Ambient levels and noise floor discussion

Discussion in 'General Audio Discussion' started by Serious, Nov 28, 2019.

  1. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,695
    Trophy Points:
    113
    Location:
    near Munich, Germany
    It's always irked me a bit that @purr1n kept referecing levels in the 30dBA range as a very quiet ambient level, about the best you can normally get to. I did some tests in the past and when he repeated the same thing again in a new thread I remembered it. I think 30dbA is louder than a bedroom should be at night.

    One quick test I did was to set white noise at 50dBA on my desktop monitor speakers for a digital full scale signal in Audacity.* Then I normalized the levels to quieter levels. At 30dBA it was still pretty loud and I wouldn't want my bedroom to be this loud at night. Subjectively the 20dBA white noise was still louder than ambient here, at least at night. 10dBA is pretty darn quiet and at 0dBA it got hard to hear, especially over my monitor's PSU, the transformer hum in the amp I have on my desk, etc. When it started drizzling it was definitely much quieter than ambient. If I heard something at -10dBA I can't say it wasn't placebo without a blind test.
    Pink noise could be a better test that's worth trying, but ambient levels aren't continuous like that. There's stuff that makes a noise periodically that I can hear here, as mentioned in the other thread. The fridge downstairs keeps cycling on and off, as does the oil heating. Sometimes a car, a train, a plane.

    I don't live in such a rural area, either. Maybe I'm underestimating how loud large urban areas in the US or Asia are. Maybe someone else could try what I did and report back.

    *This will only be as good as your volume calibration - I used my UMIK-1 calibrated with @sorrodje's mic calibrator gadget a while ago. Back then I measured my headphones with a known output level, so I still have a reference - this way I can be reasonably sure it hasn't drifted much. Measuring speaker drivers the 2.83Vrms sensitivities I get are very close to what they should be, so I'll just have to assume it's accurate.
     
  2. m17xr2b

    m17xr2b Friend

    Pyrate Banned
    Joined:
    Mar 26, 2016
    Likes Received:
    3,988
    Trophy Points:
    93
    Location:
    United kingdomland of fish and chips
    0dBA is a mosquito flying 3 meters away, I'm really impressed you can hear anything.
    30dbA is a super quiet bedroom, your description of pretty loud doesn't make sense.
    [​IMG]
     
  3. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,695
    Trophy Points:
    113
    Location:
    near Munich, Germany
    That's why I said you should try it yourself, since I don't really agree with those descriptions. Could be my volume calibration is off, but I doubt it's that off.
    Btw, I've measured my hearing threshold in a blind test at home before. Sadly at the ENT they start their tests at 0dB, even for frequencies in the 500Hz-4kHz region where it should be much lower than that. I'll have to generate a graph for that, but 0dB should definitely still be audible.
     
  4. Lyander

    Lyander Official SBAF Equitable Empathizer

    Pyrate Contributor
    Joined:
    Feb 25, 2017
    Likes Received:
    11,627
    Trophy Points:
    113
    Location:
    Philippines, The
    This is actually something I was tentatively interested in some time ago but did not delve too deeply into, so please pardon if I'm making any silly mistakes, and feel free to call me out.

    "0dB is too loud" = I may be forced to believe that there's something lost in translation or that not everyone is on the same page here because what the ballsacks is this, haha.

    "The quietest sound a humancan hear is 0 dBSPL... Zero dBSPL corresponds to a pressure wave of 0.000002 Pascals,which causes the eardrum to vibrate by approximately 10−8mm. To put thatin context, 0.000002 Pascals is less than a billionth of the ambient pressure in the air around us, and 10−8mm is smaller than the diameter of a hydrogen atom"
    https://www.ucl.ac.uk/~smgxprj/public/askscience_v1_8.pdf

    Yes ;)

    I actually have some stuff that I wanna say when I'm more able to put together a comprehensive response, but it may be worth your time to look into sensory and neural adaptation and how multiple tones tend to "blend" after a while, as per here: https://psycnet.apa.org/record/1986-10938-001, meaning that it's likely even random noise as present in most settings may over time devolve into constant, indistinguishable noise. I still easily hear birds screaming at one another or vehicles speeding by when it's relatively quiet outside, but for the most part when I'm just home I do note that my capacity to discern individual noise sources amidst a plethora of others is diminished.

