Mixing room with (coat)hangers

[jlerhun]

Hi there !
My name is Jimmy, and after reading a lot of topics this is my first post here. Unlike most, i'm not here to ask questions about my own home studio (yet), but acting as a translator for my bandmate Christophe which doesn't speak/write in english very well. That is, we are french ...

Anyway, i hope i did not forget any of the rules for posting here. The location of the home studio is near Paris, France, more precisely in Versailles (but not near the famous castle). Our goal is to install some acoustic treatment to even the response of the room, in order to mix rock music without too much artifacts. There are problems in judging the bass content of the mixes in that room, they often translate with too much low-end and lack of low-mids on other systems. Installing proper isolation is unfortunately not possible (for various reasons : rented space, budget ...) but Chris is aware that he will have to live with any outside noise leaking in. Recording of acoustic instruments will not happen here, it's purely for songwriting and mixing. The sound level will be kept rather low, i'd say less than 80 dB SPL at mixing position.

The room :
It's a simple room in a first-floor flat, and it's also a dressing room. It's used with the largest dimension as width (for better symmetry) as most of the backwall is used as a wardrobe, full of hanging clothes. This should be quite a good absorption, i guess. The thing is, the depth of the room is rather small, almost equal to the height, and further reduced by the wardrobe.
The dimensions of the room are : 4.50m width, 2.45m length, 2.47m height. There is a window on the left wall, and an open door frame (without door) in the right wall. All walls and ceiling are hard surfaces, painted plaster (probably brick and/or concrete structure). The floor is linoleum. We did a rather detailed Sketchup model of the room, but i don't have the file here for the moment, i'll post it later. Meanwhile, there is a quick and dirty map attached, as well as a few pictures.

The setup :
The monitoring consists of a pair of Adam A7X, on stands, a few centimeters away from the front wall. They are quite widely spaced to accommodate two computer screens, about 1.5m between them. In mixing position, the ears are 105cm away from each speaker, therefore the triangle is wider that the optimal equilateral. Putting the mixing position further back would be too close to the wardrobe.

The measures :
Please find here some REW results, measured with an Apex 220 mic set at the mixing position (facing the front wall, at tweeter level, as seen in the pictures), the level was calibrated at 83dBC with a soundmeter. There are three curves : left speaker, right speaker, and both. With my very limited experience i'm not sure if it looks good or bad, with all those large swings in the frequency response at some room modes : peaks around 66-70Hz, 93Hz, and deep gaps around 83Hz, 110-120Hz, 225Hz, 735-775Hz. At least there is not much ringing in the waterfall besides the bass-reflex port.

The plan and the budget :
I guess that some amount of bass-trapping is in order, to even out the aforementioned peaks and gaps in the low-end frequency response. I suppose that the upper front corners would be a good place to start. The window and door gap will be covered with a dual layer of heavy curtains, to reduce reflection there and improve symmetry.
We also plan to use some absorption at the mirror points, especially on the front wall just behind the speakers, and either on the ceiling (acoustic foam) or the floor (carpet).
The budget for all that is rather low, about 400€ (~ 550$US), but DIY is definitely an option.
For bass trapping, it could be either a couple of those kits http://www.thomann.de/fr/the_takustik_cbt_37.htm or maybe superchunks. For absorption, something like this : http://www.thomann.de/fr/the_takustik_hilo_n70.htm (size and thickness to be determined), or some DIY panels.

The questions :
- Is there a way to tame the frequency response of the room, especially in the low-end, or is it a desperate room mode issue ?
- how much bass trapping is required in that room, and where is the best place to put it ?
- how much mid/high absorption is required, and where is the best place to put it ?
- how good/bad could it end up, given the small budget ? Would some EQ help in "finishing the job" ?

Thank you for reading this long post !

[Soundman2020]

Hi Jimmy (and Christofe!), and welcome to the forum!

Our goal is to install some acoustic treatment to even the response of the room, in order to mix rock music without too much artifacts.

That's a reasonable goal.

The dimensions of the room are : 4.50m width, 2.45m length, 2.47m height.

Ouch! That's not good. Length and height are practically identical, and length is closely related, so there will be strong modal issues in there...

And the REW data confirms it. Wow! Not surprisingly, there's a huge modal peak at around 70 Hz, since axial modes in all three of your axes line up almost perfectly. That is going to take sooooommmmeeeeee treatment! And frankly, I don't think the room is big enough to be able to treat that.

It's used with the largest dimension as width (for better symmetry)

... which places the engineer's head right in the middle of that awful mode.... Not a god place at all.... I would suggest rotating the room 90° so the speakers fire down the long dimension, which would allow for slightly better acoustics. And get rid of the clothes: hang them in another room: You'll never get good symmetry with those hanging all down one wall.

