Treatment for (slightly) odd shaped mix room

[Starlight]

Source of the postYou are still going to to get the front wall SBIR with the speakers in that position.

+1

My thoughts led me to draw this version of your room, showing two options - I then noticed Jennifer suggested speakers in the corners, so sorry I did not include that.

If you move everything to the front wall to minimise SBIR then you would be sitting also relatively close given the room is almost 6m long. The 2nd option is to move the whole setup so that you are approx. 3/8ths of the length of the room from the front wall. That would put your speakers about 1.5m away from the front wall which may be far enough away not to get SBIR issues. Acoustically measuring your room will confirm which if these positions performs better.

One more thought: If your diagram really is to scale then I would suggest that when measuring your room you do once with the desk in place and once without. It looks to be a massive desk (in your diagram; I shrunk it by 25% to move the speakers closer together). It may to give your first reflection problems.

[civvie]

Source of the post Intuitively your second option seems better as it means the front and side walls can be made symmetrical. With the extra plasterboard leaf at the front wall there's a danger of the triple leaf effect reducing the sound isolation properties of the front wall though, so you may have to pay attention to that.

I agree, I'm going to explore this route primarily. I've done a little more investigation and that wall is very light. It appears to be a single layer of drywall on stud wall, no insulation at all and no drywall on the other side which is a storage area. The most basic wall you could possibly build, it's more a partition than a wall really. This is the area on the lease plan:

lease plan.jpeg (36.21 KiB) Viewed 35877 times

lease plan.jpeg (36.21 KiB) Viewed 35877 times

I'm wondering how much SBIR I'll get from such a flimsy wall? There must be a lot of bass escaping into the storage area (no one in the building cares about that thankfully) I'll do more REW tests, hopefully tomorrow.

Source of the post Certainly there will be a bass boost with them in the corner, but most speakers have the ability to roll it off.

Not the barefoots unfortunately, these are the EQ options:

Let's see what the REW tests show.

Source of the post If you move everything to the front wall to minimise SBIR then you would be sitting also relatively close given the room is almost 6m long. The 2nd option is to move the whole setup so that you are approx. 3/8ths of the length of the room from the front wall. That would put your speakers about 1.5m away from the front wall which may be far enough away not to get SBIR issues. Acoustically measuring your room will confirm which if these positions performs better.

I'd love it if speakers in the corners worked out as I'd have a lot more room. On the other hand moving more into the room would not be practically ideal, but worth testing to see what acoustic gains can be had.

The desk is big, it's actually made from a door blank. I'm going to be rebuilding the desk as it's too high, but the new one will be huge too But more symmetrical with racks on both sides, and also lower, but it will still be contributing to first reflections. Not sure how people deal with this, every studio I've been in has a fairly hefty desk between the operator and the monitors, hmmm.

Thanks all, I'll get testing asap and report back.

[endorka]

Source of the post agree, I'm going to explore this route primarily. I've done a little more investigation and that wall is very light. It appears to be a single layer of drywall on stud wall, no insulation at all and no drywall on the other side which is a storage area. The most basic wall you could possibly build, it's more a partition than a wall really.

No danger of the triple leaf effect if you add another layer to make the front of the room symmetrical then

I'm wondering how much SBIR I'll get from such a flimsy wall? There must be a lot of bass escaping into the storage area (no one in the building cares about that thankfully) I'll do more REW tests, hopefully tomorrow.

I've no idea how much SBIR it will create, but I will say this: the knee walls at the front of my room are also single layer of drywall on stud. It's the correct thing as these lead to sizeable roof eaves. How much SBIR do they create? Too much! It will be interesting to see what your measurements reveal.

You might want to check out Stuart's post about the walking mic technique on this forum for doing the REW tests; https://www.digistar.cl/Forum/viewtopic.php?f=9&t=24

It's a bit of chore to do, but by generating a really useful set of REW measurements will save time in the long run. I don't know if he's mentioned anything about the "walking speakers" test but that is useful too. You can use it to determine if a null is from the SBIR as it changes pitch as you approach the boundary.

