Building a small control room with built-in assymetry

Hi there, I would like to build a small music & video production room (with mixing capabilities if acoustics allow) in a spare space that we have in the studio. Room dimensions : 489 cm (L) x 305 cm (w) x 309 cm (h). I started by throwing a bit of fiberglass here and there and finally started reading as much as possible about acoustics and got hooked ! Now, I'd like to make this little room sound as good as possible. The room shape is not symetrical and the ceiling has a reversed U shape. I don't plan to be loud as hell but sometimes I need to pump the bass to check the low-end on mixes and/or provide the clients with a bit of power. Neighbours : my studio mate next door and an artisan on the upper floor (we are in the basement, -2 level). Testing equipment : Neutrik measurement microphone, Prism Orpheus interface/mic pre, REW. Speaker : PSI A14-M. Floor was already installed when I started working on the acoustics - it's basically 4cm fiberglass on the floor topped by floorboards on wood. Budget for upcoming development : $500-$1000 // more if result justifies. Here are some SketchUp pics : The basic room layout (as I found it) The ceiling shape I've put fiberglass in the reversed U shapes Then I've built a first corner basstrap Then a second - on the right, trying to keep some sort of symetry Now, I am about to build a third one, in the back ... and now comes the question : what's the best option for the ceiling ? I've imagined this ... but I guess this would eat up too much highs Should I then go for a solution like this ? or like that ? or would that be even better ? ... this option would lower the ceiling but ease its installation ... REW graphs will follow + full SketchUp file. ... A huge thank you in advance for your advice !! Cheers,

Ceiling : this would be the easiest (technically), but would a second insulation layer really make that much of a difference (sound-proofing-wise and acoustically-wise), compared to the loss of ceiling height ?

I am also thinking of this option : 2 side wood panels + fiberglass and a cloth that I could eventually fit with Heimholz studs ?

REW graphs : 1. The room as I got it 2. The room with the front left membrane absorber 3. The room with both front membrane absorbers ... and the waterfall view

I am a total newbie to the whole acoustics plot, so any help is warmly welcome!! From what I've read so far: - the room proportions are important to start on a good basis - the crucial frequency range is 10-300Hz - I should create a reflexion free zone at the listening spot - diffusors are not going to help (too bad, they look cool ;-) unless you have a large room - you can't get enough bass trapping - Heimholz resonators can be very effective to treat specific bass frequencies but are tricky to tune and need to be placed at the exact energy peak of the targetted frequency - membrane resonators (which I tried to build in the corners) are effective to treat specific frequencies but have a larger Q than Heimholz resonators - to further sound-proof the ceiling, I would need to use massive material (which I won't be able to do) - the reverberation time should ideally not exceed 15-20 ms and should be as consistent over the whole frequency spectrum as possible My questions: - is my understanding of the above right? - is there a chance I get a decent acoustic result out of a space like this (room proportions, asymetry, etc.)? - I have defined the sweet listening spot by using the "70% of the front wall width" distance rule, is it really the sweetest spot in this case? - where should I place my monitors, taking the non-flat profile of the front wall into account? - is it best to mount basstraps that are basically as much fiberglass as possible in the corners with a fabric to protect them, or mount membrane resonators (with as much fiberglass as possible anyways)? - John's "Corrected Heimholz spreadsheet" looks very interesting : I could build Heimholz absorbers that would reflect higher frequencies at the same time (and potentially look great)? - what would be the best ceiling option (see SketchUp grabs in previous posts)? - trying to interprete my REW graphs, I am surprised to see a drop under 60 Hz, is it due to multiple frequencies cancellation or have I reached the point of too much basstrapping (already?) Wow ... that's quite a few ones ;-)

... and the sketchUp file of the current state + ceiling options

From what I've read so far: - the room proportions are important to start on a good basis - the crucial frequency range is 10-300Hz - I should create a reflexion free zone at the listening spot - diffusors are not going to help (too bad, they look cool ;-) unless you have a large room - you can't get enough bass trapping - Heimholz resonators can be very effective to treat specific bass frequencies but are tricky to tune and need to be placed at the exact energy peak of the targetted frequency - membrane resonators (which I tried to build in the corners) are effective to treat specific frequencies but have a larger Q than Heimholz resonators - to further sound-proof the ceiling, I would need to use massive material (which I won't be able to do) - the reverberation time should ideally not exceed 15-20 ms and should be as consistent over the whole frequency spectrum as possible

Perfectly correct on all of those, except for the last one. Strictly speaking, your reverb time depends on the room volume, and is also related to the frequency. General rule of thumb: 200 to 400 ms in the range 200 Hz to 4kHz is your starting point, rising by up to 300 ms at 50 Hz, and maybe trailing off by 100 ms in the high end, above 4k. Give or take, adjust for room volume, etc. The "15ms to 20ms" figure is for RFZ designs, and refers to the minimum target delay between direct sound and first reflection arriving at the engineer's ears.

