John Sayers' Recording Studio Design Forum Archive

Hey ! I've only just seen this thread having started my UK South based practice room / studio 4 years ago, then getting buried in that and numerous other things and then only just returning here last week to finish off a much overdue build thread.

http://www.johnlsayers.com/phpBB2/viewt ... =2&t=17952

Not only also being UK South Based (Glos) but we also have other similarities i.e. trying to build a UK compliant structure which can also become part of the house again should you decide to move on. But you look to have then taken things a lot further than myself - I just needed sound isolation and less focus on sound control but its good to see you going a lot more in depth than I did.

Great to see the progress, all the very best. If there is anything I can help with just holler - on the basic / similar stuff that is as there are many on here far far more savvy on the techie stuff than I am - and good to see Stuart / Soundman is on the case for you. The devil is in the detail !

Cheers

Roger

Thanks Roger, great to read through your thread, I learnt a lot from your build.

Mulling over speaker positions…

One problem with the dimensions you have given me Stuart is that, assuming I make the main baffle as wide as possible (that is 120cm that the wood sheets come in) this means that the acoustic axis is very close to the centre of the baffle (on the horizontal plane). I don’t see there is a way round this because given the width of the speaker there is very little room on either side to play around with.

So then I was thinking that it could be better with the speakers turned round to their ‘normal’ orientation (with the woofer on the inside). This is better in the respect that the speaker angle can be bigger (35 degrees rather than 28) and so the acoustic axis of the speakers are a bit further apart (even thought the speaker boxes themsleves are in virtually the same position just reversed). But this then means that the acoustic axis is at almost exactly 25% the width of the room which is not good.

So then I went back to your dimensions as being best and just having to live with the acoustic axis being a bit too close to the centre of the baffle. Because the baffles will be heavy (18mm mdf + 18mm osb + 9mm ply) I am thinking that it won’t be too bad. Maybe most of the acoustic energy will be coming from the woofers anyway which are quite a bit more away from the centre of the baffle, I am imagining that they will be much more likely to set up resonances in the baffle than the mid range driver or tweeter anyway?
On the vertical plane at least it is easy to put the speakers well off centre but how much space should I leave top and bottom, above and below the baffle front? I am thinking maybe around 60cm in between the bottom of the baffle and floor and the the same at between the top and ceiling. See diagram below

I was imagining the ‘wings’ as large baffle like construction with the same big gap at top and bottom as on the actual soffits (so the sound waves can access the voluminous absorption behind) but then I saw some pics with slatted fronts. Would that be a better thing to do?

Anyway here is my best plan for the angles of the wings and baffle geometry. Sorry I can’t manage sketchup. I need to start buildng soon so if anything jumps out as being bad please let me know. Many thanks.

Cheers
David

ps - been working on doors and windows, a few photos coming soon I hope but it is too dark in there to take photos, the electrician is supposed to come and connect up the lights but he is very very tardy.

One problem with the dimensions you have given me Stuart is that, assuming I make the main baffle as wide as possible (that is 120cm that the wood sheets come in)

On the other hand, if you turn them sideways, then they are 240cm wide... There's also nothing wrong with putting two sheets next to each other! There are several good methods for joining sheets of wood, practically invisibly.

On the vertical plane at least it is easy to put the speakers well off centre but how much space should I leave top and bottom, above and below the baffle front?

Usually the bottom edge of the baffle itself ends up around the level of your desk surface, or maybe a bit lower.

I was imagining the ‘wings’ as large baffle like construction with the same big gap at top and bottom as on the actual soffits

Right.

but then I saw some pics with slatted fronts. Would that be a better thing to do?

It depends on what the room needs. I normally don't like doing tuned treatment at the front of the room, as I don't want any "coloration" of the direct sound. But for example in Studio 3 I did that. The wings are actually slot walls, tuned to some of the problematic frequencies. In fact, there are "slot door/walls", because there actually doors in those wings, that lead to storage space behind on one side, and a small machine room on the other side. Those "doors" are very thick, as they incorporate slot wall modules.

Sorry I can’t manage sketchup.

There are many very good, simple tutorials on YouTube about how to use SketchUp. At first it looks daunting, but it really isn't that hard, once you get the basic concepts. One key is to use "make component" each time you create geometry that belongs together as an object (such as the six faces of a stud), to prevent it getting attached to other stray bits of geometry. If you do that every time, then it becomes simple.

- Stuart -

Many thanks for your reply Stuart.

