The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Room correction ! 8 db dip at 70-85 hz . Condenser Microphones
Old 10th April 2019
  #1
Here for the gear
 

Room correction ! 8 db dip at 70-85 hz .

What is causing this big dip in my room! Have 4 corners treated with chunk bass traps, my back(behind the listener) wall has a qrd diffiusser 25cm the deeper line with size of 10 cm per line and over all size is 120cm x 200cm ! Roof is treated with 20 cm rock wool coveread with acoustic fabric! And floating floor as well! I only have a big glass dor in the middle of the side wall! Thnx for any help!
Attached Thumbnails
Room correction ! 8 db dip at 70-85 hz .-9ee56ab5-0baa-46e6-808e-f29ca119eadc.jpg  
Old 10th April 2019
  #2
Lives for gear
 

Possibly SBIR, but it's hard to say from just looking at a single frequency response graph that has been smoothed to a high degree.

I'd suggest running a full set of REW tests, and posting the MDAT file here, for the experts to analyze.
Old 10th April 2019
  #3
Lives for gear
Dimensions of the room and listening position would help as well!
Old 11th April 2019
  #4
Here for the gear
 

That is not smoothed! That is the final result after doing measurements with sonarworks and an RTA Mic (ecm8000) either when i record pink noise from listening postion with that mic the graf on a given eq spectrum looks almost the same! Im guessing the door is causing the effect! Room is 3.6m wide x 5m long and 2.5m high with the treated ceiling! Speaker are adams a77x
Old 11th April 2019
  #5
Here for the gear
 

Quote:
Originally Posted by JayPee View Post
Dimensions of the room and listening position would help as well!
Room is 3.6m wide x 5m long and 2.5m high with the treated ceiling! Speaker are adams a77x
Listening postion is 1.9m from the front wall And creating the so called triangle respecting the 38% distance! Room is symmetrical except for the door on my left side and placed exactly in the midle of the room! The exact opposite side of the door is treated!
Old 11th April 2019
  #6
Lives for gear
 

Quote:
That is not smoothed!
Ummm... yes it is. Below is what an un-smoothed frequency response graph looks like... Yours is nothing like that: it is highly smoothed. (I just grabbed a graph at random from one of my clients' projects: that's not related at all to your room. It just shows what a true un-smoothed graph looks like)

And as I said, you are only showing a low-res frequency response graph, which doesn't tell you very much at all! Frequency response is not the most important acoustic data you should be looking at. Time and phase response are more important, and provide the context for interpreting the frequency response.

Quote:
That is the final result after doing measurements with sonarworks and an RTA Mic (ecm8000) either when i record pink noise from listening postion with that mic the graf on a given eq spectrum looks almost the same!
Did you use the correct calibration file for the ECM8000? The quality control on those things is atrocious. Their frequency response is all over the place. The ECM8000 is notorious for inaccuracy. That said, the problems vaguely visible in your graph do not seem to be related to the mic, but rather to the way you did the measurement, and also to the room itself.

It is impossible to use EQ to fill in a phase cancellation null, such as SBIR, or strong reflections. That's why your EQ doesn't appear to change anything on your FR graph: it won't be able to, because it is physically impossible.

Also, pink noise is not a very good way to do acoustic testing: you cannot extract an accurate impulse response from just playing pink noise over a pair of speakers, because in a typical room at typical levels, the noise floor looks pretty much the same as the test signal... so you have poor S/N ratio... thus you have poor accuracy.

Like I said: I'd suggest running a full set of REW tests, and posting the MDAT file here, for the experts to analyze.

Here's how to do that: http://www.johnlsayers.com/phpBB2/vi...hp?f=3&t=21122 .

Quote:
Room is 3.6m wide x 5m long and 2.5m high with the treated ceiling!
That's great, but we need the actual height of the REAL ceiling, not the height of any treatment you might have on it. The biggest problems in your room are caused by the hard, solid, massive boundary surfaces of the room, not by the treatment inside the room.

In any event, assuming you did mean that the real ceiling height is 2.5m, if the length of the room is 5m and the height is 2.5m, then you clearly have a massive modal problem at 68.9 Hz! And another again at 137.8 Hz. For those two, the length and height axial line up perfectly with each other, making for very large peaks and dips. You also have coincident tangentials at 83.9 Hz and 117.9 Hz, as well as other modal issues at 34.4 Hz 47.8 Hz, 95.7 Hz, 103.3 Hz.