    Infrasonic and low frequency noise (ILFN) in urban settings and even in relatively rural ones is more or less pervasive as multiple studies have shown (http://www.engineersjournal.ie/2018/01/23/ilfn-infrasound-low-frequency-noise-turbine-health/; http://hyperphysics.phy-astr.gsu.edu/hbase/Sound/infrasound.html; this one's really interesting: https://physicstoday.scitation.org/doi/pdf/10.1063/1.883019), and these sound waves can travel ridiculously large distances.

    This next bit is something I have no literature to cite for because it either doesn't exist or I might've missed it, but is something that seems to make logical sense to me. If mistaken, anyone please do feel free to correct me, happy to learn: I wonder whether said ILFN might in turn be triggering sympathetic resonances in the environment, and the interactions therein causing a cascading effect of sorts contributing to that general "background noise" present in many settings. Pure substances are actually rare in nature, many structures here are concrete with pebbles filling in and affixed with heavily oxidised metal poles wrapped in similarly rusty mesh fences— multiple substances that are structurally bound to one another with any multiple low-frequency noise sources of sufficient intensity triggering an absolute chaos of cascading noise.


    TL;DR
    What I'm trying to get at is while I have no idea what it's like in your place, exactly, there are many, many places the world over that are subject to both anthropogenic and naturogenic sources of noise that, in conjunction with the likely effects those have on home structures and surrounds, I find it hard to believe that, unless you've taken active measures to soundproof the fajitas out of your listening space, 30dB A-weighted is so easily attainable and actually constitutes "loud noise" in that setting.

    On a related note, neighbours are firing up the karaoke machine again, there's a dog that isn't ours losing their shit, and there's a fourth plane flying overhead this past hour. Yaaaay.
     
  5. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,695
    Trophy Points:
    113
    Location:
    near Munich, Germany
    Well, I referenced my threshold tests a few times now and just put them in a nice image. I didn't want to share them since I got way too good a result for me to believe it at first. Maybe my methodology was that flawed. Note that I should have very little hearing loss, I'm 21 years old and have only been to a concert once without hearing protection (and ended up plugging my ears more than half of the time). I do like to listen at higher volume levels at home though (say 80-90dBA) and will most likely suffer hearing loss from this. I periodically test my hearing thresholds and the upper limit of my hearing, but nothing will prevent it from getting worse over time.

    I did these as ascending tests, so sine waves ramping up in level until I heard them. Used the foobar ABX comparator plugin for the tests. I measured the FR of both speakers at a known level before, so I could set 0dbfs at a known level and then generated quieter tones. I made them to 1db accuracy. I played the tone through both of my speakers and sat at the listening position, so overall listening conditions are closer to DF than FF. I only tested a few sine waves, my file for generating the plot in REW is attached. According to Etymotic there's a microstructure to the threshold that can "show threshold changes of 5-10dB when the test frequency has changed by just a few hertz.", so this is probably not the best way to test it anyway.
    Overall this was sort of a pain and I don't recommend it. Probably took me an hour at night and I'm sure this depends on your daily form aswell. I tend to hear a very quiet noise (similar to pink white noise) that seems to vary in level depending on how much sleep I get, etc. When I tested it, it was most likely a very good day.

    Anyway, with that huge disclaimer out of the way, here's what I got:
    Hearing threshold with overlaid speaker FR vs Zwicker age 20 test group.jpg
    Note that I overlaid the FR of the left and right speakers, the LR sum will have a bit of a different FR. I don't have a LR sum graph atm, though.

    EDIT: I should mention that when I did threshold tests with my UERMs a few years back they weren't as good in the midrange, but I also think that occluding the ear raises the noise floor and I don't think the UERM will have as much energy in the midrange at the eardrum.
     

    Attached Files:

    Last edited: Nov 28, 2019
  6. Priidik

    Priidik MOT: Estelon

    Pyrate BWC
    Joined:
    Sep 27, 2015
    Likes Received:
    2,339
    Trophy Points:
    113
    Location:
    Estonia
    My house is then super quiet, my hearing godly...
    Or I have freakish loud mosquitoes here.

    I can hear the bastards to the other rooms when trying to fall asleep.
     
  7. RobS

    RobS RobS? More like RobDiarrhea.

    Joined:
    Feb 21, 2019
    Likes Received:
    1,419
    Trophy Points:
    93
    Location:
    .
    LOL how big are those mosquitoes?
     