- is my understanding of the above right?

Yup! Except as noted above.

- is there a chance I get a decent acoustic result out of a space like this (room proportions, asymetry, etc.)?

Symmetry is critical for the front half of the room. If the room is significantly asymmetrical in the front half you'll find it hard to get an accurate stereo image / sound stage, and your mixes might not translate well. I would try to make the room symmetrical if possible. Apart from that, you should be able to get the room sounding reasonable.

- I have defined the sweet listening spot by using the "70% of the front wall width" distance rule, is it really the sweetest spot in this case?

Where did you find that "rule"? I haven't heard of that one before. The general recommendation is to have your ears located at roughly 38% of the room depth (distance from front wall to back wall), since that is the theoretical best position for modal response, with the least problematic modal issues. However it is NOT written in stone! It's just a guideline. A good place to start from.

- where should I place my monitors, taking the non-flat profile of the front wall into account?

I would really try to make that front wall flat. Those cavities on each side of the central pillar are going to be a big problem, acoustically. However, you could probably put those spaces to good use as part of a soffit system. Especially if you are planning an RFZ design.

- is it best to mount basstraps that are basically as much fiberglass as possible in the corners with a fabric to protect them, or mount membrane resonators (with as much fiberglass as possible anyways)?

Personally, I would go with the absorption, not the membrane traps. Absoprtion is broadband and effective to quite low frequencies. Membrane traps are tuned.

- John's "Corrected Heimholz spreadsheet" looks very interesting : I could build Heimholz absorbers that would reflect higher frequencies at the same time (and potentially look great)?

That's exactly what John's slot wall design does. :)

- trying to interprete my REW graphs, I am surprised to see a drop under 60 Hz, is it due to multiple frequencies cancellation or have I reached the point of too much basstrapping (already?)

Most likely that is simply the roll-off at the bottom end of the range of your speakers. What speakers did you use for those tests? It could also be an issue with the specs of your measurement mic: what mic did you use? Another option is the interface you used. Add to all of the above the inherent difficulty of accurately detecting, computing and plotting low frequencies. But my money is on the speakers. That's probably beyond the low end of their range. In fact, the data below about 30 Hz is pretty useless anyway: you don't need to be running sweeps down to 4 Hz. Setting your lower limit to 20 Hz is fine. It is also pretty clear that your membrane traps are doing something pretty drastic at around 160 Hz: the room response got WORSE after those were installed. It looked a lot more reasonable before. I'd take them out and replace them with simple superchunk or corner-absorber style bass traps. Also, the scale on your waterfall plot is way off, and it is impossible to see the important part of the data. - Stuart -

Thanks a ton, Stuart ! Reverb time : Thanks a lot for the precisions. Symmetry : Left vs right cavities = 141 cm vs 113 cm. Should I add 28 cm of concrete-type material in the left cavity to get closer to symmetry ? As the back of the room is not symmetrical anyway, is is worth the hassle ? Soffit mount : wow, that's a whole other topic to investigate. Is it realistic to study this and get a proper soffit mount in DIY mode ? Compensating those cavities would mean a loss of 73.5 precious centimeters ... Monitors placement : I read that 70% rule in a book (will try to find the references and post them here). I'll go with the 38% and see how it goes. Membrane absorbers vs basstraps : I noticed a drastic improvement in the bass frequencies RT60 after installing the corner membrane absorbers but, yes, curve-wise it doesn't seem to improve anything, really. Curiously, none of those membrane absorbers were targetting this 160Hz ? I need some clarification here : basstrapping would mean (in my case) replacing the front wood panel by fabric, right ? What we call basstraps is basically filling corners with fiberglass, is that correct ? REW graphs : Sorry for the messy waterfall thing - I am new to this and find it quite difficult to display what I want to see. I am using PSI 14m speakers (http://www.psiaudio.com/product/active-monitors/a14-m). I know those small speakers have a roll-off from 50hz down, but never suspected it to be so drastic. Let's add the sub and see what happens. Microphone is a Neutrik and soundcard is a Prism Orpheus (I wouldn't suspect any of those 2). Heimholz reflectors (or does this design have an official name ?) : Would it be possible to leave a few centimeters in-between the timber slats and the cloth backing and/or the insulation, or is it absolutely necessary that the timber slats are in close contact with both ? Again, thank you so much for sharing your knowledge ! Cheers,

... one more thing about those corner membrane resonators : I could turn them into broadband absorbers by letting the fiberglass touch the membrane, is that correct ? This would let the high-mids and highs be reflected by the wood panel and save the hassle/time/money loss of removing it...