Source of the post Certainly there will be a bass boost with them in the corner, but most speakers have the ability to roll it off.

Not the barefoots unfortunately, these are the EQ options:
Screen Shot 2020-09-03 at 18.43.15.png
Screen Shot 2020-09-03 at 18.43.23.png

Bummer! Can this be done in software? There certainly is very reasonably priced hardware that can do it.

The desk is big, it's actually made from a door blank.

Solid or hollow? If solid you'll be getting these nasty reflections back bigly for sure If hollow you might have a lovely resonating box a few feet from your ears!

I'm going to be rebuilding the desk as it's too high, but the new one will be huge too But more symmetrical with racks on both sides, and also lower, but it will still be contributing to first reflections. Not sure how people deal with this, every studio I've been in has a fairly hefty desk between the operator and the monitors, hmmm.

I've noticed that too, not every one, but many. It seems like something of an elephant in the room, given the havoc it wreaks on the midrange frequencies. I think they just put up with it. I've mitigated the problem by using a tiny desk, but the ergonomics of that are not for everyone. Careful design with acoustically transparent perforated panels and special attention to detail of angles can also reduce the deleterious sonic effects of the desk. Have a look at this design by Stuart;https://www.digistar.cl/Forum/viewtopic.php?f=18&t=743

I look forward to the REW tests!

Cheers,
Jennifer

[Wheresthedug]

Hi Civvie

I am undoubtedly the LEAST experienced or knowledgeable person here so treat anything I say with caution but here are my thoughts anyway. These ideas would need further consideration and analysis/testing by others to check if they will help or make matters worse.

1 - the room seems long enough to shift the entire mixing position slightly further back. This will obviously be subject to avoiding ending up with the listening position in a null. Start with a good listening position and work from there. As the walls splay outwards this would give you a slightly wider room for your speaker placement. The difference won’t be huge but every little helps and it will give you more room to move the speakers when doing REW tests for their best position. Others will be able to calculate the theoretical best position relative to the side walls.

2 - follow Eric’s concept about angling the front wall to gain symmetry. This will also remove the need to build out the right side wall and fill it with insulation. However, rather than framing out the front wall close to the existing wall, move the entire frame forward so it is hard up against the back of the speakers to address the SBIR issue. This way the speakers won’t need to be mover closer to the side walls to achieve your desired listening position.

3 - your super chunks look way to small to be very effective in that room. However, if you move the front wall forward you will now have a MASSIVE bass trap available to fill with insulation.

4 - you seem reticent to move the speakers back as this would force them wider. I assume this is due to you wanting to keep the infamous Equilateral Triangle? Have you read Stuart’s rant about this in the Myths section? Basically the equilateral triangle isn’t necessary and the position of those speakers relative to the boundaries will have a MUCH bigger impact that being slightly out of equilibrium on the triangle.


I am only throwing out some “out the box” thinking for consideration/discussion. I could be could be miles off but if I am we will all learn something so it was worth it

Alan

[civvie]

Source of the post How much SBIR do they create? Too much!

Interesting stuff Jennifer. I'm wondering more and more about SBIR and what will cause it. For instance, does the wall need to go from floor to ceiling? Or if I simple hold a 1meter square piece of drywall behind a speaker will that suffice? I'm thinking it varies on a spectrum like everything else in acoustics! But the reason I ask is if I'm to build a new wall close to my speakers (more on this below) maybe it doesn't need to go all the way to the ceiling to do the job of reducing SBIR.

I'm imagining at the end of all this, a small amount of corrective EQ might be necessary, this might help manage a bass boost by the boundary effect of speakers near corners. But from everything I've read, corrective EQ is the very last piece after all the physical treatment is in place.

