The No.1 Website for Pro Audio
Improving the acoustics of a semi-open control room (REW measurements included)
Old 1 week ago
  #1
Gear Addict
 
withintheflux's Avatar
Advice wanted for a semi-open control room (REW measurements included)

Hi all! Major thanks to all of the talented sounders who've generously shared their knowledge. I've been working on my control room for several months but feel now that I've hot a wall. I'm hoping some of you will have insights or suggestions as I try to take things even further. I'd love a second opinion about how my measurements are looking and perhaps some tips on how to combat the remaining issues.

a few details about the room (see the attached photos): Given It's the front room of my home, I'm locked into a few features that may not be ideal. One is that half of the back wall is opened into a foyer/front room. Another is that the left wall is windows so I can't hang panels there.

Dimensions:
15’1” deep (with an extra 8' of foyer on the back-right side)
11’7” wide
8’3” high

I've already done a great deal of testing and invested in lots of G.I.K. products. The current panelling is as follows:

-soffit traps with "Range Limiters" on both front wall to wall corners
-2 Monster bass traps as ceiling clouds
-monster bass traps with range limiters on back left wall to wall corner
-Monster bass trap with scatter plat on back wall
-Monster bass trap with scatter plate straddling back wall to ceiling corner
-4" panels straddling both side wall-to-ceiling corners
-4 2" panels along right wall
-24" X 24" Monster bass trap straddling back right ceiling corner.
-a few "skinny" monster traps supplementing things.

As for my testing, these are some of my major conclusions:

-With the Amphion One18s that I'm using, SBIR was VERY sever with speakers on my desk. So I had to get them as close to the front wall as physically possible. This made an enormous improvement to the low end, vastly more than all of my treatment combined.

-I found that somewhere around 38% off of the front wall was a good listening position. Any closer and I found myself in a pretty severe null around 107 hz

Remaining problems:
-My main issue is that I've got some big holes between 220 and 300hz. I've tried all several variations of adding other panels or swapping them out. Nothing seems to effect those nulls. Overall those are making things sound a little bit gutted/thin.

-around 850 hz I've got some weirdness that I presume to be related to the desk. I've currently got Neoprene not he desk to reduce reflections. Not sure what else to try.

Big appreciation in advance for anyone who takes a moment to chime in. THANKS

-Blake
Attached Thumbnails
Improving the acoustics of a semi-open control room (REW measurements included)-img_3973.jpg   Improving the acoustics of a semi-open control room (REW measurements included)-img_3975.jpg  
Attached Files
File Type: mdat ControlRoom.mdat (5.52 MB, 22 views)

Last edited by withintheflux; 1 week ago at 04:02 AM..
Old 1 week ago
  #2
Lives for gear
Quote:
Originally Posted by withintheflux View Post
Remaining problems:
-My main issue is that I've got some big holes between 220 and 300hz. I've tried all several variations of adding other panels or swapping them out. Nothing seems to effect those nulls. Overall those are making things sound a little bit gutted/thin.
Looking at the spectrogram and filtered IR, It looks like something in this range is reflecting around 14ms, which would be about a 14 foot flight path. It seems like the ceiling or floor are the most likely candidates. Looking at the pic, maybe your ceiling cloud isn't wide enough?

In this range, 4" wedge foam should be effective enough to at least figure out where it's coming from. This is much easier to work with than a full panel that has to be hung up, you can put some double stick tape on it. Then you can decide if it's worth it to get another panel if you find the source.

Although looking at the 1/3 sliced ETC, both show this reflection being as high, or higher than the source...which might mean that something is resonating. Acoustic guitars maybe? Or is that a fireplace behind the desk? Maybe stuff it with some fluffy or dacron etc? Even an old sleeping bag...

On the 800hz sliced ETC it looks like ~4ms is the biggest reflection/diffraction, then one at 14 again.

4" foam can also be somewhat useful for hunting down desk reflections, and possibly some sort of absorber between the desk and the speakers. The caveat is I've found wedge foam to only work well with a low incidence angle over about 400hz or so. As you move up in frequency and to an obtuse angle it seems the waves diffract off of the wedges. A better choice at these frequencies IME is the really plush 'cuddle fabric', it's like fleece but with longer, softer fibers. You might pick some of that up to experiment with.
Attached Thumbnails
Improving the acoustics of a semi-open control room (REW measurements included)-spec-sf.jpg  
Old 1 week ago
  #3
Gear Addict
 
withintheflux's Avatar
THANK YOU for the thoughtful reply, Ryan! I suddenly realize how much I have to learn about properly measuring and assessing this stuff.