I just added the sub to the monitor setup and - yes - the graphs obviously reflect this change. Moving the sub around also changed the bass response a lot ... one more aspect to take into account // the plot thickens !

Left vs right cavities = 141 cm vs 113 cm. Should I add 28 cm of concrete-type material in the left cavity to get closer to symmetry ?

Not necessarily. If you are going to soffit-mount your speakers then those cavities will be behind the soffit face, and thus are not really part of the room, acoustically. As long as your soffit faces are laid out symmetrically, you should be OK.

As the back of the room is not symmetrical anyway, is is worth the hassle ?

Shouldn't be a problem. It is much more critical that the front half of the room is symmetrical. The rear half is not so important.

Soffit mount : wow, that's a whole other topic to investigate. Is it realistic to study this and get a proper soffit mount in DIY mode ?

Yup! It isn't as hard as it looks. The basic principle is simple: A large sheet of something rigid and massive, such as thick plywood, OSB or MDF, with a hole cut in it for the speaker to poke through. Make a frame to support it all rigidly, and bingo! Speaker soffit. OK, it's not quite that easy, but it isn't as complicated as some people try to make it. Of course, you do need to make sure that your speakers can be soffit mounted: As long as they are not rear ported, and do have a bass roll-off control, then you should be OK. Yours are front ported and do have the roll-off, so they should be fine, theoretically.

Compensating those cavities would mean a loss of 73.5 precious centimeters ...

Not really: Most of the body of each speaker would be inside the cavities. That space is unusable as it is anyway, so it isn't really "lost". Just put to better use. :)

What we call basstraps is basically filling corners with fiberglass, is that correct ?

Generically, yes. Even though the term can also refer to tuned traps, it seems to be used more commonly to refer to large deep fibrous absorption, which could be mineral wool, fiberglass, acoustic foam, etc.

I know those small speakers have a roll-off from 50hz down, but never suspected it to be so drastic.

Yup! http://www.psiaudio.com/sites/default/f ... chdata.pdf That's common in small speakers, especially reflex (ported). Looking at your graphs, I'm wondering if you might have the roll-off control set incorrectly?

Microphone is a Neutrik

Which model? It should be an omnidirectional mic for acoustic measurements.

Heimholz reflectors (or does this design have an official name ?) :

Normally just called a "slot wall" here.

I could turn them into broadband absorbers by letting the fiberglass touch the membrane, is that correct ?

Not really: that would just damp the resonance of the membrane a bit. Yes, it would flatten the Q a bit, but not really enough to be worthwhile.

This would let the high-mids and highs be reflected by the wood panel and save the hassle/time/money loss of removing it...

The mids and highs are already being reflected by the wood panel! It's the lows that are a problem.

Moving the sub around also changed the bass response a lot ... one more aspect to take into account // the plot thickens !

Yup! :) You should look for the location where the sub gives the smoothest bass response. One way to do that easily is to it in reverse: set up the sub at the place where your head will be when you are mixing, then walk around all over the room and listen carefully, trying to find the best spot. You'll have to crawl around, of course, since the sub will probably have to sit on the floor. Once you find the spot where the bass is smoothest, that's where the sub should go. - Stuart -

Hey Stuart, Thanks a ton for your input ! You sound like the Obi-Wan Kenobi of acoustics ;-) The Padawan that I am has a few more questions but I will first post new graphs, etc. to talk on a measurable basis. The Neutrik microphone looks like a genuine measurement one (with its "earthworks"-type of shape), and was lent to me with an hardware RTA (which I don't use). My speakers were put in full-range mode / now that I have added the appropriate sub, the graphs look way more comprehensive. Following your previous reply, I decided to unmount those corner panels and stuff even more fiberglass in the corners + add a plastic protection with a finishing cloth on top. Pictures will follow. The slot wall concept looks so brilliant that I would like to try it! A few questions here : - does it also work for long distances from the wall/ceiling or is it more precise when staying in the 5-20 cm range? - does the insulation need to touch the cloth/slats or can space be allowed between the slats and insulation? - to get the optimal result, should I take measurements at the wall position where I project to place it? Again, thanks a ton for sharing your knowledge and experience! Pascal

You sound like the Obi-Wan Kenobi of acoustics

I wish! :) :oops: Not nearly, but thanks for the kind words!