Aesthetic and finacial considerations aside, practically there is only about 10cm on the left hand side of the (right) baffle spare (until the midpoint of the room) and 10cm doesn’t really make too much of a difference (putting the acoustic axis just 7.6% off centre). I suppose on the right hand side the speaker baffle could continue behind the wings but that would result in the woofer actually becoming more central and not less. I am guessing that the reason to locate the speaker off centre in the baffle is too minimize resonances within the baffle and I am guessing that because the baffle is big and heavy these resonances are going to be at a lowish frequency. If that is so then they are going to be driven primarily by the woofer and not the tweeter/midrange. Normally the acoustic axis of a speaker would go through the woofer I suppose but with KH310’s mounted horizontally the the acoustic axis in quite a way from the woofer - maybe about 20cm. So if I extended the baffle on the right by 40cm (behind the wing to the side wall) that would make the woofer exactly central.
If the woofer is the important factor even with the acoustic axis of the speaker exactly in the middle of the baffle (horizontally) the woofer is already 17% off centre!
Anyway you can see I don’t have the foggiest which is the best solution. What kind of experiments have been done which show that centralising the speaker in the baffle actually does give bad results in the real world? Or maybe is it just theoretical reasoning?

A few photos as an update ...
got some electricity in so I am not working in a dark cave anymore

I have finally managed to get around to doing a better baseline rew test for my control room, hope I got it right this time.
I followed the instructions to the letter but still had a few issues with levels so I am not quite sure it has all worked properly. Also I don’t really know what to expect as the control room is completely without treatment so it sounds pretty horrible. One thing that has surprised me is how closely the graph bumps correspond to the previously calculated modes. It amazes me when something theoretical then corrresponds to real life.

The mdat file can be found here…
https://www.dropbox.com/s/76ruwqtvec8w6 ... .mdat?dl=0

and below are a few screenshots for easy reference.

A few details on my test …
Speaker settings - not exactly sure what to set the controls on the Neumann KH310’s to, so I tried -5db bass and left lowmid and treble on 0.
Measurement mic was C414 in omni pointed up at 60 deg at the listening position (I marked the floor so I can put mic in the same place next time).
I did calibrate the soundcard but not quite sure I managed to use the calibration properly, it was basically flat anyway (-0.5db at 20hz). Interface is motu ultralight mk3
The speakers were in their final positions in boxes but without the front baffles attached.

Any advice on the results gratefully received.

I have finally managed to get around to doing a better baseline rew test for my control room, hope I got it right this time. I followed the instructions to the letter but still had a few issues with levels so I am not quite sure it has all worked properly.

The data looks fine.

Also I don’t really know what to expect as the control room is completely without treatment so it sounds pretty horrible.

And it looks equally horrible, in the graphs! Which is good, as your room's problems are clearly visible. At this stage, it is supposed to look horrible.

One thing that has surprised me is how closely the graph bumps correspond to the previously calculated modes. It amazes me when something theoretical then corrresponds to real life.

Yep! It's always nice when theory matches reality. That also gives you good confidence that the theoretical treatment will fix the real problems...

Speaker settings - not exactly sure what to set the controls on the Neumann KH310’s to, so I tried -5db bass and left lowmid and treble on 0.

You seem to be rolling off the bass a little too much, so I'd try a slightly lower setting. The others are fine, for now. We won't know too much about how the high end needs to go until you get a lot of the treatment into the room. It's mainly low treatment that you want right now, but that will have an effect on the highs, so keep your speaker tweaks in reserve until it's clear if you might need them.

Measurement mic was C414 in omni pointed up at 60 deg at the listening position

The 414 is fine for now, but it would be better to get a proper acoustic measurement mic for future tests. The 414 has a rather large body, and it can affect the measurements themselves.

(I marked the floor so I can put mic in the same place next time).

That's the X and Y location in the room, but you also need the Z location: height. All three need to be accurate.

I did calibrate the soundcard but not quite sure I managed to use the calibration properly,

It looks fine.

it was basically flat anyway (-0.5db at 20hz).

At this point in the process, the calibration is pretty much irrelevant, but as you get the room tuned closer and closer to your final goal, it becomes more important.

Any advice on the results gratefully received.

The procedure from here on is actually simple, in principle (but not so much, in practice!): Take a look at the graphs and see what the biggest problems are. Fix those, take another set of readings. Repeat. And carry on repeating until you run out of money, or time, or patience, or wall space for more treatment.

Clearly, your biggest issues right now are modal issues at 43 Hz, 83 and 123 Hz, so deal with those first: Those are very long waves, so you need very large, deep bass traps. Figure out which room axis each is associated with, and put suitable treatment on the correct walls / corners. The treatment you apply for those will also have a nice effect on some of the other low-end issues, but you'll need to take care that it does not adversely effect the high end. You don't want a dead room! You want a balanced room.