  8. Priidik

    Priidik MOT: Estelon

    Pyrate BWC
    Joined:
    Sep 27, 2015
    Likes Received:
    2,339
    Trophy Points:
    113
    Location:
    Estonia
    Thing to note here wrt to mosquitoes or high-pitched farts of other origin:
    Brain is much more receptive to stuff over the 'ambient'. Just like vision is more receptive to movement than stationary images.
    It should be significantly more difficult to detect white noise of 0dBA than some annoying high pitched, LFO modulated sound events of same loudness.

    edit:
    Now that weather gets cold here, it's beard growing season. After trimming with the beard trimmer, I have 'spectral burn' in my hearing, and everything sounds as through some vibrato effect. Perhaps the same mechanism why @MisterRogers is less sensitive to his wife.
     
  9. m17xr2b

    m17xr2b Friend

    Pyrate Banned
    Joined:
    Mar 26, 2016
    Likes Received:
    3,988
    Trophy Points:
    93
    Location:
    United kingdomland of fish and chips
    [​IMG]
     
  10. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    91,561
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    You are right. Your room ambient is 10db SPL. I don't care.

    Still doesn't change the fact that we don't need more than 80db SINAD. What difference does it make since my argument was always "60db SINAD is good and 80db SINAD if you want to be totally sure." Keeping in mind the context, original thread where you brought this up.

    Refer to my blog post. I never changed the goal posts.
     
  11. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,695
    Trophy Points:
    113
    Location:
    near Munich, Germany
    I don't care about your SINAD discussion or your "blog post", lol. That was never my point. I was just wondering if so many people really have a constant 30dB ambient level and would like to know more about other people's rooms and how quiet they get at night. Secretly I'm scared of moving to the city because I know I will never experience silence again.

    The problem is it seems to be expensive to get equipment that can measure such low ambient levels. Maybe if/when I ever build something we can make a loaner because I'm genuinely curious. Seriously, I want people to try my test. All you need is an SPL meter. Heck, a phone app and the built in speakers in a phone or laptop should be good enough.

    I feel that 0dBA white noise should be audible over most ambient noise levels. I should've just made a new thread instead of polluting the other one, I admit.

    EDIT: I also think our definitions are different. I strictly meant 30dBA is bad if it's the best the room ever gets to for a short period of time (like 30 seconds to 3 minutes). If it's 30dBA on average that's not too bad.
     
    Last edited: Nov 28, 2019
  12. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    91,561
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    You did your experiment wrong.

    56db SPL. It's just drizzling outside.. and actually not that loud. This is 4ft from the door.
    IMG_20191128_094651.jpg

    I then close the door to the outside and set the mike here around where my head usually is.
    This is white noise at near 70db SPL. The ARTA Gen is at 0dbFS.
    IMG_20191128_094920.jpg
    upload_2019-11-28_9-53-24.png

    Then I did this: reduced the level down -40dbFS in ARTA to get approximately 30db SPL (Rat Shack meter won't go below 50db SPL)
    upload_2019-11-28_9-53-44.png

    I can't hear the white noise over ambient (the door is still closed), unless I put my ear up against the speaker 12". I'd say it's just a little louder than the inherent hiss from the Kali LP6 monitors (switching the white noise on and off).

    "Faint" is a good word.

    There's probably too much ambient noise today with the drizzle outside, although with the door closed, I'm not sure how much this can be perceived. But as I've already said, the quietest room in the house (my garage) is worse than the loudest sound edit bay at any of my work locations.

    EDIT: My ears seem to have adjusted to the quietness of the environment after 10 minutes. I was originally in the house: breakfast and all with kids attacking each other and cats begging for food. Interesting how our ears adjust and go into "protect" mode and then loosen up for extra sensitivity in "safer" environments. The drizzle just died down too, so I can now hear the 30db SPL white noise. It's still faint.

    BTW, Rat Shack meters are fricking accurate, or accurate enough for professional work. Engineers at the sound houses still use them. The old ones with meters are highly sought after. The new ones have inferior digital meters.
     
    Last edited: Nov 28, 2019
  13. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,695
    Trophy Points:
    113
    Location:
    near Munich, Germany
    Well, then I must've really done something wrong. I generated noise in audacity (white at first, but I also just tried with pink again) and normalized it to 0dB. I set the 0dBfs noise to 50dBA according to REW with my UMIK-1.
    At -20dBfs it's impossible to not hear. At -50dBfs I could see it being masked by ambient, but not even at -30dBfs or -40dBfs. What's going on now?

    Going to check my volume calibration. Seriously, my 30dB is at the point where I might be able to pick it up in the next room over with the door closed, so maybe something is wrong. It's similar to the level of a modern "quiet" laptop fan.
     
  14. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    91,561
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Check calibration, gain, volume pot, etc. Get a Rat Shack meter to get quick reference readings. Such "appliances" come in handy and don't require computers, USB, software sliders, OS sliders, etc.
     