The monitoring consists of a pair of Adam A7X, on stands,

Excellent! Fantastic choice of speaker. Are those stands very heavy? They don't look very heavy. If not, make them heavy. Or replace them with something heavy. Many people use bricks to build up a stand, or hollow metal stands filled with sand.

a few centimeters away from the front wall.

You'll need to put thick absorption in between the speaker and the wall. 10cm of OC-703 would be ideal.

Putting the mixing position further back would be too close to the wardrobe.

It's already way too close! It's in the worst possible location, in a room with really bad low-end acoustics, which is why the mixes don not translate at all well. No matter how much treatment you put in that room, the mixes will never translate until you fix the basic layout and geometry.

With my very limited experience i'm not sure if it looks good or bad, with all those large swings in the frequency response at some room modes

The data itself is fine: it is valid, and is exactly what I would expect from such a room, so you did the calibration correctly, and ran the tests correctly. However, what the data SHOWS is pretty ugly... Major modal issues, large reflections, very long and unbalanced decay times, etc. Not pretty.

There are three curves : left speaker, right speaker, and both.

And there are rather large difference bet ween the left and right speakers, which indicates that there is already a symmetry problem in the room.

At least there is not much ringing in the waterfall besides the bass-reflex port.

Ummmmm, sorry, but I think we must be looking at different watefalls! On the one I downloaded, there is MASSIVE ringing all over the low end, and none of it is related to the A7X reflex ports. It is all modal. For example:

jlerhun-waterfall-20-500.jpg

That's the waterfall for the low end, for both speakers together. There's a huge modal peak at around 70 Hz (0,0,1 axial, plus 0,1,0 axial, plus 2,0,0 axial), another at 92 Hz, (probably tangential), another at 114 (3,0,0 axial), another around 128 (likely oblique), another at 153 (4,0,0 axial), and a whole lot more. It's a text-book classic case of a bad room with the listening position in the worst location. Huge, enormous modal ringing below 200 Hz. Here's the RT-60 plot:

jlerhun-rt-60.jpg

And the Spectrogram for the low end:

jlerhun-spectrogram.jpg

That's about as bad as it gets.

I guess that some amount of bass-trapping is in order,

Right, but you'll need a lot more than just "some"! Theoretically, you need nearly 300 sabins of absorption to get the room usable. That implies treating over 50% of ALL of the surfaces in the room. You can't treat the floor, obviously, nor the doors and windows. The remaining surface area will have to be covered nearly 100% with treatment.

I suppose that the upper front corners would be a good place to start.

You'll need to do every single corner that you can get to in that room. Ideally, all 12 of them will need major bass trapping. You probably can't do all 12, but I would go for as many as you can, and at least 6 of them.

The window and door gap will be covered with a dual layer of heavy curtains,

heavy curtains are not going to do much. Your problems are in the part of the spectrum where wavelengths are measured in several meters! So you need very thick, deep, large, porous absorbers over almost the entire surface of the room, and then they need to be covered with something reflective, to keep the highs in the room.

and either on the ceiling (acoustic foam)

You are going to need more than a bit of foam on the ceiling! I would suggest a hard-backed cloud, hung at a steep angle.

or the floor (carpet).

Forget carpet, and don't do anything to the floor. Carpet is useless as acoustic treatment in studios, which is why you practically never see it in professional studios!

The budget for all that is rather low, about 400€ (~ 550$US), but DIY is definitely an option.

5oo dollars is not going to buy you enough insulation and other materials to do this job. It will buy some, yes, and that "some" will make a difference, but in order to really get it under control, you'll need to increase your budget.

For bass trapping, it could be either a couple of those kits ... or maybe superchunks. For absorption, something like this ... (size and thickness to be determined), or some DIY panels.

Superchunks, for sure, on at least six of your corners. And DIY panels will be a lot cheaper than purchased foam.

- Is there a way to tame the frequency response of the room, especially in the low-end, or is it a desperate room mode issue ?

You can tame it to a certain extent, but it would be far better to use a different room. This one is basically just a big closet, with barely 10m2 of floor area. The minimum for a good control room is about 30m2 of floor area. So moving to a larger room would be the best option. But if you treat this room and arrange it properly, it can be a lot better than it is now. At present, it is unusable.

- how much bass trapping is required in that room, and where is the best place to put it ?

Put superchunks in at least 6 of the room corners, and more if you can figure out how to do that. Do at least the front vertical corners, the the rear vertical corners, and the top and bottom corners of the front wall. That's the minimum.

- how much mid/high absorption is required, and where is the best place to put it ?