Quote:
Originally Posted by RyanC View Post
Looking at the spectrogram and filtered IR, It looks like something in this range is reflecting around 14ms, which would be about a 14 foot flight path. It seems like the ceiling or floor are the most likely candidates. Looking at the pic, maybe your ceiling cloud isn't wide enough?
Interesting. I had wondered about the width of my ceiling clouds now that my speakers are wide, although visually they do seem to cover the mid point between myself and the speakers. Also, the flight trip off of that ceiling reflection would be 10' maximum, so that seems to rule out the 14' reflection you've noticed.


Quote:
Originally Posted by RyanC View Post
Although looking at the 1/3 sliced ETC, both show this reflection being as high, or higher than the source...which might mean that something is resonating. Acoustic guitars maybe? Or is that a fireplace behind the desk? Maybe stuff it with some fluffy or dacron etc? Even an old sleeping bag...
weird. Is it possible that it's a room modes and not reflections causing such an increase of energy there? When I send a tone of 298hz, I can very easily find a null in my listening position where the tone almost vanished. Moving my head back or forward increases the volume drastically.

Yes, that is a fireplace. Currently I have it stuffed with Auralex corner traps. There is one acoustic guitar int he back of the room but it doesn't seem to resonate when sending a tone in that range, although I'm not sure that such a basic test disqualifies it as a culprit.

Ryan, would you mind show me how you got that spectrogram you've posted? It's very useful but I can't seem to create it on my end in REW.

I'll do some experiments and follow up. I really appreciate it.
Old 1 week ago
  #4
Gear Addict
 
withintheflux's Avatar
Interestingly, 14' seems to be the exact flight path distance of the right speaker reflecting off of the left wall (which is glass windows). I'll experiment with making sure that is being treated with a freestanding 2" panel.
Old 1 week ago
  #5
Lives for gear
Quote:
Originally Posted by withintheflux View Post
Interestingly, 14' seems to be the exact flight path distance of the right speaker reflecting off of the left wall (which is glass windows). I'll experiment with making sure that is being treated with a freestanding 2" panel.
Actually looking at sliced ETC's again (this is the filtered IR, with the 1/3 oct filter setting). It looks like the first, and largest one is actually more like 5ms.

Could be the floor?

It's a little tough to get *exact* timings from the sliced ETC because what you see is rounded there (lack of faster treble to have a more precise point).

It shows up in both L and R, and more so in L...
Old 1 week ago
  #6
Gear Addict
 
withintheflux's Avatar
Quote:
Originally Posted by RyanC View Post
Actually looking at sliced ETC's again (this is the filtered IR, with the 1/3 oct filter setting). It looks like the first, and largest one is actually more like 5ms.

Could be the floor?

It's a little tough to get *exact* timings from the sliced ETC because what you see is rounded there (lack of faster treble to have a more precise point).

It shows up in both L and R, and more so in L...

ah ok. I would assume that's the desk, TBH. 5 ms would be something like 5 feet, right? If so, probably the desk. Incidentally, there's no neoprene on the left side but there IS not he right.

Blake
Old 1 week ago
  #7
Lives for gear
 

Quote:
"5 ms would be something like 5 feet, right? If so, probably the desk."
You are probably already aware of this, but here's a reminder anyway: That 5 feet is the DIFFERENCE in path length between the direct sound and the reflected sound. So the total flight distance for that reflection is the distance from the speaker to the mic, plus 5 feet. I'm pretty sure you already got that: the comment is more for other people who might be following your thread, and not understanding the 5 foot thing.


- Stuart -

Last edited by Soundman2020; 1 week ago at 02:22 AM..
Old 1 week ago
  #8
Gear Addict
 
withintheflux's Avatar
Quote:
Originally Posted by Soundman2020 View Post
"5 ms would be something like 5 feet, right? If so, probably the desk." You are probably already aware of this, but here's a reminder anyway: That 5 feet is the DIFFERENCE in path length between the direct sound and the reflected sound. So the total flight distance for that reflection is the distance from the speaker to the mic, plus 5 feet. I'm pretty sure you already got that: the comment is more for other people who might be following your thread, and not understanding the 5 foot thing.


- Stuart -
Thanks Stuart! I actually was NOT clear on that, so very much appreciated. Easy to get your head twisted when thinking about this. so the reflection in question is around 5 feet away from the mic. That is actually the distance to me left wall which is glass windows. I'd better experiment with covering that to see how it's impacted.

