The No.1 Website for Pro Audio
measurements - bad sounding space
Old 26th April 2014
  #1
measurements - bad sounding space

Hi all
I am measuring my basement as it sounds awful after lots of sound treatment, bass traps, diffusers, clouds, etc.
I use it for recording and mixing, my music.
I am posting the rew data, floorplans and photos. There are both speakers, sub and sum measurements in it. Second file has separate left and right speakers.
The place is 6 x 4.5 x 2.7 meters (LxWxH).

What's your opinion about measurements? doing something wrong or this place just sucks..?probably the first..hehe

Ideally I want to correct the room acoustics and then place my speakers where it is best. and then phase align my sub. just like everyone does tbh..asking too much?heh

any opinions would be much appreciated! thanks
Attached Thumbnails
measurements - bad sounding space-floorplan.jpg   measurements - bad sounding space-floorplan-current-setup.jpg   measurements - bad sounding space-ceiling.jpg   measurements - bad sounding space-frontside-center-view.jpg   measurements - bad sounding space-frontside-left-view.jpg  

measurements - bad sounding space-frontside-right-view.jpg   measurements - bad sounding space-frontside-ceiling-center-view.jpg   measurements - bad sounding space-frontside-ceiling-left-view.jpg   measurements - bad sounding space-frontside-ceiling-right-view.jpg   measurements - bad sounding space-backside-center-view.jpg  

measurements - bad sounding space-backside-left-view.jpg   measurements - bad sounding space-backside-right-view.jpg   measurements - bad sounding space-back-side-ceiling-view.jpg   measurements - bad sounding space-horizontal-column-ceiling-traps-front-.jpg   measurements - bad sounding space-horizontal-column-ceiling-traps-back-.jpg  

measurements - bad sounding space-360view.jpg  
Attached Files
File Type: mdat almaspkr.mdat (7.81 MB, 168 views) File Type: mdat almaspkr2.mdat (5.20 MB, 140 views)

Last edited by almaelectronix; 26th April 2014 at 08:01 PM.. Reason: media update
Old 26th April 2014
  #2
Lives for gear
 

posting a floor plan with location of current treatment and some pictures would help the GURUs ... otherwise they have to charge extra ...
Old 26th April 2014
  #3
Sure I will post some pics!
truth is I asked and acoustician to check the place and he did and I am still waiting for his offer..It's been 2-3 months since though..
So I decided to mess around myself a lil bit, but I am not sure exactly what I am reading on rew to be honest, the only thins that i can understand is the spl and I believe if the lines of phases between sub and mains go together then they are in phase..?not sure
Old 26th April 2014
  #4
Lives for gear
 

I am not sure I understand your dimensions above, is it more than one room?

I suspect as a "basement" you have very solid walls? This is going to make the bass end more difficult to control.

It would be useful also if you provide measurements for "just" left and right individually?

That being said, as a quick look/see your "bass" trapping is not doing much trapping, the delay below 100Hz is too long for my liking! I would imagine it sounds boomy at the low end? Additionally you would get benefit from moving your relative loudspeaker/listening position set-up. There are many threads on GS that give advice on setting up and positioning to reduce peaks and nulls. Small changes in set-up can make large differences in perceived and measured response.
Old 26th April 2014
  #5
Lives for gear
 
sheggs's Avatar
I'm with ice cube here, floor plans and pics would be very useful and it is always advisable to test the left and right monitors desperately.
The more info you can post the more everyone can help. What type of bass trapping do you currently have?
Old 26th April 2014
  #6
Icecube it is just one room, sorry for the confusing numbers, I corrected that.
Yes all the walls are solid and all around are tons of pure dirt. The only wall that's "light" is the ceiling and a part about 2m of the front right wall.
It is boomy, was worst with the heavy rag on the floor I had before. Lots of standing waves in the room, much of energy way down to 20-30hz which is audible and disturbing. Listening music in there is confusing..

So, I updated the original post with one more rew file which contains separate left and right speaker measurements.
Also attached a floorplan, a plan of my current setup for sidewalls and ceiling, and lots of pictures.

sheggs my traps are mostly 4inch rigid fiberglass, but also 2 and 8 inch. corner, 1st reflections and clouds. All spaced except from the 2'' and 8'' traps. Check the images with my setup I updated on the first post
Old 26th April 2014
  #7
Gear Guru
Bad?

