The No.1 Website for Pro Audio
Valley at 100hz, Partial Success!
Old 4th February 2019
  #1
Gear Head
 
Tristana's Avatar
 

Valley at 100hz, Partial Success!

Hey all-

I recently acquired a Sonarworks XREF20 microphone for use with their Reference 4 corrective eq software, and noticed a terrifying valley at 100hz on my first pass through.

https://www.dropbox.com/s/sr0uc4tbsg...fore.png?raw=1

I poked around the forum here for a bit, learned about SBIR, and huzzah!
After placing my monitors much closer to the back wall, the valley improved by ~6dB:

https://www.dropbox.com/s/b2e6mnbm8l...fter.png?raw=1

Things certainly sound better in the low-end, even more-so with Ref4's correction applied.

I'm curious as to your opinions on a couple things:
1. Is this a small enough deviation to let corrective EQ do the rest of the work in a home studio environment?

2. What further treatment options might further help the freq response? Absorption panels behind the monitors to further reduce SBIR? Are the corner traps or window perhaps contributing to the problem?

Here is what the space looks like:
https://www.dropbox.com/s/43ieqselg2...20PM.jpg?raw=1

The room is 9’4” x 15’5” ft (2.8 x 4.7m), with a height of 7' 9.5" ft (2.4m).

My listening position is 5'5" (1.65m) from the wall and forms an equilateral triangle with my monitors.
Old 5th February 2019
  #2
Lives for gear
 
Jens Eklund's Avatar
A 6 dB dip might be fixable with DSP. Best way to find out is to try it and check the results (also check the distortion values in the problematic range before and after correction).

Assuming your speakers are close to the front wall, any SBIR related dips due to front wall only will be higher up in frequency, most likely above 200 Hz or so (assuming a distance of about 0,4 m or so from front of speakers to front wall), so adding treatment behind the speakers will probably not further improve the 100 Hz dip. Possibly even the opposite since any treatment behind the speakers will force you to move them further away from the front wall. Broadband treatment behind the monitors might on the other hand improve imaging if it removes early reflections bouncing of the front wall and this depends on the directivity of your monitors (small monitors radiates a large portion of sound also to the rear in the lower midrange and below).


To further improve your situation, you need to identify the cause of the problem. More here:

Need help to tune my control room. (First Measures Included)

Axial Mode..
Old 5th February 2019
  #3
Quote:
Originally Posted by Tristana View Post
I'm curious as to your opinions on a couple things:
1. Is this a small enough deviation to let corrective EQ do the rest of the work in a home studio environment?
Frequency response is one thing but I'm wondering if the desk in front of the speakers is hurting your impulse response. Can you use Room EQ Wizard and post a .mdat file?

Quote:
Originally Posted by Tristana View Post
2. What further treatment options might further help the freq response? Absorption panels behind the monitors to further reduce SBIR? Are the corner traps or window perhaps contributing to the problem?
I am guessing something like the GIK Monster Panel on the side walls would help.
Old 5th February 2019
  #4
Gear Addict
 

Quote:
Originally Posted by Tristana View Post
I recently acquired a Sonarworks XREF20 microphone for use with their Reference 4 corrective eq software, and noticed a terrifying valley at 100hz on my first pass through.
It's very common to have a dip at 100Hz! So rest assured you're in good company

I had a similar issue in my own room. What helped to solve it was:
  • Moving my speakers up against the front wall (pushing SBIR to a higher, more treatable frequency).
  • More bass trapping.
  • Adding a subwoofer.
  • Use of DSP.


Quote:
Originally Posted by Jens Eklund View Post
A 6 dB dip might be fixable with DSP. Best way to find out is to try it and check the results (also check the distortion values in the problematic range before and after correction).
Good point about distortion resulting from use of DSP. I use Dirac Live to help 'fix' a 100Hz dip in my room, which it does surprisingly well at the expense of distortion - see plot below (the 1kHz peak is present with/without DSP).