Question for RyanC (Or Stuart if you happen to know): What settings got you the spectrogram image that you posted in your original response? I somehow can't seem to see my data in just that way.

And Stuart, I wonder if you might chime in to offer your sense of whether this reflection we've discussed might be responsible for the problematic dips I'm seeing in my SPL graph in both Left and Right speakers. I'm hoping I can address it and it's not merely room modes that can't be tamed.

THANKS.
Old 1 week ago
  #9
Lives for gear
Quote:
Originally Posted by withintheflux View Post
Thanks Stuart! I actually was NOT clear on that, so very much appreciated. Easy to get your head twisted when thinking about this. so the reflection in question is around 5 feet away from the mic.
Just to make sure here because it can be tricky to explain in just text, the delay on the reflection is the red (or yellow) minus the blue. I was spacing this out yesterday, (did not sleep)...So not just distance to the mic.

A couple other things in your hunt here, an acoustic wavefront expands as it travels. So you can't think of reflections like a pool ball or laser. A better analogy is to think of the reflection point as being at least a bit more like a 2nd pair of speakers.

Also an object can only reflect frequencies that are the same size or smaller than the objects shortest dimension that is perpendicular to the wavefront. So for 250hz your desk would have to be 55" in it's shorter dimension. IOW a desk that is 55" x 55" can reflect a 250hz wave, a desk that is 20"x55" won't reflect that frequency- instead it will diffract around it.

I'll post a picture of the spectrogram settings here a little later.
Attached Thumbnails
Improving the acoustics of a semi-open control room (REW measurements included)-reflections.png  
Old 1 week ago
  #10
Lives for gear
Here are the settings I used
Attached Thumbnails
Improving the acoustics of a semi-open control room (REW measurements included)-screen-shot-2019-12-05-2.32.11-pm.png  
Old 1 week ago
  #11
Gear Addict
 
withintheflux's Avatar
Thanks Ryan! super useful.

I've been doing more tests. I have found ways to improve a dip 235 hz. Largely by using a deeper panel at my left wall first reflection and by laying 2 1'X4" monster traps on the ground under both 1st reflection points. Those both gave me unmistakable improvements.

Now my biggest problem which I can't seem to fix is a dip at 287hz, which is more pronounced from the right speaker. I'm starting to suspect room modes rather than any specific reflection. I'd love to get your eves on these new measurements if possible.

Blake
Attached Files
File Type: mdat BlakeRoom2.mdat (5.53 MB, 10 views)
Old 1 week ago
  #12
Lives for gear
Quote:
Originally Posted by withintheflux View Post
Now my biggest problem which I can't seem to fix is a dip at 287hz, which is more pronounced from the right speaker. I'm starting to suspect room modes rather than any specific reflection. I'd love to get your eves on these new measurements if possible.
This is a pretty high frequency for it to be modal, especially with the back of the room being open like it is... Still seeing that same ~5-6ms in the sliced etc in this range, and then something again at 14, but improved for sure.

I would try the ceiling to the sides of the cloud, and the bare part of the rear wall first.
Old 1 week ago
  #13
Lives for gear
 

Quote:
Originally Posted by withintheflux View Post
Thanks Stuart! I actually was NOT clear on that, so very much appreciated. Easy to get your head twisted when thinking about this. so the reflection in question is around 5 feet away from the mic. That is actually the distance to me left wall which is glass windows. I'd better experiment with covering that to see how it's impacted.
Sort of! If you look at the diagram that Ryan posted, it might get a bit clearer. The reflection isn't 5 feet from the mic, but 5 feet FURTHER from the mic than the speaker is.

There's a "trick" you can do with a piece of string to help you find where a reflection is coming from. Measure the direct straight-line distance from the speaker to the mic, and add 5 feet to that. Cut a piece of string that long (add an inch or so on each end so you can attach it!). Tape one end the the face of the speaker (carefully! Don't damage the speaker!), and tape the other end to something sturdy set up exactly where the mic tip was. Now you have a piece of string that is sagging down a lot in the middle, and it is exactly the right length for the reflection. All you have to do now is to move the sagging loop in the string all around the room until it touches a hard reflective surface, and ... Bingo! You found the culprit! However, there might be several such surfaces, so there's one more item to add: you need to get the "saggy loop" stretched out so that both sides of it are practically straight lines, by pulling gently with your finger, and you want your finger to be roughly in the middle of the string, so that as your finger touches the suspicious surface, the taught string makes roughly the same angle on each side of your finger (the side towards the speaker, and the side towards the mic). That's the orange line on Ryan's diagram. You can see that the bounce point is about half way along the string, and the angle that the string makes is about the same on each side. That orange line in Ryan's diagram is showing the floor bounce, but in your case it might also be one of the walls, or the ceiling, or something else in the room.

