Odd shaped stone wall basement studio acoustical challenges

[Gunner]

Hi everyone-
Thanks in advance for the help. I got a lot of advice from this forum while the project was under construction. Now I'm trying to get the monitors set up and I have a few questions.

Please, no comments on the layout and construction choices. I realize this is not even close to an ideal monitoring room. The design goals for the studio were to make a room that everyone would want to be in, to make a room that sounds awesome to play in, and to make an acoustically accurate monitoring situation, in that order. The first two goals seem to be good, the goal now is to make the room as accurate as possible given the current situation.

Some details on the construction: Floor is concrete. Walls are about a 50/50 mix of rough pointed stone and mahogany. Ceiling is open joists with a mix of 2 and 4" OC 703 panels suspended about 3 inches from the upstairs floorboards. The hope was for those panels to control bass without having any OC on the walls and to stop reflections into drum overheads from the 8'6" ceiling. It sounds awesome in the room, I think because of the combination of the absorptive ceiling and the diffusive qualities of the stone.

I included a floor plan and some pics of the finished room. Speakers are behind the acoustic cloth screens behind the desk. If you remove the cloth screens it's just an open cavity. The area under the desk is open too. I attached a pic with the screens removed. The yellow baffles are OC 703 panels I was experimenting with. The stands are PVC pipe filled with sand.

So far I've experimented with moving the monitors forward and back in the cavity and it seems they are best almost all the way to the cloth. I threw panels of OC 703 into the cavity and tested after each one. The more the better there. I put some OC on the left and right walls at the reflection points. There's one 4" panel of OC on the back wall center. That's about as much stuff as I really want to put up on the walls. I included plots from REW of what I have done so far. Couldn't attach the mdat file, too big. I used the behringer ECM8000 mic at the listening position.

I think in the end I'll fill almost all the space in the cavity with OC 703, including 2" on the front wall behind the speakers. I figure that'll stop it from resonating and function as a bass trap in a place I don't have to look at it or walk around it. I'm also considering building a proper little stud wall to the left of the left speaker so the cavity the speakers are in is at least symmetrical.

Two questions:

1) My plot seems to say that everything from 500hz on down is WAY quieter than anything from 500hz up. My ears say there's no way that 500 on down is 30db quieter than everything else. What's going on here? Anything else of note in the waterfall or RT60? Nothing is too simple to point out, I'm not an expert. There are high and low shelf filters on my Mackie HR824's. Conceivably I could boost the lows and reduce the highs to balance the plot out to some degree.

2)What does anyone suggest to improve the room? I'm open to any suggestions, keeping in mind the room is built and is not changing.

If anyone is building a similar basement studio in an old house please feel free to contact me for advice. This project started with a filthy moist basement with cast iron pipes at eye level through the whole thing. It was a challenging project and there's lots of things I would do differently that would be cheaper, faster and better if I had it to do over again. If you're in the new england area I can point you towards some contractors who are excellent and some to stay away from. The architect is Paul Privitera at http://www.tera-form.com. It would be great to give a little back to the forum.

Thank you!

Recording-Studio-Boston.jpg

Recording-Studio-Design-Architect.jpg

Basement-Recording-Studio.jpg

Screen Shot 2013-02-01 at 5.44.16 PM.jpg

Machinegun Studios_Master Plan_01.pdf

speaker cavity.JPG

SPL plot.jpg

RT60 plot.jpg

Waterfall Plot.jpg

[Soundman2020]

Please, no comments on the layout - the goal now is to make the room as accurate as possible given the current situation.

I think you are asking the impossible. You say you want us to comment on how to fix the room acoustically, but at the same time you say you don't want any comments on how to fix it acoustically!

To be brutally honest, and also brutally ignore your request to not talk about the layout: The layout is the problem. You cannot get an acoustically usable environment without symmetry, and you don0t have symmetry, since that was part of the original decision to make the place look nice, at the expense of acoustics.

Short of fixing the layout, there is nothing that can be done to make that into a place where mixes will translate well. I could go into the details of why that is so, but it sounds like you don't want to hear the explanations about what is wrong with the layout, so I'll refrain from that, and just comment on the existing treatment.

The hope was for those panels to control bass without having any OC on the walls

The graphs show that this was not successful. The hope didn't work out, since it was based on unrealistic expectations and misunderstanding of the principles of physics.. Room modes terminate in corners, not against surfaces, so bass trapping is dozens of times more efficient when placed in corners than when places in the open. There simply is not enough area on the ceiling of a room to provide the amount of bass trapping that a small room needs. You don't say what the dimensions of that room are, so I can't tell you how man sabins of absorption you would need to bring it within ITU specs, but generally it is much more than twice the area of the ceiling, for a typical room.

Speakers are behind the acoustic cloth screens behind the desk. If you remove the cloth screens it's just an open cavity.

That falls under "layout", so I'll refrain from commenting on that, since you asked us not to, except to say that it is a big part of the problem you are seeing in your graphs, and the problems you are hearing.

The stands are PVC pipe filled with sand.

That's good!

So far I've experimented with moving the monitors forward and back in the cavity and it seems they are best almost all the way to the cloth.