Testing will reveal more of course. I'm quickly realising that the testing I need to do is quite a chore and will ideally require me moving the (very big, very heavy, very connected with a thousand cables) desk which essentially means packing up a lot of my studio and stopping work. I'm working all the time but am booking a few weeks off soon to do the build. So in reality I might not be able to do a comprehensive set of tests until just before the build, but that raises the problem of "what build?" as without the tests I might not know everything I need to know in order to design and order the materials. So, I'm going to try and get a friend to help and do all the testing in a day much sooner.

Source of the post your super chunks look way to small to be very effective in that room. However, if you move the front wall forward you will now have a MASSIVE bass trap available to fill with insulation.

The triangles I drew on my sketch are 60cms accross the front, which I've read is a size many go for, but of course bigger is likely better, I'll build them as big as is recommended!

But your second point has me very intrigued. If I build a new front wall, angled symmetrically with the room, could the cavity between the walls act as a giant bass trap? Is there any reading I can do on this idea? I was aware that large gaps between walls is good for soundproofing and stoping bass energy transferring to outside the room, but does it effect the bass response within the room too? How much? What if the wall doesn't go all the way to the ceiling, leaving the top (and maybe bottom) open? Leaving the top and bottom corners open and exposing a cavity full of insulation might allow more bass to be trapped? I don't know!

But I like this idea because it solves 3 things in one. I can keep my monitors roughly where they are which is ergonomically great for me, I improve the symmetry and hopefully the stereo imaging, and I trap and improve the bass response.

[Starlight]

Source of the post... if I'm to build a new wall close to my speakers (more on this below) maybe it doesn't need to go all the way to the ceiling to do the job of reducing SBIR.

If I build a new front wall, angled symmetrically with the room, could the cavity between the walls act as a giant bass trap?

If you are going to build a new front wall to create symmetry and reduce SBIR, why not just build a wall into which you can flush mount your speakers?

Source of the postThe triangles I drew on my sketch are 60cms accross the front, which I've read is a size many go for, but of course bigger is likely better, I'll build them as big as is recommended!

In a previous place I made my corner traps in the superchunk style (completely filled triangular traps), 60cm sides and an 85cm hypotenuse (front). Bigger is better simply because bigger traps lower frequencies than smaller.

Then I stacked four 120cm x 60cm x 17cm broadband traps to create a rear wall bass trap. They are each light enough that I could slide the stack of four out of the way of the door easily enough.

[Wheresthedug]

Source of the post

The triangles I drew on my sketch are 60cms accross the front, which I've read is a size many go for, but of course bigger is likely better, I'll build them as big as is recommended!

Yes, the bigger the better. All waves terminate in corners hence the reason why bass traps tend to be placed in room corners, wall to ceiling intersections and wall to floor intersections. The triangular shape of superchunks is a compromise for getting a lot of treatment without taking up too much space. But bigger is still better.

Source of the post

But your second point has me very intrigued. If I build a new front wall, angled symmetrically with the room, could the cavity between the walls act as a giant bass trap? Is there any reading I can do on this idea? I was aware that large gaps between walls is good for soundproofing and stoping bass energy transferring to outside the room, but does it effect the bass response within the room too? How much? What if the wall doesn't go all the way to the ceiling, leaving the top (and maybe bottom) open? Leaving the top and bottom corners open and exposing a cavity full of insulation might allow more bass to be trapped? I don't know!

But I like this idea because it solves 3 things in one. I can keep my monitors roughly where they are which is ergonomically great for me, I improve the symmetry and hopefully the stereo imaging, and I trap and improve the bass response.

There is lots of reading available. Reading isn’t a problem. I do plenty of reading. It’s understanding that I have a problem with

A good starting point is Rod Gervais book. The chapter on treatment cover this. For velocity devices, “the thicker the material and the deeper the air space, in general, the more control you will have at lower frequencies”.