Your 1/3 octave Topt is close to 200mS throughout most of the audio band.
It rises to 400/500 mS at LF. These numbers are close to those recommended by International Broadcast bodies. There does appear to be substantial subsonic information below 30Hz. Not sure that a sub is a great idea in this room, but do try it in the null of that longest lowest mode. Fire up the Signal Generator in REW, Frequency follows cursor. Play the sine and find the null. Put the sub there.
You could also use Eq to minimise this rumble.
The frequency response of the mic in a Sound Level meter is generally extremely poor.

DD
Old 26th April 2014
  #8
hi, thanks for the reply.
Could you please explain how to do this? sorry i am new on this..
I understand i am playing the signal generator on 21hz which is the lowest null.
but what do you mean "find where the null is"? actually moving inside the room and find the exact spot where I cant hear the 21hz?

And what about the phase of the sub to the mains, I thought that is essential on sub placement

yes I know about the sound level meter, I am about to buy a proper one. Also used the calibration file for the meter, not sure if that helps.
Old 26th April 2014
  #9
Gear Guru
Where

measurements - bad sounding space-screen-shot-2014-04-26-22.49.44.jpg

Quote:
actually moving inside the room and find the exact spot
Yes. It is likely to be half way along the length of the room. If so, room centre would be inconvenient, so try the sub at the centre of the side wall.


DD
Old 26th April 2014
  #10
Quote:
Originally Posted by DanDan View Post
try it in the null of that longest lowest mode. Play the sine and find the null. Put the sub there.

DD
Sorry Dan but I am a little confused,
the lowest null is around 21Hz but the screenshot is pointing a 26Hz peak. Which one should I try?
Old 26th April 2014
  #11
Gear Guru
Words

The lowest thing which looks like a mode is around 27Hz, which is about right for a 6 Metre long room.
When you play this signal you should find extreme level peaks at the front and back walls and almost silence at mid length.
That's how modes roll.

DD
Old 26th April 2014
  #12
oh I see now what you mean, I am not yet familiar with the acoustic terms in English. Thanks for the suggestion, I will try this tomorrow.

I will also try changing the speaker placement and measure. I currently have them 1.70m apart and 80cm from front wall. Tweeter height at 1.22.

More I will space the 1st reflection 4"side traps and the 8"front center trap which are currently without air gap.
Old 26th April 2014
  #13
Gear Guru
Front Wall

Genelec recommend placing speakers in one of two likely best zones.
0-20cm from the Front Wall or more than 2.2 Metres from it.
Your speakers are thus in a no fly zone. Try them very close, pretty much touching the front wall and see if the big dip in your Frequency Response improves.

DD
Old 26th April 2014
  #14
I will try that, I read somewhere that it is actually pushing the SBIR on lower mid band which can be treated more easily.
maybe it was you that wrote that tbh heh
Also my mains are not ported so less worries..
Old 27th April 2014
  #15
Lives for gear
 
sheggs's Avatar
Hi, Short article here on SBIR but yes as you move the speakers closer to the wall you can move through the frequency range into the mids and make it more easily treatable - Speaker Boundary Interference Response SBIR GIK Acoustics
Old 27th April 2014
  #16
speaker position tests

So, I tried some new placement for mains, I am attaching the rew data. Check the notes on each file for more info.

Actually moving the speakers closer to front wall improved the nulls and peaks on the region 80-500hz.

Below 80hz decay times grew bigger as peak and nulls also did.

Also removed the traps behind speakers which seems to affect 220hz and 440hz nulls.
Placement also does that, different placement removes the 220 but grows bigger the 440hz or opposite.

20-80hz seems to be impossible to control with placement only.

I will experiment more with bass trap placement and maybe try adding the sub to see if I can interact down low. but after dinner..
Attached Files
File Type: mdat spkrpos1.mdat (7.81 MB, 135 views) File Type: mdat spkrpos2.mdat (5.21 MB, 116 views)
Old 27th April 2014
  #17
Gear Guru
Fix

A concrete room will have strong LF modes and SBIR. The lower LF can be affected only by very large traps. A full metre of light fibre in a frame covering the whole back wall would be great. If you don't want to lose that amount of space, VPR traps can achieve a lot in only 100mm depth. You will still need a large area of them though. You could consider a dividing wall made of relatively light panelling. The area inside would be mostly damped with cheap fibre. A huge wall panel trap.
You could install computers and other noisy equipment into a small part of the space, creating a silent CR. A very useful free recording space.