Nothing at all! the bass traps will already be sucking out far too much of the highs, which is why they will need partially covering with reflective panels. However, you will need absorption on the first reflection points, behind the speakers, and across the entire back wall.

- how good/bad could it end up, given the small budget ? Would some EQ help in "finishing the job" ?

It is unlikely that EQ will be much use here. Your problems are in the time domain, and the phase domain, not the frequency domain. EQ can give the "final touches" to a good room that has been treated as far as possible, but in this case, you won't be able to get to the point where EQ would be useful.

- Stuart -

[jlerhun]

Hi Stuart,
Thank you very much for your detailed answer, it will help a lot !
I had no idea that the situation was that bad. Measurement is indeed the way to go, with the help of a knowledgeable guy to interpret them.

I will now translate it for Christophe. He will for sure be dissapointed, but now we know where the problem lies, and what can be done.

It's used with the largest dimension as width (for better symmetry)

... which places the engineer's head right in the middle of that awful mode.... Not a god place at all.... I would suggest rotating the room 90° so the speakers fire down the long dimension, which would allow for slightly better acoustics. And get rid of the clothes: hang them in another room: You'll never get good symmetry with those hanging all down one wall.

Would the window, centered in the (new) frontwall, be a problem ? I guess the clothes could be relocated in a symmetric way, those are two identical pieces of furniture.

Are those stands very heavy? They don't look very heavy. If not, make them heavy. Or replace them with something heavy. Many people use bricks to build up a stand, or hollow metal stands filled with sand.

The stands are quite heavy already (thick steel tube and large base), and filling them with sand is the plan, probably done already.

10cm of OC-703 would be ideal.

I'm not sure if i can get real OC-703-branded product here in France. Any rigid fiberglass or mineral wool of similar density would do ? The main brands here are Isover, Rockwool, and maybe Knauf also.

It's in the worst possible location, in a room with really bad low-end acoustics, which is why the mixes don not translate at all well. No matter how much treatment you put in that room, the mixes will never translate until you fix the basic layout and geometry.

OK, that makes sense, geometry first. Of those two conflicting requirement : avoiding center where room modes are worst, and perfect symmetry for good stereo imaging, am i right that the latter is less critical ?

At least there is not much ringing in the waterfall besides the bass-reflex port.

Ummmmm, sorry, but I think we must be looking at different watefalls! On the one I downloaded, there is MASSIVE ringing all over the low end, and none of it is related to the A7X reflex ports.

Obvisouly, i need to learn where to look in a waterfall plot. I guess i did set it up with longer time range, wider bandwidth, and much higher amplitude floor. I saw no long ridges extending in the time domain there.

You'll need to do every single corner that you can get to in that room. Ideally, all 12 of them will need major bass trapping. You probably can't do all 12, but I would go for as many as you can, and at least 6 of them.

Ok, that's indeed a large amount of bass-trapping, definitely over the initial budget. And i don't know how to hang horizontal superchucks on the ceiling/wall corner ... Drilling the ceiling to hang a heavy cloud is probably not an option, it's a rented flat. I'm not the one who decides, and there are alternatives to consider before treating extensively that bad room.

- Is there a way to tame the frequency response of the room, especially in the low-end, or is it a desperate room mode issue ?

You can tame it to a certain extent, but it would be far better to use a different room. This one is basically just a big closet, with barely 10m2 of floor area. The minimum for a good control room is about 30m2 of floor area. So moving to a larger room would be the best option. But if you treat this room and arrange it properly, it can be a lot better than it is now. At present, it is unusable.

One option would be to switch the bedroom and this room. I don't know yet if Chris is willing to do that ... first thing would be to measure the bedroom and put that into a room mode calculator.

[Soundman2020]

Would the window, centered in the (new) frontwall, be a problem ?

No, not a problem. Well, it's not ideal, and you might need to cover part of it with absorption panels, but it is much better to have the room oriented that way, than the way it is now. The minor problems from reflections off the front window are less important than the major problems with the mix position right now.

I guess the clothes could be relocated in a symmetric way,

Any chance they could be relocated to a different room?

I'm not sure if i can get real OC-703-branded product here in France. Any rigid fiberglass or mineral wool of similar density would do ? The main brands here are Isover, Rockwool, and maybe Knauf also.

Isover, Rockwool, and Knauf are all good manufacturers. They all make the type of product you need. But it isn't "rigid" fiberglass insulation that you need: it is "semi-rigid". You can easily crush it and bend it with your hands. What you need is fiberglass with a density of about 30 kg/m3, or if you prefer to use mineral wool, then it should have a density of about 50 kg/m3. If you can't get semi-rigid, then ordinary "fluffy" insulation is also fine, as long as it is in that range of densities.