One other thing: The good news is that your left and right channels are fairly similar: so you should have a good stereo image. However, your left speaker is slightly louder than your right speaker: you need to turn down the left by about 1/2 a decibel. That's not easy to do at this point, with the room the way it is, so don't try yet. Wait until you have some of the treatment in place.

- Stuart -

Thanks so much for your reply Stuart, it is good to know that I haven’t completely messed up the test.

Soundman2020 wrote: Clearly, your biggest issues right now are modal issues at 43 Hz, 83 and 123 Hz, so deal with those first: Those are very long waves, so you need very large, deep bass traps. Figure out which room axis each is associated with, and put suitable treatment on the correct walls / corners. The treatment you apply for those will also have a nice effect on some of the other low-end issues, but you'll need to take care that it does not adversely effect the high end. You don't want a dead room! You want a balanced room.

I am sure I need some serious absorbtion on the rear wall and corners. I am wondering what is the best thing to do here. Some options I am considering are are …

- —Rockwool - Just fill as much space as possible with this.
I bought rather too much rwa45 rockwool and now have more left than I can probably use. I still have a few rolls of the fluffy loft insulation left so I could use that as well, but I was thinking I might use that behind the soffit wings and maybe underneath the speakers in the soffit bottoms, but I could also use some as part of the rear wall treatment if it would be better than the denser rockwool. I am confused becasue there is so much conflicting information about what depths and densities should be used.

—— Or I could try and build some hangers for the rear wall. I am planning a few hangers as part of the soffits but don’t quite know what material to use as the core. I can’t seem to find Hermosote over here and cork of that size is very pricey. I was considering heay duty cardboard but at this point in my build I have a fair amount of left over materials that I am keen to use - plenty of 6mm ply, 12mm ply, 11mm osb, 18mm osb, hardboard etc so I am wondering if I can use any of these for the hanger cores? Maybe the 6mm ply would work ok?

—— tuned traps . I have been discounting the idea of tuned traps as I have no faith that I could make them succesfully. I could try in an emergency situation but not too keen unless everything else fails.

— I have discounted diffusion panels as probably my room is too small for those to help

— heavy duty waveguides in the rear corners. Interesting idea but can’t find out much info on them.

I have too much choice for my limited knowledge. I can see some options but lack the ability to choose between them!

Any help gratefully received

Here is a small sketch of the dimensions of the back wall …

I am sure I need some serious absorbtion on the rear wall and corners. I am wondering what is the best thing to do here.

I would suggest a combination of pure absorption right up against the wall, then acoustic hangers in front of that. I've had decent success with that. It's one of John's very effective concepts.

—— Or I could try and build some hangers for the rear wall.

I am planning a few hangers as part of the soffits but don’t quite know what material to use as the core.

John recommends Homasote. It's a sft fiber-board product, that is commonly used to make office notice boards for pinning up notes with push-pins ("thumb tacks").

I was considering heay duty cardboard but at this point in my build I have a fair amount of left over materials that I am keen to use - plenty of 6mm ply, 12mm ply, 11mm osb, 18mm osb, hardboard etc so I am wondering if I can use any of these for the hanger cores?

I have heard of some people using that type of core, but it's a bit too dense for my liking. Lighter, less dense stuff is better.

I have been discounting the idea of tuned traps

Smart man! They are a lot harder to tune than the text books tell you about. If you work carefully, and make acoustic measurements along the way then they can work, yes, but I'm not a big fan for general home studios.

— I have discounted diffusion panels as probably my room is too small for those to help

What are the final interior dimensions of the inner-leaf? I'm sure you have that some place in your thread, but I'm lazy to go looking for it, and it would be good to get a final update. And some photos, too!

— heavy duty waveguides in the rear corners. Interesting idea but can’t find out much info on them.

Actually, it turns out that hangers act like wave-guides. That's part of the reason why they work so well. And they don't need to be tuned (although you could if you really wanted to). Just following John's basic advice on how to build them and place them, and they'll work fine.

Here is a small sketch of the dimensions of the back wall …

OK, but how about the rest of the room? All of the dimensions are important...


- Stuart -

Thanks for your reply Stuart, very helpful as always.

Soundman2020 wrote: What are the final interior dimensions of the inner-leaf? I'm sure you have that some place in your thread, but I'm lazy to go looking for it, and it would be good to get a final update. And some photos, too!

