive large diaphragm mic with basically no self noise
     
  4. stjohnson

    stjohnson Guest

    I can actually hear the drone. My wife acts like she's straining to hear it, and then says it wouldn't bother her. I guess we have differing levels of tolerance, but it is definitely a noticeable frequency present. It's just like someone has a high voltage 50-60hz transformer attached to the frame of your home in the basement, and the hum from the transformer is being transmitted into the walls which you can pick up on from other rooms. That's the best I can describe it. I guess I'm more sensitive to it being more of an audiophile than my wife.

    I've been pointed to a Neumann TLM 103 as a good mic to pickup low freq with a very low noise floor. I might give that a shot, but the problem is that it won't be representative of the actual db or sound level of the noise. I need to have in addition to a recording sweetened to highlight the hum, some sort of measurement of the sound. So I'm researching SPL meters now. The one I have from Radio Shack is not going to cut it. But perhaps they make one that I can tune to a specific freq (or something, maybe octive) and get a measure of the actual sound and frequency. I think this will hold up more in court for my client than a general recording would. Or at least it would add some merit to the recorded track.
     
  5. Boswell

    Boswell Moderator Distinguished Member

    You need a measurement or calibration microphone coupled to a recorder with an adequate low frequency response.

    The Behringer ECM8000 is a perfectly adequate measurement microphone, but needs phantom power, which limits your portable recorder options. Don't get the Zoom H4 recorder even though it can supply phantom power, as its low frequency response is not up to the task in hand.

    Conventional microphones such as the the Rode NT4 have a low 3dB point of 20Hz and can be battery powered, so one of those running into a Sony MZ-RH1 minidisc recorder would do your job.
     
  6. JoeH

    JoeH Well-Known Member

    Actually, you'll need a SMALL Diaphragm microphone, something like the earthworks or the DPA 4006's. THese microphones were originally calibration mics, and go all the way down to 20 HZ and a bit lower, and you'll need tha accuracy when doing something like this. (DPA used to be B&K - which is now exclusively a calibration & testing microphone company.)

    Believe me, I understand EXACTLY what you're trying to capture here. I'm currently sufferening from something very similar to this. About .2 miles from here is a new construction site where they are building a 2 or 3 story office campus. It will be fairly eco-friendly and will be tucked behind a huge berm and lots of trees, but in the meantime, the construction process is intense/insane.

    Any given day, there are at least two to four large earth movers, backhoe's and levellers pushing dirt around and digging trenches. The beeping is one thing, but the combined low end is absolutely astounding. The morning hours are the worst; I can barely sit in my studio (which at the moment is a semi-finished garage, while I'm buidling my new space). Delivery people have stood at my front door and remarked how bad the bass is at my house (which seems to work as a resonator) and how they can't hear it on the street - even CLOSER to the construction site! If you stand outside, in front of my house, it's not bad, but when you come INSIDE and feel the resonance, it's ungodly.

    Fortunately, it won't last forever, and they're done working most days by 4, so I'm trying to take it all in stride.

    But in your case, you've got to document this acoustically AND visually. (With charts and graphs, etc.)

    If there's an attorney or class action suit involved, you should get some "Up front" money to cover your expenses, perhaps from the law firm handling this.

    You're going to need the following to do this properly:

    Professional Mic's and pre's with a Firewire or USB (Digital!) interface to your computer's recording software. (Do NOT use the analog inputs, no matter how tempting) to capture this low end properly. I'd suggest renting it all for a couple of days, esp if you're not going to need this beyond the capture/documentation stage.

    Software to properly show the frequency response in a vivid, visible way. Any good spectral editor will really wow them. I have Algorithmix (which is about $3k stand alone, you can probably get the same thing from Adobe's Audition - a spectral editor/viewer that does similar stuff for a lot less). You will want to do a screen capture and print it out too.

    The low end stuff will be immediately visible to anyone, including audio amatuers, with just a little bit of help: The spectral analyzer shows them frequency, time and amplitude in a very easy to understand way. (Left to right is time, Top to bottom is freqs., and COLOR is the amplitude/intensity.)