In the shorter term, playing around with the sub position and crossover should be able to shorten and weaken that sub rumble by optimising positions.
Eq is quite often used to tweak these LF monsters.

DD
Old 27th April 2014
  #18
I added 2x 4"inch and 1x 8"inch 180x65cm panels on back side center and corners and slightly made the LF peak-null problem worst. Details changed but adding the rew data anyway.
I will try the sub implementation to see what more can be achieved.

Dan, If I understand correctly what you mean is shorten the length of the room one meter on the back side and the whole space fill it with low density wool? If so what density?
I am generally using rockwool. I think 30kg/m3 is the lowest available. There is also a 13kg/m3 fiberglass available.

I don't mind losing some space if that's gonna solve the problem below 80hz.
I use a fanless pc with ssd's so no noise in my cr.

.
Attached Files
File Type: mdat TRAPING_PLACEMENT1.mdat (7.81 MB, 138 views) File Type: mdat TRAPING_PLACEMENT2.mdat (7.81 MB, 80 views)
Old 28th April 2014
  #19
Quote:
Originally Posted by DanDan View Post
Not sure that a sub is a great idea in this room, but do try it in the null of that longest lowest mode. Fire up the Signal Generator in REW, Frequency follows cursor. Play the sine and find the null. Put the sub there.

DD
I did that tests today. tried lot of positions. one worked pretty good for peak and nulls, not for decay though.Of course in addition with the LPF and phase.
rew data from subwoofer implementation attached.
Attached Files
File Type: mdat sub implementation1.mdat (7.81 MB, 326 views) File Type: mdat sub implementation2.mdat (7.81 MB, 275 views)
Old 29th April 2014
  #20
Lives for gear
 
ciro's Avatar
 

I found a little strange that you seems to have good treatment ("no FRK" - in the gif of post #1) but several strong reflection points in the ETC plot (3, 4, 5, 8, 9 ms).

Ciro
Old 29th April 2014
  #21
the front floor to ceiling corner traps and sides wall to ceiling corner traps are FRKed. Maybe these are the ones causing the reflections?
Or maybe the floor? It's just tiles cause rug was too boomy

If I am not mistaking 3ms is about 1metre, 9ms is 3 meters, what would that be?
Speakers are 2.33 apart and 1 meter from side walls

the door (metal) which is right on the side of the listener and the opposite wall are left untreated. I ve got some flutter on listening spot. Just a little.
Not sure how to fix that or reading the etc to be honest. I tried DanDan's "Measuring Room Acoustics" and Jen's "Before posting your measurement results" but still can't figure out how to read it. Any tips or a link with further explanation on it would be great!

Edit:
Oh, I just found that:
Envelope Time Curve - ETC - Impulse
Old 29th April 2014
  #22
Lives for gear
 
ciro's Avatar
 

https://www.gearslutz.com/board/6161854-post31.html

and one of your ETC plots:


(the spikes before the initial impulse possibly indicates that something itยดs wrong with the measurements, sorry if I donยดt know exactly what - maybe someone could explain better).

Ciro
Attached Thumbnails
measurements - bad sounding space-etc.jpg  
Old 29th April 2014
  #23
Gear Guru
Spikes

The information before the initial impulse is valid and relates to distortion. It is needed to relate THD and the individual harmonic distortions. From memory this is covered in the REW manual. The spikes shortly after the initial are the extra paths which the 'Impulse' has travelled. From the pictures, I'm guessing floor.
DD
Old 29th April 2014
  #24
Gear Guru
 
Glenn Kuras's Avatar
You need more trapping in corners but I would focus on the back wall. Either go with something broadband that is THICK or tuned in the back
https://www.gearslutz.com/board/7597560-post146.html
40Hz, go to hell!
Old 29th April 2014
  #25
Gear Guru
Odd

As ciro said, those reflections are very strong. We should see nothing above -20dB in the first 20mS or so.
Also all of these measurements show big vague lumps of energy below 30Hz. More so without the sub in some cases. Is the metal door ringing, or are you near a motorway or airport or train station? Your mic could have LF noise also.
How loud are your sweeps?
DD
Old 29th April 2014
  #26
Quote:
Originally Posted by Glenn Kuras View Post
You need more trapping in corners but I would focus on the back wall. Either go with something broadband that is THICK or tuned in the back
https://www.gearslutz.com/board/7597560-post146.html
40Hz, go to hell!
Maybe I should finally go with the huge trap DD suggested, a whole meter of the back room filled with wool floor to ceiling. Still trying to figure out the density, maybe 13kg/m3 fiberglass is enough for 1m deep trap.