That's correct, yes. That will force some of the artifacts up to a high enough frequency where they are less noticeable. Particularly SBIR and comb filtering should be less noticeable. They won't go way, of course, but they should be less offensive like that. Did you set the bass roll-off control correctly for quarter-space radiation?

I threw panels of OC 703 into the cavity and tested after each one. The more the better there.

Yes, that's correct. Placing more directly between the speakers and the wall would also help with SBIR and combing.

Couldn't attach the mdat file, too big.

Post it on a file-sharing service, so we can download it and check it. The graphs you show are insufficient.

I figure that'll stop it from resonating and function as a bass trap

It won't do either of those, actually. 2" flat against the wall is nowhere near enough to affect bass much. Rather, it will help somewhat to reduce the SBIR effect, and also to reduce somewhat the comb filtering and other phasing issues caused by having speaker in close proximity to reflective surfaces. It will also help mildly with front-wall first-order reflections, but not much since the speaker geometry is not set up correctly to start with.

I'm also considering building a proper little stud wall to the left of the left speaker so the cavity the speakers are in is at least symmetrical.

The problem is not on the left. Take a look at your photos and diagram, and you should be able to see that the issue with symmetry is on the right, not the left.

1) My plot seems to say that everything from 500hz on down is WAY quieter than anything from 500hz up. My ears say there's no way that 500 on down is 30db quieter than everything else. What's going on here?

Not necessarily quieter, just badly distorted response. You are showing swings over over 40 dB in the low end, so there's either some major modal issues going on, or major reflection issues going on, or both.

However, it almost looks like you have the speaker response controls set up incorrectly.

Is that graph taken from the left speaker or the right speaker? My guess is that it comes from the left speaker.

Anything else of note in the waterfall or RT60?

Impossible to say: there isn't enough detail on there, as the display parameters are not set correctly. Something is going on, yes, and the basic plot is typical of serous modal response issues, but there's not enough detail visible in the graph to be able to tell.

There are high and low shelf filters on my Mackie HR824's. Conceivably I could boost the lows and reduce the highs to balance the plot out to some degree.

That would the wrong thing to do. You switches should be set up as follows:

ACOUSTIC SPACE: Position "A" quarter space
LOW FREQ: Position "37Hz".
HIGH FREQ: Position "0".

The reason for this is simple: your speakers are positioned in the worst possible location: room corners. Mackie provides a way for compensating for the drastic power imbalance, so adjusting those switches will help, but there's nothing that a speaker manufacturer can do beyond that to fix all the other issues. So that will at least fix the power imbalance, but the other artifacts will remain.

The lack of bass is probably due to having the speaker so close to the front wall: it is a rear-ported speaker, and those usually need a lot of space behind them to perform properly.

2)What does anyone suggest to improve the room? I'm open to any suggestions, keeping in mind the room is built and is not changing.

Bluntly, if you are not willing to change the layout, then there is nothing that can be done. No amount o acoustic treatment can fix the problems created by the layout.

The architect is...

My congratulations to the architect, and also to the contractors: That place looks very nice, and the workmanship looks very good. If it wasn't for the layout issue and the lack of treatment, then it could be used as place for mixing too. Unfortunately, it is about as good as it can be, acoustically, since you don't want to change the things that need changing.

Sorry to be so blunt, but I'm sure you came here to find the truth about what is wrong with your setup, and what can be done to fix it. That's what I'm giving you, nothing more.


- Stuart -

[Gunner]

So, to make this thread useful for other people who are trying to make an odd shaped stone room as usable as possible I'm going to comment on a few things soundman 2020 wrote.

First though, I've finished the room and the monitoring sounds pretty fantastic. Despite soundman's opinion you actually can make a clearly noticeable difference in a room like this with treatment. All is not lost if you room is not symmetrical. I ended up adding a 4" OC absorber on the walls to the right and left and skipped the back wall. I built the extra stud wall to make the cavity symmetrical if not the room and stuffed the cavity almost completely full of OC 703.

The plot in the post above was actually an REW error. I rebooted the program and the plot was totally different with the same setup. All that stuff below 30hz disappeared, which makes sense because the mackies don't go that low. The rest of the plot smoothed out and the waterfall completely changed. Keep that in mind if REW is ever giving you weird results. I wasted quite a few hours on wacky plots before I figured it out.

On soundman's post.

Quote:
Please, no comments on the layout - the goal now is to make the room as accurate as possible given the current situation.
I think you are asking the impossible. You say you want us to comment on how to fix the room acoustically, but at the same time you say you don't want any comments on how to fix it acoustically!

To be brutally honest, and also brutally ignore your request to not talk about the layout: The layout is the problem. You cannot get an acoustically usable environment without symmetry, and you don0t have symmetry, since that was part of the original decision to make the place look nice, at the expense of acoustics.

Short of fixing the layout, there is nothing that can be done to make that into a place where mixes will translate well. I could go into the details of why that is so, but it sounds like you don't want to hear the explanations about what is wrong with the layout, so I'll refrain from that, and just comment on the existing treatment.