  15. Senorx12562

    Senorx12562 Case of the mondays

    Pyrate
    Joined:
    Oct 26, 2015
    Likes Received:
    3,336
    Trophy Points:
    113
    Location:
    Bird-watcher's paradise
    I have never lived in a place where, with all windows, doors, etc. closed and air movement devices like fans, furnace, etc. off, I could get below 30 dB ambient spl, no matter the device used to measure it, from an actual meter like purr1n's to an app. And right now I live in the woods, with no traffic or businesses, and nothing but trees within at least 100 meters.
     
  16. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    91,561
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Note edit:

    My ears seem to have adjusted to the quietness of the environment after 10 minutes. I was originally in the house: breakfast and all with kids attacking each other and cats begging for food. Interesting how our ears adjust and go into "protect" mode and then loosen up for extra sensitivity in "safer" environments. The drizzle just died down too, so I can now hear the 30db SPL white noise. It's still faint.

    Now I do wonder how the dynamic range is for hearing. I bet after exposure to 100db peaks at a concert, I probably wouldn't be able to detect anything less than 50db for at least 20 minutes to an hour.
     
  17. Luckbad

    Luckbad Traded in a unicorn for a Corolla

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    3,412
    Trophy Points:
    113
    Location:
    Holly Springs, NC
    If you're talking about recording software, 0dB is 100% volume. Anything over that is clipping. It's not 0 actual dB.

    I just did a quick test. My room is extremely quiet and acoustically treated. When there are no drips of rain outside and everything is turned off, I'm at 29dB. That's basically "wtf the power went out at 2AM" levels.

    My ultra quiet computer brings that up to 33dB from 1m away. A car passing a few streets away bumps it up several more dB.

    Either you literally live in a multi-million dollar anechoic chamber, you don't understand SPL, or your measurement setup is jacked.

    Out of curiosity, what volume does your computer measure at and what's in it?
     
  18. Priidik

    Priidik MOT: Estelon

    Pyrate BWC
    Joined:
    Sep 27, 2015
    Likes Received:
    2,339
    Trophy Points:
    113
    Location:
    Estonia
    I trust Marvey's RatZac more than Serious' assumed calibration.
    You could be still both right.
    Genelec 8050A's self noise is supposed to be less than 10dBA/1m on axis.
    In a really quiet night I can hear it very slightly when I switch off everything else.
    Tiny transformers in my dac make more noise. Here, another anecdotal datapoint.
     
  19. Hands

    Hands Overzealous Auto Flusher - Measurbator

    Staff Member Pyrate MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    12,300
    Trophy Points:
    113
    Location:
    Colorado
    Home Page:
    Me: My house seems pretty quiet right now

    Electricity: *shuts off due to power outage*

    Me: *sound void causes overwhelming existential dread*
     
  20. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,695
    Trophy Points:
    113
    Location:
    near Munich, Germany
    @Luckbad no normal meter will detect much below 30dB. Also why can you hear your computer? I can't hear my computer at all. Noctua fans at 200RPM are the only moving parts. My UMIK-1 has a noise floor at 37dBA. Plugging in my HDD is unacceptably loud and it measures in at 38dBA now, so I assume even the HDD is quieter than 37dBA.
    My PC is quieter than my Surface Book 1 when the SB's fans aren't blowing.
    My CRT monitor and plasma TV do make a noise, though. Less than 37dBA since my UMIK-1 can't detect it, but definitely audible. Even my LCD monitor makes a noise.

    Okay, I don't have a normal SPL meter, so math is the only way:

    First test I did: Amp set to 4Vrms for a 0dBfs signal. Speakers are 103db/4Vrms/1m, but two speakers so 106db/4Vrms (or 109 depending on the frequency). My listening position is at a 2m distance. In-room there won't be much of a level difference between 1m and 2m, but let's assume free field conditions, so it's back to 103db for a 0dBfs signal. 800Hz tone set to -46dBfs, so it should read 57dB. I got around 58dB average when moving the mic around. Spatially averaged (to get rid of room-modes) it seems to check out.

    Second test: I have a file from 2017 where I saved an HD800 measurement plugged into the GO450 at 21dBfs. At 500Hz I got 86.4dB back then, now I get 86.6dB. I guess this means my mic at least hasn't drifted since I did my calibration with @sorrodje's gadget. I'll see if I can find a cheap enough SPL meter that I can buy, since those things are handy anyway, but so far I doubt my volume calibration is that far off.
     
    Last edited: Nov 28, 2019

Share This Page