But there might also be a "double bounce" or "second order bounce", which is when the sound bounced off two surfaces before it got to the mic (not shown in Ryan's diagram), as well as the third-order bounce shown in yellow in that diagram. You'll need more fingers, and some acrobatic skills, to get your piece of string lined up for those, but fortunately they are rather unlikely: the most likely thing is that you only have a first-order bounce, which you can find fairly easily with this "string trick".


Going back to something you said earlier:

Quote:
SBIR was VERY sever with speakers on my desk. So I had to get them as close to the front wall as physically possible. This made an enormous improvement to the low end, vastly more than all of my treatment combined
Yup! That's why we recommend that speakers should always be set up like that in a small room! SBIR is a big issue, and fixing it is fairly easy: just move the speakers!

Also:

Quote:
-I found that somewhere around 38% off of the front wall was a good listening position. Any closer and I found myself in a pretty severe null around 107 hz
There's a procedure I developed for helping to find the best spot for the listening position, that I call the "Walking Mic" test. If you don't mind getting very bored, and have some time on your hands, then here's how to do it: http://www.digistar.cl/Forum/viewforum.php?f=9 It's the third one down on that list.

Quote:
And Stuart, I wonder if you might chime in to offer your sense of whether this reflection we've discussed might be responsible for the problematic dips I'm seeing in my SPL graph in both Left and Right speakers. I'm hoping I can address it and it's not merely room modes that can't be tamed.
I doubt that your issues at 230 and 280 Hz. That's above the Schroeder frequency for your room (about 130 Hz, assuming decent treatment, which you seem to have), so it's unlikely to be modal. Plug your room dimensions int Bob Gold's calculator:
http://www.bobgolds.com/Mode/RoomModes.htm


It could be a reflection, but my money would be on either SBIR or desk-related issues. If you happen to have a large piece of 4" (or thicker) insulation on hand, then lay that flat on top of the desk, completely covering it, and see what results you get.

By the way, I'm seeing your reflections at more like 6ms, not 5. I get around 6.5 ms, 7.3 ms, 10.4 ms, 14.8ms, 20.9 ms, and a few others. See the image below. That's for your left speaker. The second image is for both speakers, where you can see other stuff as well (lots of it!), but the biggest peak is still around 6.5ms. So when you do the "string trick", make it about 88" longer than the speaker-to-mic distance. (88" is the distance that sound travels in 6.5ms, assuming typical temperatures). If you want to test the others too, then the distances would be 98 inches for 7.3ms, 140 inches for 10.4ms, 200 inches for 14.8 ms, and 282 inches for 20.9 ms. The last couple are likely double-bounces, or perhaps even triple bounces.

However, not everything that looks like a reflection there is actually a reflection, and not everything needs great attention.... Overall, your room isn't too bad: With only a couple of exceptions, you are within about +/- 9 dB for each speaker by itself at 1/12th octave smoothing, which is pretty reasonable for a home studio. You might find that the "walking mic" procedure can get you to a better result without needing too much additional treatment...


- Stuart -
Attached Thumbnails
Improving the acoustics of a semi-open control room (REW measurements included)-withintheflux-rew-reflections.png   Improving the acoustics of a semi-open control room (REW measurements included)-withintheflux-rew-reflections-lr.png  
Old 1 week ago
  #14
Gear Addict
 
withintheflux's Avatar
Quote:
Originally Posted by Soundman2020 View Post
Sort of! If you look at the diagram that Ryan posted, it might get a bit clearer. The reflection isn't 5 feet from the mic, but 5 feet FURTHER from the mic than the speaker is.