However, you say that your problem is between 70 and 85 Hz, but that doesn't line up with any of the above. The only thing in that region is a tangential at 77 Hz, but that's not likely. Tangentials are lower intensity, but still fairly sharp (high Q).

So I'd suggest that my original "diagnosis" is still likely correct: It's probably SBIR, or maybe floor bounce. You can't fix either of those with bass traps or diffusers. It seems too broad to be floor bounce (that's usually a very narrow, deep dip), and its also a bit tool low, so my best guess is SBIR. Although it could be a combination of problems. It could also be some other form of phase cancellation, but with only your smoothed graph to go on, there's just not enough information to be sure.

Quote:
Im guessing the door is causing the effect!
Very unlikely... It is possible it could have some effect if it is a hollow-core door that happens to be tuned to that frequency, but I doubt the effect would be that big, just from a door.

Quote:
And floating floor as well!
How did you float that? It's quite possible that the floor itself is causing that dip.

Run REW, and post the MDAT. And when you do run REW, make sure that you have all EQ set flat, and all "room correction" toys turned off: you want only the actual response of the room itself.

- Stuart -
Attached Thumbnails
Room correction ! 8 db dip at 70-85 hz .-rew-unsmoothed-fr-18-22k.png  

Last edited by Soundman2020; 11th April 2019 at 06:13 AM..
Old 11th April 2019
  #7
Lives for gear
Quote:
Originally Posted by Erion View Post
Room is 3.6m wide x 5m long and 2.5m high with the treated ceiling! Speaker are adams a77x
Listening postion is 1.9m from the front wall And creating the so called triangle respecting the 38% distance! Room is symmetrical except for the door on my left side and placed exactly in the midle of the room! The exact opposite side of the door is treated!
Try to measure at 135cm from front wall (real wall). Please post the measurement (mdat would be better than screenshot).

I can already tell that your rear wall is not treated enough. It will be clearly apparent on your waterfall.

Pics would help as well.
Old 11th April 2019
  #8
Lives for gear
Hello,



Where and how are positionned your speakers : distance to the front wall, on a desk or freestanding.

The 38% rules doesn't work with a lack of modal support.

1.9 vs 3.6 vs équilateral triangle is the center and it is not good. The standing waves love this situation.
Old 11th April 2019
  #9
Here for the gear
 

That is the result before eq ing anything! I dont get it when you say that my fr looks like that! I see u have a problem with those toys but thats what ingot for the moment! I have the ecm8000 cal file !
Old 11th April 2019
  #10
Lives for gear
let the mic where he should be : in his box.

you have not treated the side walls.

your listening position is far from your speaker.

And the rest of my previous post.

Last edited by dinococcus; 11th April 2019 at 11:41 AM..
Old 11th April 2019
  #11
Gear Addict
 

Quote:
Originally Posted by Erion View Post
That is the result before eq ing anything! I dont get it when you say that my fr looks like that! I see u have a problem with those toys but thats what ingot for the moment! I have the ecm8000 cal file !
I have experience of using Sonarworks. It functions ok for the purpose of speaker correction; but it's of very little use for measuring room acoustics. You want to be using REW as others have mentioned above (Fuzzmeasure is an alternative if you're on Mac and having problems getting REW to recognise your mic).

Also Soundman2020 is absolutely right about your measurement being smoothed. For example, here are some old measurements of my room, before I finished adding acoustic treatment:

This is the measurement generated by Sonarworks (blue line shows room without Sonarworks EQ; green line shows suggested EQ curve):


And this is the same measurement taken using Fuzzmeasure, with 1/24 octave smoothing:


Quite a difference! The -10dB dip in the Sonarworks plot is really more like -30dB when shown at a higher resolution (1/24 octave smoothing compared to what looks like 1/3 or maybe 1/6). So worth noting that the frequency dip in your room is almost certainly much deeper than it appears.

PS - here's a thread explaining how I treated that low frequency dip in my room:
Big frequency dip in treated room - what are my options?
Attached Thumbnails
Room correction ! 8 db dip at 70-85 hz .-180107_measurement_the-rock-mkii.png   Room correction ! 8 db dip at 70-85 hz .-freq_full-range_24.jpg  
Old 11th April 2019
  #12
Lives for gear
 

Quote:
I dont get it when you say that my fr looks like that!
Your frequency response curve is SMOOTHED! Therefore it does not show any detail! The curve I posted is an example of what a REAL frequency response curve looks like, when it is NOT smoothed: Just the real, actual, raw data, showing everything! If you use REW to test your room, you too will see real unsmoothed frequency response curves! Then you will be able to identify the problems in your room!