    And last but most importantly, if you're going to show anyone (a judge, an attorney, or the company owner) what it all FEELS like, you're going to have to get a subwoofer (with calibration documentation to prove you're not hyping it!) to reproduce this sound in a neutral environment, away from the er, crime-scene. Perhaps a local studio or someplace that's very quiet and still. - when you switch on the Low end noise, it should be dramatic and very impressive.

    Then while the noise is on, perhaps you can attempt to show them what it's like to live with this sound day in and day out. (Maybe a tv show, phone call, or even a simple meal with this going on.) If you're creative, you can possibly show something even VIBRATING to the low end to give it visual impact.

    I would think the first person you should talk to is the attorney for the person with the machinery. A couple thousand in rubber motor mounts or factory retro-fits would be alot cheaper than a lawsuit, then LOSING the lawsuit, and then paying for the repairs anyway.

    Good luck, and drop me a note if you want any more ideas on what it's like to live with constant low end noise. :roll:
     
    audiokid likes this.
  7. MadMax

    MadMax Well-Known Member

    I'm not an expert in forensic audio, but these are the issues as I see em';

    Your recordings will have to be made for at least for 2-24 hour periods...

    You will have to record the rooms/structures in their "natural state"... e.g. all standard appliances, etc running. Then you will have to record with all of said appliances off.

    Then you will have to produce waveforms that show the difference.

    As you pointed out, you are looking for a microphone that has a much lower frequency cut-off than your typical mic. This isn't going to be available in your typical banjo center. If the frequency is actually as low as I think it is, you are likely looking at a fairly sophisticated transducer that would be better attached to the structure than a microphone in free air.

    The first thing to do is forget doing this yourself. Hire a licensed acoustical firm to come do the testing. Since they do not have a vested interest, they will give you both the source of the noise and a plausable solution. Besides, they already have all the gear to monitor and test.

    Split the cost of their services among the neighbors who have a complaint. You should get reimbursed by the court when you take a class action suit against the warehouse.

    Even if you are able to record this accurately, but then enhance the problem frequency in the wrong way, or present it in the wrong way, you will lose your case as soon as it's presented.

    The frequency of the noise is the issue. Think about how low this frequency is. It's traveling over a quarter of a mile! This falls in the range and classification of industrial noise. Realistically, you don't stand a chance of fighting this without proper professional assistance.
     
  8. stjohnson

    stjohnson Guest

    Boswell, this is a good point. I am looking at studio mic's. I should tune my search a bit, so to speak. I'm ok with needing phantom power, as I have field gear and equip that should be fine if I need to go away from an outlet.

    Joe, that is insane info. Incredibly useful. You do understand the noise! It's exactly like that for about 4 households that have stepped up. I'm not sure if there are more, but for now, it's not a huge legal issue since it's such a quiet drone. People are more passive about it than they should be in my opinion. I have the simple Adobe CS3 sound booth software and it does have the sonar feature you mention (which I couldn't live without). I was definitely planning on using that to filter the high frequency sounds. GREAT freken idea with the showing people what it's like. I definitely don't have a good enough sub, but I could stop by the local guitar shop and rent one of those for a demo if needed. It would be clunky and difficult, but man that would be awesome!

    Max, thanks for the insight. I don't think I'll be able to sample people's houses for 24 hours! But from what I'm told this shouldn't be necessary. Just proof of the sound and level is what is needed. I'm doing this myself because I have technical knowledge of audio engineering, but just not the gear to accomplish this task. Also, it hasn't escalated to a class action yet. This is the basework to establish a case, but will be used primarily at this point to solicit the company owning the warehouse to modify their ventilation systems. If no go, then this becomes real and suits will be filed.