If you look at some of the treatment designs for front and back walls in Stuart’s posts and many of the diagrams on the old John Sayers forum you will see many people use hanging insulation slabs behind walls to give maxim depth so maybe something along those lines would be worth exploring.

Another good resource is Alton Everest’s Masters Handbook of Acoustics or his cut down studio builder version.

Source of the post If you are going to build a new front wall to create symmetry and reduce SBIR, why not just build a wall into which you can flush mount your speakers?

If you can manage this it would seem to be the way to go.

[endorka]

2 - follow Eric’s concept about angling the front wall to gain symmetry. This will also remove the need to build out the right side wall and fill it with insulation. However, rather than framing out the front wall close to the existing wall, move the entire frame forward so it is hard up against the back of the speakers to address the SBIR issue. This way the speakers won’t need to be mover closer to the side walls to achieve your desired listening position.

Source of the postIf you are going to build a new front wall to create symmetry and reduce SBIR, why not just build a wall into which you can flush mount your speakers?

Superb idea! It is common to see speaker soffit baffles with gaps above and below to allow the bass trapping within them to work properly, so wonder if that would work well for this case too.

Cheers,
Jennifer

[civvie]

Source of the post If you are going to build a new front wall to create symmetry and reduce SBIR, why not just build a wall into which you can flush mount your speakers?

Very fair point. It's still an IF at this point as it wasn't in my inital plan, but I was aware that as soon as I posted here then sofits would be strongly suggested! (I know for good reason, I've read a lot of John Sayers and Soundman2020 posts over the years).

Source of the post A good starting point is Rod Gervais book.

I've got this book but I can't locate it at the minute and it's driving me mad, I have a sinking feeling I've lent it to a friend and I'll probably need to buy it again.

Source of the post people use hanging insulation slabs behind walls to give maxim depth so maybe something along those lines would be worth exploring.

Interesting idea, I've seen people using hangers. One concern is ventilation for the back of my monitors, there is a warning in the manual about overheating, so any solution will have to provide enough airspace around the back of them.

Source of the post It is common to see speaker soffit baffles with gaps above and below to allow the bass trapping within them to work properly, so wonder if that would work well for this case too.

If you have any links to this type of design I'd love to see them.

OK I just did another REW test with my new Barefoot monitors (in same position) and with my desk and front part of room cleared of junk

The bass is a little worse than the last tests which I'm putting down to the Barefoots going much lower than my previous Dynaudios. But across the board it doesn't seem wildly different, which I'm taking as a good sign that I'm measuring correctly.
I'd be interested in knowing what you all make of the differences in L and R response. I guess I was expecting these to match better which started me worrying about imaging and asymmetry, but I don't know much about analysing this type of data so I'd love to know how good/bad it all looks just based on the data alone. I also don't know how to interpret the phase data.

Here are the mdat and screenshots:

civvie Sep 4.mdat

(11.06 MiB) Downloaded 1176 times

civvie Sep 4.mdat

(11.06 MiB) Downloaded 1176 times

[endorka]

Caveat as others have stated: I am far from the most experienced on this board, especially in asymmetrical rooms. Here are some thoughts though.

I think it is useful to look at the L and R channels independently, especially when there is asymmetry. It's also useful to focus on the low end first;

First thing that springs to mind is the null on the left around 61Hz. That is a nightmare, and must really mess with the imaging and perceived level of crucial elements like the resonance of the kick drum. In the absence of photos of the measured setup it is difficult to guess at the cause, but perhaps something to do with the primary width mode of your room?

I plugged the dimensions of your room into this room mode calculator, with an average of the width, and it identifies the primary width mode as 62Hz.
https://amcoustics.com/tools/amroc?l=595&w=276&h=326&re=EBU%20listening%20room
As the speaker and / or listening position moves across the width of the room the volume of this frequency will go up and down. Perhaps the room asymmetry is causing an uneven response at the monitor and / or listening position? It would be interesting to see the result of measures with you moving the listening position left and right. Although this may be moot if you even up the symmetry of the room.