Thankfully this is only just above the noise floor in my room so it's not audible - at least not to my ears - but it's something I always now keep in mind when using DSP: an improvement in one area nearly always leads to a problem in another; the trick being to best manage the compromise
Valley at 100hz, Partial Success!-left.jpg
Attached Thumbnails
Valley at 100hz, Partial Success!-left.jpg  
Old 5th February 2019
  #5
Lives for gear
 
Jason Foi's Avatar
 

Hmmm... i must be missing something. She said the dip at 100hz filled in 6db after moving the speakers. If it was SBIR, wouldnt the null have changed frequency all together. Seems like she should be looking to ETC on this one to me.
Old 5th February 2019
  #6
Lives for gear
Try moving your microphone on length, width, and height axis. Take notes of changes. It will help you identify your problem.

For example, by moving on length axis, if the frequency doesn't change it's a room mode, related to these walls.
If the freq. changes it's a SBIR, related to...I don't know yet.

Seeing your pic, I would start by replacing these foams by a proper treatment.

J.
Old 5th February 2019
  #7
Lives for gear
 
Jens Eklund's Avatar
Quote:
Originally Posted by Jason Foi View Post
Hmmm... i must be missing something. She said the dip at 100hz filled in 6db after moving the speakers. If it was SBIR, wouldnt the null have changed frequency all together. Seems like she should be looking to ETC on this one to me.
Did you see the graphs?

The ETC is often not very useful when tracking low frequency related issues.
Old 5th February 2019
  #8
Lives for gear
 
Jens Eklund's Avatar
Quote:
Originally Posted by JayPee View Post
Seeing your pic, I would start by replacing these foams by a proper treatment.
+1
Old 5th February 2019
  #9
Lives for gear
 
Jason Foi's Avatar
 

Quote:
Originally Posted by Jens Eklund View Post
Did you see the graphs?

The ETC is often not very useful when tracking low frequency related issues.
No, im not at a computer.
Old 5th February 2019
  #10
Lives for gear
 
Jason Foi's Avatar
 

And, now i looked at graphs and pictures. No treatment what so ever. My bad. +2 to jaypee
Old 6th February 2019
  #11
Lives for gear
 
akebrake's Avatar
 

Quote:
Originally Posted by Jason Foi View Post
... If it was SBIR, wouldnt the null have changed frequency all together...
Difficult to conclude from the uploaded (Sonar works) measurement plot.
The shape of the ”valley” have changed. There are also modes involved...

Moving the speaker forward means the distance to all boundaries are changing more or less. The woofer also moves away from the axial length standing wave null. Mode 3.0.0 (109Hz) null happens around 0.78m from the front wall. (17% of room length)

Small movements of speaker (or listener) makes a huge difference when a room is only sparsly treated.

I was curious to see what it looked like in REW Room Sim. Listener pos not changed(?)
Just swap between the two speaker positions pics 1& 2

Note: The softw. cannot model small corner traps or a mid wall window but rather an average front wall abs coeff. (Estimated by me )

Best
Attached Thumbnails
Valley at 100hz, Partial Success!-dist.jpg   Valley at 100hz, Partial Success!-close.jpg  
Old 6th February 2019
  #12
Gear Head
 
Tristana's Avatar
 

Thank you for all the responses, I'm working my way through them!

Quote:
Originally Posted by johannburkard
Can you use Room EQ Wizard and post a .mdat file?
- What measurement settings should I use to generate this? The default SPL measurement 0 to 200hz sweep (or should I go to 20khz)?
- Do I take the measurement from my listening position, and is an omni-directional mic suitable?
- Will I need to calibrate the software SPL meter with a physical one, and what meter might you recommend?
- Is it important to set up a loopback for the timing coefficient?

Quote:
Originally Posted by JayPee
Try moving your microphone on length, width, and height axis. Take notes of changes. It will help you identify your problem.
What measurement software would you recommend? The graphs in my initial post were created from 36 measurements around my listening position using the Sonarworks Measure software; I'm uncertain how those results were interpolated but the program doesn't allow one-off measurements.