    But ALAS I've come across a company that will set me back ~$1k, but has the exact tool I'm looking for. Just thought I'd post it here for anyone else's info.

    http://www.linearx.com/files/pdf/RTAjr_Brochure.pdf

    These guys have an ancient ISA-Bus architecture product called the RTAjr. It looks like they can not only give me SPL per frequency but also the dB per freq (dBspl). This is exactly what I'm trying to accomplish. With the right filters on this audio I can also record to an offboard recorder through their line out. It sounds too good to be true, but I'll find out more very shortly. My last concern is the mic. The transducer might in fact be a better route to go, but I'm going to go trial and error on that one and find out what gives the best overall and repeatable response.
     
  9. Boswell

    Boswell Moderator Distinguished Member

    If you want to go down to really low frequencies, then a velocity device like a Geophone is the thing to go for. It's used for such things as seismic investigations and quantifying the vibration effects of trains in tunnels on buildings above and around the tunnel. Best used with a data acquisition board (which goes down to d.c.) rather than an audio interface.

    Click for help files

    Don't bother with the RTAjr gear, unless you are tuning up for a car stereo loudness contest.
     
    audiokid likes this.
  10. stjohnson

    stjohnson Guest

    I don't really need low frequency recording. The frequency will be 50-60hz. The amplitude will be very very low. Some of the hardest signals to pick up are the low freq low volume ones. Looks like the RTA people will allow me to eval a system before buying. Funny what they say about car stereo engineering, but I still think it might be a good solution if it does the other 1/2 of what they claim.
     
  11. RemyRAD

    RemyRAD Well-Known Member

    I would have also recommended the B. & K./DPA/Earthworks microphones, since their early existence was that of calibration and noise pollution measurements.

    The problem with what everybody is recommending here is that, you really aren't going to be able to play back this annoying sound as it appears in people's homes. This is akin to trying to record the extremely low-frequency communications of whales & elephants. First, like we recommended, you need to record the source sound in the offended locations. Then, that recording needs to be speeded up along with a spectral display and appropriate scientific explanation with the double speeded audio played back so that it can be properly audibly realized.

    Trying to give people the experience of what this sounds and/or feels like is like trying to reproduce the sound of a Shuttle launch, in a courtroom. Right, impractical at best. And that's why we have all of this wonderful equipment to be able to manipulate and display sound in a scientific way. Why not just have some idiot with an overblown car stereo play some rap and/or hip-hop while driving by the courthouse outside? Maybe that's what it is?

    I believe filtering anything out would make your demonstration invalid. All of the noise can be scientifically identified in spectral displays/analysis. And so, you really need to contract a forensic audio expert from the NTSB that can validate this noise problem.

    Thinking outside of the noisy box
    Ms. Remy Ann David
     
  12. VonRocK

    VonRocK Active Member

    You are not being subjective with this statement, and in my opinion, not justly serving your clients.

    You have stated that some people can not hear this noise, as it is too low in frequency. Then you state that you really don't need low frequency recording?

    It is logical to assume that a large part of this noise is lower than most people's ability to hear, as you stated that your trained ears can hear it, yet your wife's cannot, and some people do, and some people don't.

    Either way, a fully subjective study of the issue is desired.

    Perhaps you should pass this job on to somebody else. Seeing as you "don't need low frequency recording"

    Is there any such thing as a forensic audiologist? At the very least, I'm sure these people would not be too impressed with their audio expert seeking microphone advise on a public forum.

    I know at least one lawyer that would tear you "expert opinion" apart based on this alone.
     
  13. VonRocK

    VonRocK Active Member

    Crap, Remy's was typing and posting before I saw her....

    the gist of the message is, "it is beyond your scope of expertise" and buying a microphone is not going to change that.
     
  14. RemyRAD

    RemyRAD Well-Known Member

    Yes VonRocK, is basically saying the same thing that I said. Great minds think alike especially when they're professionals like we are. You have a fool for a lawyer if you are attempting this yourself. You need it backed up by a professional from a professional.