Quote:
Originally Posted by DanDan View Post
As ciro said, those reflections are very strong. We should see nothing above -20dB in the first 20mS or so.
Also all of these measurements show big vague lumps of energy below 30Hz. More so without the sub in some cases. Is the metal door ringing, or are you near a motorway or airport or train station? Your mic could have LF noise also.
How loud are your sweeps?
DD
When setting up REW I did the necessary level checks and 75db etc
Now I see that the sweeps are around 73db without the sub.

I believe The metal door is not ringing, at least I haven't heard something all this time.
Not sure if it affects the sound in some way acting like a trap or something.
I made that myself using 1mm steel for the housing and the door. It is filled with rockwool and the inner side of metal is tarsheet covered. It is 22cm thick and around 210x90cm big

No airport , station, motorway near but around 5 meters further from the back of the room and on the ground floor is a road and cars are passing by. The sound of the sewer's metal cover can be heard if cars step on it but not the sound of the cars. Also if there is lots of wind blowing some noise can be heard from the exhaust of the fireplace which I currently use for ventilation running a motor on top of it, on the 3rd's floor roof.
But I was carefull to make the measurements on quiet backround and it really is very quiet in there. REW's meter shows around -57db just before hitting "measure".

spl meter's mic is generally considered low quality as you already know..
Attached Thumbnails
measurements - bad sounding space-image_7849.jpg   measurements - bad sounding space-image_9985.jpg   measurements - bad sounding space-image_2806.jpg   measurements - bad sounding space-image_1462.jpg  
Old 29th April 2014
  #27
Gear Guru
SPL

What is the SPL meter reading as background noise? Are you on C?
Try a sweep with the speakers turned off and see of that LF rumble is still there.
Ignore REW warnings about low level. You should be able to get a Waterfall of just the noise.
I would crank the playback level by 10dB at least when measuring to get above your mic noise if nothing else.
DD
Old 29th April 2014
  #28
Lives for gear
 
Jolida's Avatar
 

Quote:
Originally Posted by DanDan View Post
You could consider a dividing wall made of relatively light panelling. The area inside would be mostly damped with cheap fibre. A huge wall panel trap.

DD
Further explanation/links would be appreciated...
Old 29th April 2014
  #29
Gear Guru
Free

Hi Jolida. A boundary comprised of panelling, gypsum, wood, or even steel which the BBC used, on studs, obviously vibrates easily at LF. If this is damped we have a huge LF absorbent trap. avare calls this 'Invisible Alpha' He holds the world patent rights.....;-)

DD
Old 29th April 2014
  #30
Quote:
Originally Posted by DanDan View Post
What is the SPL meter reading as background noise? Are you on C?
Try a sweep with the speakers turned off and see of that LF rumble is still there.
Ignore REW warnings about low level. You should be able to get a Waterfall of just the noise.
I would crank the playback level by 10dB at least when measuring to get above your mic noise if nothing else.
DD
Yeah I am on C, slow.
The spl meter's lowest range is -50 and it is not picking something, fast or slow response.
On ipad i can see the digital spl meter around 40db, but that mic is not probably picking the extreme lows and highs.

Noise rew file attached. Things seem quiet. Two measurements made. Quiet while no cars passing by and loud with cars and sewer's cover stepped by them sounding and my dog chipped in . An average file too

Someone please increase the mdat file size..16-18mb would be wonderful instead of splitting the files every time..
Attached Files
File Type: mdat noise.mdat (5.96 MB, 157 views)
๐Ÿ“ Reply

Similar Threads

Thread / Thread Starter Replies / Views Last Post
replies: 86 views: 1211
Avatar for mrface2112
mrface2112 7th February 2006
replies: 286 views: 31988
Avatar for deedeeyeah
deedeeyeah 22nd November 2018
replies: 83 views: 17620
Avatar for africantigercow
africantigercow 11th August 2017
replies: 8 views: 1420
Avatar for Brad Lunde
Brad Lunde 30th July 2018
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
๐Ÿ–จ๏ธ Show Printable Version
โœ‰๏ธ Email this Page
๐Ÿ” Search thread
๐ŸŽ™๏ธ View mentioned gear
Forum Jump
Forum Jump