Quote:
Originally Posted by JayPee
Seeing your pic, I would start by replacing these foams by a proper treatment.
I take it then that proper corner traps are the best first step for acoustic treatment? Would GIK Tri-traps be a significant improvement or should I look to something else?
Old 6th February 2019
  #13
Lives for gear
I much prefer REW - Room EQ Wizard Room Acoustics Software for measurement's results. Much more infos.


Worth reading these threads as well:

Measuring Room Acoustics

Before posting your measurement results

Well, everything is better than foam.
So yes, GIK Tri-traps would be an improvement. But I'd encourage you to DIY.

Corners are good because you treat partially 2 walls in 1 shot.
But, from my experience and many many threads I read, I found center of wall is the best place to start.

In your case, I'd start by backwall, and early reflection points (cloud, and your L & R walls).
The more surface treated, the better.

If you don't want to end with 50 cm of treatment on each walls (inc. ceiling) I'd recommend to read this thread, specially Jens' posts:

Velocity based vs. pressure based absorbers


Once you post your measurements (post the .mdat, will be easier for @ akebrake to do his job hehe) and we can help you further.

Cheers,

J.

Last edited by JayPee; 7th February 2019 at 12:25 AM..
Old 6th February 2019
  #14
Lives for gear
And, above all these recommandations I'd suggest to hire a professional acoustician of course.
Old 6th February 2019
  #15
Quote:
Originally Posted by Tristana View Post
- What measurement settings should I use to generate this? The default SPL measurement 0 to 200hz sweep (or should I go to 20khz)?
I always do 20-20000, with a 1 M length (or more).

Quote:
Originally Posted by Tristana View Post
- Do I take the measurement from my listening position, and is an omni-directional mic suitable?
Yes to both.

Quote:
Originally Posted by Tristana View Post
- Will I need to calibrate the software SPL meter with a physical one, and what meter might you recommend?
I didn't and quite frankly, I don't care.

Quote:
Originally Posted by Tristana View Post
- Is it important to set up a loopback for the timing coefficient?
I don't think I did that but I had REW measure my soundcard at some point for a calibration profile.

Quote:
Originally Posted by Tristana View Post
I take it then that proper corner traps are the best first step for acoustic treatment? Would GIK Tri-traps be a significant improvement or should I look to something else?
Nothing against tri-traps but I have moved most of my GIK panels twice or three times last year and Monster Bass Traps in the corners will probably be not much different but inherently more flexible.
Old 6th February 2019
  #16
Gear Head
 
Tristana's Avatar
 

Alright, attached *should* be an MDAT containing 3 measurements (Left, Right + Sum, 20-20khz sweep of 1m length from my listening position)
Attached Files
File Type: mdat Tristana_Feb6.mdat (7.39 MB, 65 views)
Old 7th February 2019
  #17
I think the frequency response isn't bad overall, it's quite symmetric. That's without the dip from 70-100 Hz.

Your desk is in the way acoustically though, and it shows.

The decay time in the room is a bit uneven but I suppose that will get better once you have some thick absorption panels to experiment with.
Old 7th February 2019
  #18
Gear Head
 
Tristana's Avatar
 

Quote:
Originally Posted by johannburkard View Post
I think the frequency response isn't bad overall, it's quite symmetric. That's without the dip from 70-100 Hz.

Your desk is in the way acoustically though, and it shows.

The decay time in the room is a bit uneven but I suppose that will get better once you have some thick absorption panels to experiment with.
Thank you for the analysis!

What particular factors were you looking at to determine the desk is a problem?
Old 7th February 2019
  #19
Quote:
Originally Posted by Tristana View Post
What particular factors were you looking at to determine the desk is a problem?
The impulse response tab.
Old 8th February 2019
  #20
Lives for gear
Yes even better than ETC is ETC filtered.
But according to your last measurement, there is no problem here (no specular reflection).