    I understand, you don't need low-frequency recording. But you need to record the low-frequencies, so that the spectral energy, along with all of the other crap, can be properly displayed by an audio forensic professional. And yes, forensic audio professionals exist! The NTSB has enlisted forensic audio professionals to determine particular cockpit voice recorder sounds in the background prior to a catastrophic incident. It was even suggested to me to go into that field by Robert Hager, of NBC News, because I was an NBC audio engineer that was already working with computer audio back in the mid-1990s, before NBC was playing with computer audio devices. I wasn't interested much. I'd rather record rock-and-roll than listen to people losing their lives. Maybe that's why I'm not much into death metal? Been there, heard that.

    Enjoying lively rock-and-roll
    Ms. Remy Ann David
     
  15. MadTiger3000

    MadTiger3000 Active Member

    Switch out subjective for objective in VonRock's otherwise great thoughts.
     
  16. JoeH

    JoeH Well-Known Member

    We should all realize, of course, that most of this will come down to the attorney for the suffering parties sitting down with the attorney for the factory with the fans, having lunch, and figuring out the best/easiest/cheapest way for the factory people to stop the noise.

    If they have any brains, that is.

    The mark of a truly failed negotiation is litigation in a coutroom. At that point, it's just a slug-fest, and the one with the best representation wins.

    As others have pointed out already, maybe you should get an attorney invoved FIRST, before you buy or rent anything, spell it out for him/her as to the problem & solution, and see what they say. An hour of an attorney's time up front may save you and your friends a LOT of hassle in the long run.

    One of my clients does forensic audio for criminal law in NJ and NYC, and most of the projects he brings in for documentation and analysis never make it trial; once the lawyers look at the evidence (or lack thereof), many of these things are settled quietly, far away from a judge or courtroom.

    Hopefully your noisy neighbor will see the light (or hear the rumble) and take care of it before things get ugly.
     
  17. stjohnson

    stjohnson Guest


    haha.. i love the Internet and all the opinions out there. Its a beautiful place.

    I've never said that the low frequencies are invalid, but I don't need them. For all of you shadetree lawyers out there, do some research and you will find that the range of human hearing is between 20hz-20Khz (don't shoot me down if I'm not 100% accurate here, haven't looked in a while but I know I'm close). Your lawyer that would tear my "expert opinion" down because I didn't include a full spectrum would be immediately put in his place when I was crossed and revealed that any missing information in my presentation isn't even audible to us! Most of these geophones being recommended are insanely low frequency sensitive, and not valuable for this case. The lowest I need to go is 20hz. Not 4, not 7 or 11 like some of these mics and people are suggesting. I would dare anyone to bring a lawyer to challenge that. So I reiterate, I don't need low frequency recording, I need down to about 20hz. This is not such a tall task. The hard part as I've said previously is the noise floor of the mic. Just because it's sensitive to 20hz, doesn't mean it can hear 20hz @ -80dB. The spec for human hearing is a range of frequencies, not amplitudes. The same holds true for Mic's based on their noise floor.

    The idea of duplicating for play back is far fetched, like you say Remy, but it is not far from reality. The idea of getting a subwoofer and reproducing it identically is of course not going to happen. Seems like most people can't hear the tone anyway, so why reproduce something people can't hear? That would be a funny piece of evidence. In all actuality, upon displaying the soundprint spectrally, and identifying it's specific wavelength and dBspl at different sites. it would be easy to play it back at an amplitude that was easily audible. And it would not be invalid as it would simply be those frequencies that would be boosted, rather than filtering out the surrounding noise. This is perfectly valid for demonstration purposes. It doesn't filter or alter in any way the evidentiary sound, rather just emphasizes the documented sound for a judge, should it get that far.

    You won't hear me say that the NTSB or some other "forensic audio firm" is not valid, they are valid, if only for their accreditations. But for now, there's not much they can do that we can't. If it gets to the point where we need to go into a court of law, yeah, it makes sense that if you put two people on a stage, one has a degree, so to speak, and the other doesn't, the people probably will tend to believe the one with a degree. That's a flawed perception, but unfortunately true. We would get our results certified in that case.
     