Second thing is the peak on the impulse charts of both L and R;

Ideally all the peaks should be below -20dB after the direct sound. There are really nasty peaks much louder than this at round 0.7 milliseconds in both speakers. This means the path for that particular reflection is about 24cm longer than the direct path from speaker to the measurement mic. Any idea what could be causing these? You can use the string trick described by Stuart to find them; (from post https://www.digistar.cl/Forum/viewtopic.php?f=14&t=208&p=1373&#p1373)

I have placed the cursor at the biggest issue, which is a very strong reflection at about 3.1ms after the direct sound, but you can see several other reflections around that area. The ideal is to have no reflections larger than -20 dB for the first 20 ms, and as you can see, you have several. Each vertical spike on that ETC graph is one specific reflection. That big one at 3.1 ms is nasty: just a couple of dB down from the direct sound, in fact. I mentioned this in an earlier post, and what it does to the psycho-acoustic ability of your ears and brain to correctly interpret sound.

So I would suggest that you try to identify that guy, and get rid of him! Fortunately, there's an easy way to do this, which I call the "string trick". You are going to need a long piece of string, some masking tape, and a marker pen. From looking at the REW data, it's clear that your reflection is arriving at very nearly 3.08ms after the direct sound. Since sound travels at 343m/s, that means that it must have traveled about 1100mm further than the direct sound did, to arrive at the mic 3.08ms later. So, all you need to do is to find out which surface in your room creates a path that is 1100mm longer than the direct path from the speaker to the mic! Simple. In other words, the reflection is traveling about a meter longer than the direct distance, which is why it is delayed by 3.08ms.

So, first carefully measure the distance from the acoustic center of the speaker to the tip of the mic, then get your piece of string and measure out that same distance PLUS 1100mm along the string! Leave a little bit of string free at one end so you can tape it to the speaker, make a mark on the string with your marker pen, then measure exactly the distance "speaker to mic + 1100mm", and make another mark. That distance is how far your nasty reflection is traveling. Now gently tape one end of the string to the front of your speaker as close to the acoustic axis as you can get, (gently! don't damage the speaker!!) such that the first mark on your string is right up against the front face of the speaker. Now attach the other end of the string to a mic stand (without the mic! Don't risk messing up your mic...), set up so that the second mark on your string is at the exact spot where the measurement mic was when you did the REW tests. So no you have your string danging down, with one mark on the speaker axis and the other where the mic was: all you need to do now is to take the middle of that dangling loop of string and move it around the room, to see what surface it touches. Bingo! That's your reflection point. It might be on the wall, or the ceiling, or some piece of furniture, or the desk... try many spots all over the place: you might even find that there are several spots that you can reach with the middle of the loop: put pieces of masking tape on all of those, with a visible mark at the point where the string just touches the surface. You might need help with this, so you don't accidentally pull over the mic stand, or pull the string off the speaker.

Cheers!
Jennifer

[civvie]

Source of the post First thing that springs to mind is the null on the left around 61Hz. That is a nightmare, and must really mess with the imaging and perceived level of crucial elements like the resonance of the kick drum. In the absence of photos of the measured setup it is difficult to guess at the cause, but perhaps something to do with the primary width mode of your room?

I plugged the dimensions of your room into this room mode calculator, with an average of the width, and it identifies the primary width mode as 62Hz.
https://amcoustics.com/tools/amroc?l=59 ... ing%20room
As the speaker and / or listening position moves across the width of the room the volume of this frequency will go up and down. Perhaps the room asymmetry is causing an uneven response at the monitor and / or listening position? It would be interesting to see the result of measures with you moving the listening position left and right. Although this may be moot if you even up the symmetry of the room.