Your problems are your room modes, calculated by amroc, it looks like this:

1 36.49 Hz D1 1-0-0 ax
2 61.25 Hz B1 0-1-0 ax
4 71.46 Hz D2 0-0-1 ax
5 72.98 Hz D2 2-0-0 ax
11 109.47 Hz A2 3-0-0 ax
13 122.5 Hz B2 0-2-0 ax
19 142.92 Hz D3 0-0-2 ax
21 145.96 Hz D3 4-0-0 ax
35 182.45 Hz F3# 5-0-0 ax
36 183.75 Hz F3# 0-3-0 ax
54 214.38 Hz A3 0-0-3 ax
57 218.94 Hz A3 6-0-0 ax
76 245 Hz B3 0-4-0 ax
84 255.43 Hz C4 7-0-0 ax
112 285.83 Hz D4 0-0-4 ax
119 291.91 Hz D4 8-0-0 ax
137 306.25 Hz D4# 0-5-0 ax
166 328.4 Hz E4 9-0-0 ax

Check your waterfall and spectrogram and compare your measurements to the values above.

Treat the surface concerned. But not before reading this thread I linked in my previous post.

Cheers,

J.
Old 8th February 2019
  #21
Lives for gear
 
akebrake's Avatar
 

Modal demo

Quote:
Originally Posted by Tristana View Post
Alright, attached *should* be an MDAT containing 3 measurements ...
Fine! Your mdat looks like valid measurements.

The WaterFalls show Strong Room modes as JayPee said which is part of the "valley problem".

Measured Room modes (long sloping ridges in a WF) seem to correlate well with calculated modal frequencies.
Shown here as coloured bars along the freq scale. Same colours as the modes in REW Room Sim post 11
(Room dimensions can be inserted in the EQ-tab /Modal analysis
Valley at 100hz, Partial Success!-left-wf-w-comment.jpg

Cursor at 72Hz (blue) Mode 0.0.1 (floor-ceiling). Cloud needed.
Close to that 73Hz (red) mode 2.0.0 (front-back) and
a little lower 61 Hz (green) mode 0.1.0 (left-right walls) Corner traps

I believe Sonarworks needs a little "help" to do it's job by treating some modes for a start (and also first reflections).

Is this a possible option?

What’s the rear of the room like? Pic?
Is there a window/opening on the right wall? Are the walls of similar density? Brick/gypsum/ wood?

Cheers
Attached Thumbnails
Valley at 100hz, Partial Success!-left-wf-w-comment.jpg  
Old 8th February 2019
  #22
Gear Head
 
Tristana's Avatar
 

Quote:
Originally Posted by akebrake View Post
Is this a possible option?
I don't have the finances to immediately treat everything in one pass but I can certainly make a checklist and work towards treatments one step at a time.

Quote:
What’s the rear of the room like? Pic?
Here's the front again, after populating with some instruments (...additional reflection points, I know! These were present during my REW measurements)
Dropbox - front.jpg

And the back as requested:
Dropbox - back.jpg

It's pretty barren atm, and there is the a 3x2.5 ft alcove to the exit you can see to the right. Been meaning to put a movable barrier there.

Quote:
Is there a window/opening on the right wall?
There is. Would balancing this out be best approached by LF absorption on the opposing wall, or perhaps a removable plywood barrier to close off the window?
Old 8th February 2019
  #23
Lives for gear
 
akebrake's Avatar
 

Quote:
Originally Posted by Tristana View Post
...Here's the front again, after populating with some instruments (...additional reflection points, I know! These were present during my REW measurements)
OK
Quote:
And the back as requested...
Thanks!

Quote:
It's pretty barren atm, and there is the a 3x2.5 ft alcove to the exit you can see to the right. Been meaning to put a movable barrier there.
Hard to say how much the alcove / window contributes to the "valley" at this point in time. There is a lot going on and I'll dig a little deeper.

Best
Old 2 weeks ago
  #24
The graph reminds me of my room a bit. The null at 100hz and the peak around 150Hz.
My room is much more unsymetrical and has alot more acoustics in it. But we still have that 100hz null and 150hz peak.
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
Forum Jump
Forum Jump