  18. VonRocK

    VonRocK Active Member

    I love the internet too!

    You asked, and you received opinions from this forum. These opinions all range from complete amateurs (like me) to very knowledgeable recording professionals. All of them, a valid contribution with varied point of views.

    The subject has peaked interest here, and in so doing created individual thoughts and a discussion concerning very low frequencies. I will be doing some of my own experiments for my own curiosity later on today with very low frequencies, and exploring an avenue with my equipment that I may have never thought to go. Thanks!

    Now, back to your issue;

    I really don't think that you should be neglecting to look for offending frequencies below the spectrum of human hearing. I can't see the flu virus, it's beyond the ability of of my human vision, but it sure has had an affect on me this last week!

    None the less, I hope that you are able to find an adequate solution, and would appreciate a continued dialog and updates as this issue develops.
     
  19. Boswell

    Boswell Moderator Distinguished Member

    The focus of this thread seems to have moved from audio recording to capture of legal/forensic evidence. Even if you disagree with that, I think there are several flaws in your argument.

    1) The point of this thread is that you are being troubled by a low-frequency acoustic emission which you believe comes from a warehouse in the neighbourhood. You "need to record for legal documentation" this acoustic emission. The annoyance factor of low-frequency emissions is not just in the range normally assigned to human hearing. A large fraction of the acoustic energy is at frequencies lower than 20Hz. The annoyance is caused not just by what you can hear but what you can feel. If a diesel-engined truck pulls up outside my office and the driver leaves the engine running, I often have to get up and go elsewhere until it departs. This is because of acoustic energy (that I have measured to be in the 7-10Hz range) that becomes painful. The levels of emissions above 20Hz (the nominal lower limit of hearing that you quote) are much smaller. You need these lower frequencies to support your case.

    2) Even if you were concerned only with frequencies of 20Hz and above, you would not record and measure these convincingly with equipment that itself was struggling to reach down to 20Hz. You would need 20Hz to be in the passband of the recording equipment, which generally would imply 2 octaves above its nominal lower -3dB point (so 5Hz in this case). A competent lawyer would shred a case where there was doubt about the equipment's ability to capture the offending sounds.

    3) Notwithstanding the hearing/feeling split form (1), the Fletcher-Munson curves for human hearing show that the sensitivity of the ear falls off rapidly at low frequencies compared with what it can detect in the region around 1KHz. Recording equipment does not follow the F-M curves, and any recording gear appropriate for this task would have plenty of headroom above the noise floor at low frequencies compared with the ear.

    4) You must not filter or remove frequency ranges from the recording. Firstly, it would be regarded as tampering with evidence if you wanted to bring a legal case, and secondly, you are pre-judging what it is in the emission that is causing the annoyance. You said "...It's just like someone has a high voltage 50-60hz transformer attached to the frame of your home in the basement, and the hum from the transformer is being transmitted into the walls which you can pick up on from other rooms." It seems you have made this analogy and now you want the recordings to fit your model. Similarly, you want opinions from this forum to back up what you have already decided.

    As JoeH, VonRocK and others have said, we are here to help, and we take time and effort to give you what advice we can based on our knowledge and experience. It's your choice whether you take any of it on board, but if you are looking to get some action against your warehouse, the opinions here in this forum are friendly, but only the first test of your case. You will meet much sterner examination if you take it further down the legal route.
     
    audiokid likes this.
  20. MadMax

    MadMax Well-Known Member

    Boswell's dead on the money.

    You cannot quote me or even refer to my reply (or any of em' for that matter) to this post with any legal credibility, but one thing I think you will find is that the real frequency of your noise is going to be in the sub 20Hz range.

    You are probably looking at nodal issues where the buildings are acting as node points like the endpoints of a vibrating string. There, the energy is transfered into the structure, and causes the structure to vibrate at a harmonic of some fundamental within the individual structures.

    This is a seriously complex industrial noise issue.

    If you want to go ahead and jeopardize your case, proceed. If it were me and my money... I'd hire a professional.
     

Share This Page