Yes I thought that looked nasty too! As it's so low in frequency I probably don't notice the imaging shift as much, but I'm going to do some more tests to try and work out the source. It's interesting the correlation with the 62hz the room calculator suggested, though as you point out, that's a width mode and the width at the listening position would be less than the average width as I'm positioned more towards the front of the room. Would you suggest doing more tests with just moving the mic left and right width wise?

Source of the post Ideally all the peaks should be below -20dB after the direct sound. There are really nasty peaks much louder than this at round 0.7 milliseconds in both speakers. This means the path for that particular reflection is about 24cm longer than the direct path from speaker to the measurement mic. Any idea what could be causing these?

Oh fascinating, 24cms is almost certainly the desk. You can see it's worse in the right speaker, well the left hand side of the desk has a 19" rack on it while the right side is clear, giving more flat desk surface to bounce off. I'll do more tests to confirm this but this is great info to take into consideration for my new desk design. I wonder if desk reflections can be subdued by using absorption panels to block the first reflection sound path?

Many thanks for your analysis and input!

[endorka]

Yes I thought that looked nasty too! As it's so low in frequency I probably don't notice the imaging shift as much, but I'm going to do some more tests to try and work out the source. It's interesting the correlation with the 62hz the room calculator suggested, though as you point out, that's a width mode and the width at the listening position would be less than the average width as I'm positioned more towards the front of the room. Would you suggest doing more tests with just moving the mic left and right width wise?

Definitely - see the walking mic link test link for a method to do this rigorously. It can also be illuminating getting REW to emit a steady pitch (sine wave) from its generator at the problem frequency. There's an option on the generator to emit the pitch at the SPL graph cursor so you can do this very accurately. Emit the pitch, then wander about the room, and see what your ears tell you. Perhaps a walk from left to right will tell you if this is a room mode or not.

Oh fascinating, 24cms is almost certainly the desk. You can see it's worse in the right speaker, well the left hand side of the desk has a 19" rack on it while the right side is clear, giving more flat desk surface to bounce off. I'll do more tests to confirm this but this is great info to take into consideration for my new desk design. I wonder if desk reflections can be subdued by using absorption panels to block the first reflection sound path?

You can certainly use acoustic panels to help confirm the diagnosis. If you put a panel on the likely reflection point and it reduces the peak, then bingo. They probably help to some extent, but would they make the desk difficult to use? I wonder if angling the desk or making it from an acoustically transparent material like perforated panel would give a better combination of good acoustic performance and usuability?

Cheers,
Jennifer

[ericwisgikl]

Hey civvie, I've been thinking, while looking at the last measurements: what if you twist your desk and monitors a little, in order to end up centered between side walls, regardless front and back walls?

I don't really know how much difference could it make, but it sure will be interesting to compare with measurements from the current layout. Maybe the low end doesn't change at all, but I hope to see some less difference between L and R monitors.

Best regards,

Eric

[civvie]

Source of the post Hey civvie, I've been thinking, while looking at the last measurements: what if you twist your desk and monitors a little, in order to end up centered between side walls, regardless front and back walls?

I don't really know how much difference could it make, but it sure will be interesting to compare with measurements from the current layout. Maybe the low end doesn't change at all, but I hope to see some less difference between L and R monitors.

Hi Eric. I'm doing more testing today and plan on trying this. My issue is, how might I calculate the correct angle to create symmetry between my L + R walls? Architecture and geometry are not my strong suit! I need to find the angle of the front-right corner but not sure how to do this accurately.

[ericwisgikl]

I put your measurements into Sketchup, traced side walls lines till they meet, measured the angle, which is about 11.5º, and traced a line al 5.75º, the half of them both. Then I took measurements from right side wall to this center line, and that is your new center.

I almost match the center of both front and back walls, so if you trace a line between the front wall center and back wall center, you will get your center line anyway.

When twisting your desk, your goal is to get your head in the center between side walls, as well as the monitor, and them at equal distance from each side wall.

Cheers,

Eric