There's a "trick" you can do with a piece of string to help you find where a reflection is coming from. Measure the direct straight-line distance from the speaker to the mic, and add 5 feet to that. Cut a piece of string that long (add an inch or so on each end so you can attach it!). Tape one end the the face of the speaker (carefully! Don't damage the speaker!), and tape the other end to something sturdy set up exactly where the mic tip was. Now you have a piece of string that is sagging down a lot in the middle, and it is exactly the right length for the reflection. All you have to do now is to move the sagging loop in the string all around the room until it touches a hard reflective surface, and ... Bingo! You found the culprit! However, there might be several such surfaces, so there's one more item to add: you need to get the "saggy loop" stretched out so that both sides of it are practically straight lines, by pulling gently with your finger, and you want your finger to be roughly in the middle of the string, so that as your finger touches the suspicious surface, the taught string makes roughly the same angle on each side of your finger (the side towards the speaker, and the side towards the mic). That's the orange line on Ryan's diagram. You can see that the bounce point is about half way along the string, and the angle that the string makes is about the same on each side. That orange line in Ryan's diagram is showing the floor bounce, but in your case it might also be one of the walls, or the ceiling, or something else in the room.

But there might also be a "double bounce" or "second order bounce", which is when the sound bounced off two surfaces before it got to the mic (not shown in Ryan's diagram), as well as the third-order bounce shown in yellow in that diagram. You'll need more fingers, and some acrobatic skills, to get your piece of string lined up for those, but fortunately they are rather unlikely: the most likely thing is that you only have a first-order bounce, which you can find fairly easily with this "string trick".


Going back to something you said earlier:

Yup! That's why we recommend that speakers should always be set up like that in a small room! SBIR is a big issue, and fixing it is fairly easy: just move the speakers!

Also:

There's a procedure I developed for helping to find the best spot for the listening position, that I call the "Walking Mic" test. If you don't mind getting very bored, and have some time on your hands, then here's how to do it: http://www.digistar.cl/Forum/viewforum.php?f=9 It's the third one down on that list.

I doubt that your issues at 230 and 280 Hz. That's above the Schroeder frequency for your room (about 130 Hz, assuming decent treatment, which you seem to have), so it's unlikely to be modal. Plug your room dimensions int Bob Gold's calculator:
http://www.bobgolds.com/Mode/RoomModes.htm


It could be a reflection, but my money would be on either SBIR or desk-related issues. If you happen to have a large piece of 4" (or thicker) insulation on hand, then lay that flat on top of the desk, completely covering it, and see what results you get.

By the way, I'm seeing your reflections at more like 6ms, not 5. I get around 6.5 ms, 7.3 ms, 10.4 ms, 14.8ms, 20.9 ms, and a few others. See the image below. That's for your left speaker. The second image is for both speakers, where you can see other stuff as well (lots of it!), but the biggest peak is still around 6.5ms. So when you do the "string trick", make it about 88" longer than the speaker-to-mic distance. (88" is the distance that sound travels in 6.5ms, assuming typical temperatures). If you want to test the others too, then the distances would be 98 inches for 7.3ms, 140 inches for 10.4ms, 200 inches for 14.8 ms, and 282 inches for 20.9 ms. The last couple are likely double-bounces, or perhaps even triple bounces.

However, not everything that looks like a reflection there is actually a reflection, and not everything needs great attention.... Overall, your room isn't too bad: With only a couple of exceptions, you are within about +/- 9 dB for each speaker by itself at 1/12th octave smoothing, which is pretty reasonable for a home studio. You might find that the "walking mic" procedure can get you to a better result without needing too much additional treatment...


- Stuart -
Stuart...thank you SO MUCH. Really appreciate you taking your time to give educated insights.

I pursued the string trick and wrote down potential culprits for that first reflection. My best guess is a second order bounce off of the ceiling and then desk. Every single order bounce that I can find are wall surfaces that are behind the listening position and are angled such that I would expect sound to bounce into the back of the room. One exception is the possibility of the topmost wall directly to my sides.

As for the dips I'm seeing close to 300 Hz you are totally correct in your guess about SBIR. I ruled out the desk causing those problems by covering it with foam and noticing that it didn't alter that dip even a tiny amount. In fact, none of my tests had improved that dip significantly UNTIL I started moving the speakers in small increments. I found that the most dramatic improvement came when I pulled the speakers off of the from wall 2 inches. I noticed that it didn't alter the low end much at all (gladly) but improved that 300Hz area by 5-10 DB. I then experimented with the speakers left/right dimension, finding that bringing them inward a touch helped. In terms of stereo image It's less ideal (these Amphion One18s seem to prefer greater width). But the improvement to the freq response seems worth it to me.

I'm attaching REW measurements that include the before and after. If you happen to notice anything that is indeed worse now, let me know. Much appreciated.

Blake
Attached Files
File Type: mdat BlakeRoom3.mdat (11.05 MB, 2 views)
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