L - 410cm
W - 320cm
H - 280 cm
(rough sketch and photos below).

Soundman2020 wrote: John recommends Homasote. It's a sft fiber-board product, that is commonly used to make office notice boards for pinning up notes with push-pins ("thumb tacks").

Can’t find it anywhere in the uk. Cork is far too expensive.

Soundman2020 wrote: I have heard of some people using that type of core, but it's a bit too dense for my liking. Lighter, less dense stuff is better.

I think I should probably use the cardboard then, which seems the only thing which is light enough.

Soundman2020 wrote: Actually, it turns out that hangers act like wave-guides. That's part of the reason why they work so well. And they don't need to be tuned (although you could if you really wanted to).

I thought the idea of waveguides was that they were big and dense and so would reflect sound to the required place. I can’t see cardboard with a bit of rockwool reflecting many of the troublesome low frequencies. As long as they have a beneficial effect on the low frequencies then that is what counts I guess, all the theoretical stuff is well above my pay grade anyway.

Soundman2020 wrote: Just following John's basic advice on how to build them and place them, and they'll work fine.

I am wondering whether it would be better to use the denser rwa45 or fluffy rockwool for the hangers? I noticed one description that said to use 2.5 cm of rockwool on each side glued on but I can’t see gluing the fluffy insulation to cardboard is going to work well. And the rwa45 I have is in 10cm batts - I can cut it but probably not neatly to 2.5cm. Would 4/5 cm on each side be ok-ish? Glued but with some gaffa tape reinforcement or maybe string?

Having had a quick look at the space available I think I can get in about 3 or maybe 4 full length hangers on each side in the rear corners, about 240 x 45 cm, with at least 10cm of rockwool against the rear wall and ceiling and floor behind the hangers. In addition maybe I can fit some smaller hangers against the top of the rear wall - I think there is space for 5 or 6 at around 60 x 35 cm. Do you think that is looking in the right ballpark?

cheers
David

I've been moving the rock wool in, lots of it.
Here are a few update photos ...
More REW tests coming soon.
cheers
David

So finally I have got round to doing some more REW testing in the control room now that (hopefully) all the absorption is in. I will post all the results below, I am not really sure how to evaluate them but they don’t look too bad - frequecy response is flat within +-6dB without any smoothing (in the low frequencies) and there is not that much low boomy decay going on. I guess I can fine tune a bit but the most important thing right now is to get the speaker settings correct before I finally screw all of the soffitts and wings together. I put a single layer of soffit front on just to run the tests (but not the wings yet).

These tests were done with my KH310a’s set with just a bass cut of 2.5dB but I am wondering if I should set the low mid cut to -1.5db as well to tackle the boost in the graph between 150 and 325hz. I can’t seem to find any mention of at what frequecies the controls act at and it is a very large hassle to take down the soffits and deconst the speaker boxes everytime I want to change a setting so it is pretty important to make the final decisions right now.

The other setting I have no idea about is the input/output level settings on the speakers.
At the moment I set the output level switch to 100dB spl and the input gain control to 0 and have no idea if this is suitable or not. It sounds good to me but as I haven’t got a desk in yet it is problematic to know if it will work suitably when I do. I can’t work out how to send a reference level from my laptop/motu interface either. The REW levels seem all messed up to me, it keeps telling me that my level is far too low but the results seem to look ok, to me at least. I have to turn the laptop output level down to around half in the systems preferences in order to get the sound at 80dB (C) but have no idea what actual level that is. I am not working in broadcast or film so I guess it is not really critical as long as I get plenty of volume from the speakers and don’t clip when sending full level.

There are 3 tests I did ...

1 - without the soffit front
mdat 1
https://www.dropbox.com/s/11grca7u8zq0m ... .mdat?dl=0

2 - with the first layer of soffit front
mdat 2
https://www.dropbox.com/s/3b9vssgqr0fh3 ... .mdat?dl=0

3 - and with an additional couple of absorption panels on the floor at reflection point between speaker and mic.
mdat 3
https://www.dropbox.com/s/f4nx895xy85ut ... .mdat?dl=0


There are few graphs below from mdat 3 just to give a quick idea

In other news I finally got the live room up and running and had my gand piano moved in yesterday. Still a lot of organising and soldering to do. Photos should come in the next week.

Any help gratefully received.

cheers
David

Or alternative frequency response view ...

The room seems a little dry but maybe there will be more reflections when I have wings and some other hard surfaces
cheers
david

a couple of photos of the live room. great to finally have some instruments and musicians in there
cheers
david