My latest tests confirm that there's a lot you can do to a room like this to make it better if not perfect, aside from knocking down the walls and starting again with symmetry. My final plot was accurate to about a 12db range, and without the treatment I added it was more like 50db.

The graphs show that this was not successful. The hope didn't work out, since it was based on unrealistic expectations and misunderstanding of the principles of physics.. Room modes terminate in corners, not against surfaces, so bass trapping is dozens of times more efficient when placed in corners than when places in the open. There simply is not enough area on the ceiling of a room to provide the amount of bass trapping that a small room needs. You don't say what the dimensions of that room are, so I can't tell you how man sabins of absorption you would need to bring it within ITU specs, but generally it is much more than twice the area of the ceiling, for a typical room.

The ceiling panels are in corners everywhere the ceiling meets the wall. The panels also made a huge difference in the sound of the room to track in, both in reducing boominess and largely eliminating phasey reflections from the ceiling. Again, the goal was to improve the room, not make it perfect.

Quote:
1) My plot seems to say that everything from 500hz on down is WAY quieter than anything from 500hz up. My ears say there's no way that 500 on down is 30db quieter than everything else. What's going on here?
Not necessarily quieter, just badly distorted response. You are showing swings over over 40 dB in the low end, so there's either some major modal issues going on, or major reflection issues going on, or both.

However, it almost looks like you have the speaker response controls set up incorrectly.

Is that graph taken from the left speaker or the right speaker? My guess is that it comes from the left speaker.

This was all REW errors in the plot. They all went away with a reboot.

Bluntly, if you are not willing to change the layout, then there is nothing that can be done. No amount o acoustic treatment can fix the problems created by the layout.

I didn't ask how to fix them, I asked how to improve them given what the room is.

To be brutally honest, and also brutally ignore your request to not talk about the layout: The layout is the problem. You cannot get an acoustically usable environment without symmetry, and you don0t have symmetry, since that was part of the original decision to make the place look nice, at the expense of acoustics.

Actually it was designed to be a nice place to be in, in terms of usability and workflow. It's possible we could have made a tiny symmetrical drywall box to sit in, but what's the point of having a hyper accurate room to mix in if no one wants to come inside it and make music? In the end it came out to be a super awesome room that makes people want to buy recording time as soon as they walk in. The acoustic treatment I did made a noticeable difference in the accuracy of the monitoring, totally worth doing. Hopefully this thread helps some other people working on the same sort of thing.

[Soundman2020]

I'm glad it worked out for you, but how about posting the MDAT file from the final REW test? If it's too big to post here on the forum, then upload it to a file-sharing services, such as Dropbox, and post the link here. It's very easy to make the claim that the room worked out great, but then not provide any evidence to demonstrate it.

Also, some photos of the finished room would be great, so folks can see what you did, and how it turned out.

I ended up adding a 4" OC absorber on the walls to the right and left

Which you originally said you did NOT want to do...

I built the extra stud wall to make the cavity symmetrical

So you DID take my advice, even though you said you didn't?

The plot in the post above was actually an REW error.

Ummm.... No it isn't. The data is perfectly valid and correct, and shows exactly what one would expect from such a room.

All that stuff below 30hz disappeared, which makes sense because the mackies don't go that low.

Yes they do. Mackie states the following in the owner's manual: "±1.5 dB, 37 Hz to 20 kHz". Ported speakers generally have roll-off of 12 dB/octave (although it is possible to have other rates), and your REW graph shows you pumping out nearly 80 dB at 37 Hz, so at half an octave below that, 28 Hz, the level would be just 6 dB lower, or around 72 dB. A full octave below the "cut-off", at 19 Hz, your speakers would still be putting out 68 dB. So yeah, your speakers actually DO put out plenty of energy below 30 Hz, and below 20 Hz too. You say that the 30 Hz stuff "disappeared": that means that you screwed up. The graph shows what would be expected of that room and those speakers, so if you then did something that made all the energy below 30 Hz "disappear", that means you screwed up your measurement or signal chain somehow. In reality, the energy did NOT disappear: it is still there.

My final plot was accurate to about a 12db range,

That isn't possible in that room. Sorry.

This was all REW errors in the plot. They all went away with a reboot.

Once again, if all the EXPECTED information that is clearly visible in those graphs "went away", then you screwed up. Those graphs are showing what would be expected of the room you describe. If that all "went away" as you claim, then you screwed up.

In the end it came out to be a super awesome room

I'm finding that VERY hard o believe. The only way you'll convince anyone (other than yourself) is to post the MDAT files (both "before" and "after"). "super awesome" would mean that it meets the ITU BS.1116 spec: I'm betting it doesn't even get close. Not even vaguely in the ball-park.


- Stuart -

[Soundman2020]

I'm glad it worked out for you, but how about posting the MDAT file from the final REW test? If it's too big to post here on the forum, then upload it to a file-sharing services, such as Dropbox, and post the link here.

Also, some photos of the finished room would be great, so folks can see what you did, and how it turned out.

Five years have passed, and still no MDAT, no data, no graphs, no photos, no nothing to back up the claims.

I wonder why?

In light of the lack of evidence, I'll stand by my original advice, and my original comments.


- Stuart -