Quote:
I see u have a problem with those toys but thats what ingot for the moment!
I don't have a problem with "room correction" software when it is used correctly in a properly set up and treated room, but as dinococcus said, you are not using it like that: Your room is not set up correctly, and you have not treated it correctly. Therefore, you should not be attempting to use EQ: it is not applicable to your room. First you have to set up the speakers and mix position in the correct locations, then you have to measure the room response like that, so see what the problems are, then you have to treat those problems, with suitable acoustic treatment devices placed at the correct positions around the room. Once you have all that in place, with the room properly treated acoustically. THEN you can get out your room correction software and use it for some final tweaks.

Quote:
thats what ingot for the moment!
Then get something else! I already gave you the link with the instructions for download REW (it is free), and setting it up (also free), and using it correctly to get the initial tests (also free). It will give you the data you ACTUALLY need, to determine what is wrong with the room. What you are showing right now is just the frequency response graph, which is not very useful. You ALSO need the rest f the acoustic data from the room, so you can understand the frequency response graph. Apart from not having high enough resolution, you graph also has no context: You can's see how the phase is changing at the points you suspect are problems, and you can't see how the levels of each frequency are changing over time. So there's no context to your curve, and this it isn't any use. Frequency response alone does not tell you anything useful about the room, or the treatment, or the speakers. Frequency response combined with time-domain response and phase response, tells you a hell of a lot more. That is one of the reasons why I consider many "room correction" products to be "toys". They only show and adjust frequency response, without considering the real context of that data, then attempt to fix acoustic issues with EQ, when it is actually impossible to "fix" those issues with EQ. You can ONLY fix a frequency response issue if there is no underlying serious phase issue, or time domain issue. For example, you CANNOT use EQ to fix SBIR, or strong reflections, or some modal issues, and you certainly can't use it to fix the decay times, or create a diffuse field. Those are ACOUSTIC problems, and need ACOUSTIC solutions. Only then is EQ possible.

- Stuart -
Old 11th April 2019
  #13
Here for the gear
 

Ok this is where my listening position is measured today ! Speakers are equally distanced 125 cm from each other , measured at the center of tweeter! and are aprox 1.1m from side walls ! And they are placed 40 cm from their back wall(front wall) ! And my head is sitting at around 1.9 - 2m from the same wall(the wall im looking at) distance from my chair to the back wall is aprox 3m! So me triangle is set correct to my knowledge!
Old 11th April 2019
  #14
Lives for gear
 

Quote:
Speakers are equally distanced 125 cm from each other , measured at the center of tweeter! and are aprox 1.1m from side walls !
That doesn't make sense. You said the room is 3.6m wide. If each speaker is 1.1m from its own side wall, then they must be 3.6 - 1.1 - 1.1 = 1.4m apart. But you say they are only 1.25m apart. The math doesn't work.

Quote:
And my head is sitting at around 1.9 - 2m from the same wall
So this is your current setup then? I drew that up based on what you said:
Attached Thumbnails
Room correction ! 8 db dip at 70-85 hz .-erion-speaker-setup.png  
Old 11th April 2019
  #15
Here for the gear
 

Quote:
Originally Posted by Soundman2020 View Post
That doesn't make sense. You said the room is 3.6m wide. If each speaker is 1.1m from its own side wall, then they must be 3.6 - 1.1 - 1.1 = 1.4m apart. But you say they are only 1.25m apart. The math doesn't work.

So this is your current setup then? I drew that up based on what you said:
Yes exactly! Ok may be my numbers aren’t precise but aprox are there!
Old 11th April 2019
  #16
Lives for gear
 

Quote:
Originally Posted by Erion View Post
I have the ecm8000 cal file !
Just to make sure, do you have the cal file of your ecm8000 or a generic ecm8000 calibration file?
Old 11th April 2019
  #17
Here for the gear
 

Quote:
Originally Posted by JEK3 View Post
Just to make sure, do you have the cal file of your ecm8000 or a generic ecm8000 calibration file?
Im telling the mic isn’t the problem here! Even when first put my neumann u87 at omni the graf was almost the same! The dip is there no matter what!
Old 12th April 2019
  #18
Lives for gear
 
Jason Foi's Avatar
 

Quote:
Originally Posted by Erion View Post
Yes exactly! Ok may be my numbers aren’t precise but aprox are there!
Your dip is due to SBIR, from having the speakers 40cm off the front wall is what i would wager. Push them right up against the front wall, or 100mm off the front wall with some 703 between the speaker and wall, and watch that dip go up in frequency.
Old 12th April 2019
  #19
Lives for gear
 

Quote:
Im telling the mic isn’t the problem here!
You seem to be missing the point: we are not saying that the dip is caused by the mic, but rather that the mic you have is inaccurate. It's one of the few mics where you need the correct calibration file for that specific mic, even to use it for simple, rough measurements.