Of those two conflicting requirement : avoiding center where room modes are worst, and perfect symmetry for good stereo imaging, am i right that the latter is less critical ?

Wellll..... not really! Your head MUST be on the center-line from left to right across the room, but not up and down, and not front to back. The theoretical best location is 38% of the distance from the front wall to the back wall, and not centered vertically, if possible. That's not possible sometimes, since normal sitting height is 1.2m for your ears, but many ceilings today are 2.4m high.... That's why it is recommendable to hang a "cloud" over the mix position, and make it hard-backed, and angled, so that it breaks up the floor-ceiling dimension a bit.

But left-right symmetry is very, very important.

guess i did set it up with longer time range, wider bandwidth, and much higher amplitude floor.

Yup! For small rooms, all of the problems will be below about 300 Hz, so setting the upper limit of the graph to 500 Hz is a good idea. And setting the time scale to show about half a second gives you a really good idea of what the room is doing, in terms of modes, SBIR and similar problems. With amplitude, you want to be able to see from the peaks down to about 50 or 60 dB below the peaks, since all of that is very audible. Ideally, you want to see that most of the modal ringing has died away by about 500 ms, and that the RT-60 graph is as flat as possible, close to about 300ms (less for smaller rooms). That is not achievable in the vast majority of studios, even high-end professional ones, but the idea is to get as close as you can to that goal. Here's the graphs for a large professional studio that I'm busy tuning right now:

waterfall-after.jpg

rt60-after.jpg

It's not finished yet: we are still working on flatting those curves a little more.

Your studio won't be able to get like that, of course, since this is a large commercial studio with lots of treatment, but you should be able to do a lot better than what you had in the graphs I posted for you above!

And i don't know how to hang horizontal superchucks on the ceiling/wall corner

They are not very heavy: You can make frames to hold them, then put a nail or screw every foot or so (every 30cm, aprox.) up near the ceiling to hold them in place. Then if you take them down, you just plug the holes with plaster, and re-paint.

Drilling the ceiling to hang a heavy cloud is probably not an option,

Could you set up tall stands on each side of the room, and support the cloud on those?

first thing would be to measure the bedroom and put that into a room mode calculator.

And also do a test in there with REW!


- Stuart -

[jlerhun]

Hi !
Thanks again for your answers, i'm learning a lot.

Chris envisions three options :
1- use the current room, lengthwise. It would still require a lot of treatment as suggested above.
2- use the bedroom, which happens to be smaller still.
3- do nothing, and use headphones to mix the low-end.

The dimensions of the bedroom are 2.88m x 3.13m and 2.50m height.
It seems to me that it's much better in the room modes calculator, but again i may miss the point.
At least the dimensions are not near multiple, but apparendlty quite a lot of absorbtion would still be required, about 260 sabins.

Is it worth moving the bed out to perform a REW measurement ?

[Soundman2020]

1- use the current room, lengthwise. It would still require a lot of treatment as suggested above.
2- use the bedroom, which happens to be smaller still.
3- do nothing, and use headphones to mix the low-end.

I would go with Option #1. If the only other room is even smaller than this one, then it's not a good option.

Also, I wouldn't try to do the final mixing or critical listening on headphones: You don't get the proper stereo sound field like that. With correctly set up speakers, your left ear does hear the right speaker, and your right ear does hear the left speaker, so you have much better spacial reference of how the mix really sounds. With headphones, your left ear hears ONLY the left channel, and your right ear hears ONLY the right channel, which greatly exaggerates the sound stage separation. Headphone mixing is only useful if your mixes will only ever be heard on ear-buds. Even though that's the way a lot of music is hear these days, the same music is still heard far more often on car radios, home stereos, notebook speakers, smart-TV speakers, tables, clubs, restaurants, etc., all of which have two real speakers that are both heard by both ears.

When I mix, I do check how it sounds on headphones and ear-buds, but you can't mix accurately like that. Mix on good speakers in a good acoustic environment, then check the mix on headphones, to make sure it works well there too.

Is it worth moving the bed out to perform a REW measurement ?

Do a REW test with the bed in, just to see if it might be suitable. I wouldn't bother taking the bed out, though.


- Stuart -

[jlerhun]

Hi!
Just a quick update on this one:
Chris did change the orientation of the room, and that was indeed a big improvement, thanks a lot for the advice!
He also added a fair amount of treatment, including bass-traps in corners (floor to ceiling). With the addition of the ARC-2 software, this makes the room usable for mixing.

He's starting a side-business of field-recording and home-mixing, called SWAT (Sound Workshop and Audio Tactics).