The Neutrik microphone looks like a genuine measurement one (with its "earthworks"-type of shape), and was lent to me with an hardware RTA (which I don't use).

OK, sounds like it should be the right type then.

- does it also work for long distances from the wall/ceiling or is it more precise when staying in the 5-20 cm range?

It is tuned, since it is based on the same old Helmholtz principle. So changing the cavity depth (distance inside, from slat to back panel) changes the frequency of each slot. You can therefore "tune" each slot by also varying the gap size and thickness of the slats. And you can make it broadband by varying the depth: make it deeper at one end than at the other. In other words, it will work for a range of cavity depths, yes, but you should take care to figure out the highest and lowest frequencies to ensure that you are covering a useful range.

- does the insulation need to touch the cloth/slats or can space be allowed between the slats and insulation?

Yes, it should touch. You might find these interesting: http://www.johnlsayers.com/HR/index1.htm http://www.sae.edu/reference_material/a ... encies.htm

- to get the optimal result, should I take measurements at the wall position where I project to place it?

Yep! :) - Stuart -

those links !! thanks a ton !

I've made some decisions on this project : - go for the lower ceiling option (easier to mount) - unmount my panel absorbers and go for regular basstraps (with potential additional slats) - leave the cavities as is for the moment and take time to study the soffit-thing more in detail before I go for it. I've made pictures and will try to share as soon as I have a minute.

A few pictures of my first basstrap constructions ;-)

A few more pics. Basstrap with cloth (I may add slats according to the measurements once the ceiling is completed). Making of the ceiling modules: I've added a layer of black thin plastic just to help with the consistency of the cloth color (as it's white, the structure and mineral wool were a bit visible)

I am trying to see the end of this project. The lessons learnt so far : - measuring is a must - next time, I'll start right away with proper measuring equipment - for a small room, base the treatment on absorbtion - basstrapping the corners is probably the best starting point - you'll use more mineral wool than expected - make physical measurements of the room once, well and precisely - making a SketchUp plan is very helpful - as much as possible, think about the whole concept from early measurements to the finishing touches - details take much more time than expected - think about the ventilation, power, audio and all other cables & connections early enough

Very true, and very useful tips! - Stuart -

I've put this plastic layer underneath the finishing cloth and I must say I am very impressed on how live the room sounds! With an all-mineral wool ceiling, I was expecting a much tighter sound but I actually like it better as it is. Pictures & graphs will follow... I find it quite difficult to handle all the parameters to keep those acoustics under control : the measurements with or without me standing/sitting at the sweet spot are pretty different ... same thing when I add a couch at the back of the studio, etc. In any case THANKS A LOT for all the shared knowledge and time spent providing guidance, tips, background info, etc.! I have a new control room (a bit more serious this time) project coming soon - will post for sure.

I find it quite difficult to handle all the parameters to keep those acoustics under control : the measurements with or without me standing/sitting at the sweet spot are pretty different ... same thing when I add a couch at the back of the studio, etc.

Another possibility is that maybe you are not getting the mic in the exact same position very time you measure. That is important. It's a good idea to measure the precise distance from the four walls, floor and ceiling to the top of the mic, and make sure that it is back in that exact same position every time, accurate to within a 1/2" or so. If the mic is in a different place for each measurement, then you can't really compare the graphs as they are showing the results for different locations. - Stuart -

Thanks Stuart, The mic is mounted on a stand and wouldn't move at all...

The mic is mounted on a stand and wouldn't move at all...

So the measurement mic is a permanent fixture in your room? You never, ever move that mic, as you take the sofa in and out, or move the treatment around the room? It will always be in that exact same location, which is the same place where your head is as you are mixing? I'm not quite sure how you can do that... - Stuart -

Yes, the mic is mounted on a stand and does not move or very very slightly (I put a mark on the ground). I got consistent measurements until now, so I would not question the microphone placement too much. I will post some graphs to express what I'm trying to understand for your consideration.

Attached are waterfalls with/without myself in the room ... Empty: With me