As Jason said: your dip is likely SBIR. Possibly from the front wall (although it does seem a bit low for that), but it could also be from the side walls.

With your current layout, your speaker toe-in angles are a little shallow: only about 24°. If that were my room; I would spread them further apart, and move the mix position forward just a bit, then adjust the angles as needed.

Then I would measure the room using REW, and post the resulting MDAT file here on the forum, so the experts can look at your actual data, all of it, rather than just the smoothed frequency response curve on it's own, which isn't much use, and doesn't tell you anything about the room (except that it has problems).

- Stuart -
Old 12th April 2019
  #20
Lives for gear
 
PaulMac's Avatar
It's pretty similar to my Sonarworks correction before I treated my control room a tonne more!
Attached Thumbnails
Room correction ! 8 db dip at 70-85 hz .-screen-shot-2019-04-12-00.29.58.jpg  
Old 12th April 2019
  #21
Lives for gear
 
Jeff Hayat's Avatar
 

O!M!G! This thread! Is! Funny!

!
Old 12th April 2019
  #22
Here for the gear
 

Quote:
Originally Posted by Soundman2020 View Post
You seem to be missing the point: we are not saying that the dip is caused by the mic, but rather that the mic you have is inaccurate. It's one of the few mics where you need the correct calibration file for that specific mic, even to use it for simple, rough measurements.

As Jason said: your dip is likely SBIR. Possibly from the front wall (although it does seem a bit low for that), but it could also be from the side walls.

With your current layout, your speaker toe-in angles are a little shallow: only about 24°. If that were my room; I would spread them further apart, and move the mix position forward just a bit, then adjust the angles as needed.

Then I would measure the room using REW, and post the resulting MDAT file here on the forum, so the experts can look at your actual data, all of it, rather than just the smoothed frequency response curve on it's own, which isn't much use, and doesn't tell you anything about the room (except that it has problems).

- Stuart -
Well i see i have to sit and learn the REW and give you proper perspective! Sorry for my bad english!
Old 4 weeks ago
  #23
Here for the gear
 

Back with rew

So i opened rew and did a measurement and here are the files! Mic positioned at Tweeter level and at sitting point!
Attached Thumbnails
Room correction ! 8 db dip at 70-85 hz .-44c76392-1811-497e-b3c6-2630828fa3a6.jpeg   Room correction ! 8 db dip at 70-85 hz .-276a063f-2ef2-49fb-b648-396b8f82f375.jpeg  
Attached Files
File Type: mdat room measurment for gs forum.mdat (2.53 MB, 14 views)
Old 4 weeks ago
  #24
Lives for gear
Have you measured the two speakers in the same set?
Old 4 weeks ago
  #25
Here for the gear
 

Quote:
Originally Posted by dinococcus View Post
Have you measured the two speakers in the same set?
Yes both speakers on !
Old 4 weeks ago
  #26
Lives for gear
 

Quote:
Originally Posted by dinococcus View Post
Have you measured the two speakers in the same set?
Right. That's just both together. And not calibrated. We need to see each speaker individually, as explained in the instructions at the link I gave you, and REW needs to be set up correctly, in the manner explained in the instructions at the link I gave you.

But your data does look valid, which is good.
Old 4 weeks ago
  #27
Lives for gear
 

Quote:
Originally Posted by Erion View Post
Yes both speakers on !
I think you misunderstood. IN ADDITION to having both speakers on, we also need to see the separate measurements for each speaker by itself.

And REW also needs to be set up properly, as described in the link I gave you.

- Stuart -
Top Mentioned Products
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…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Similar Threads
Thread
Thread Starter / Forum
Replies
junya-eskimo / Studio Building / Acoustics
3
Okatvius / So Much Gear, So Little Time
4

Forum Jump
Forum Jump