John Sayers' Recording Studio Design Forum Archive

Greetings all,

This is my first post here but I've been lurking for some time now. Goes without saying but I'm so thankful this forum exists as I've learnt so much!

I've decided to build some soffit mounts in my small bedroom for my Mackie HR824 (mk1) speakers.

I plan on eventually treating the whole bedroom (you'll notice the murphy bed at the rear wall that I built specifically for this purpose) but I figured soffits was a good place to start (I can do a REW test straight after and that will determine what I do next).

The room is 390 x 221 x 290. Here are some pics from the Sketchup model I made:
(The door by the bed is the entrance door & the door with the windows around it is the balcony door)

From my months of lurking on here & digistar.cl I've gathered the following:

- Listening position should be around 35-40% of the room length
- Speaker angle can be anywhere between 25-40° (sometimes even slightly outside that window)
- Speakers should be roughly 55% of the room width apart
- Speakers be pointed towards around 12-16 inches behind the head (i.e. pointed at the ears, not the eyes)
- The soffit baffle width should be 3-5x the width of the speaker woofer
- The speaker should be positioned around 3/5 of the way down the width of the baffle
- A proper RFZ should have 1st reflections no closer than a foot to the listening position

(Please correct me if any of these are wrong!)

With these in mind, I whipped up the following basic soffit shape (for now I only have questions regarding the design from above):
I wanted to achieve the best possible RFZ, given that the soffit wings can't extend further than the balcony door. The numbers you see above produce the following 1st reflections:
And here are the rays closest to my listening position:
As you can see, the problematic reflections are caused by the parallel walls (they're around 6" away from the listening position i.e. well within a foot).
Unfortunately the balcony door opens inwards & towards the speakers so I'm unable to extend the soffit wing so its partly on the door.
So my first question is, could they be remedied by absorbers placed right where the soffit wings end (the left one would be attached to the balcony door), like so:
I tried decreasing the angle (i.e. moving the speakers back) to offset the parallel wall reflections, but the more I do that the closer the reflections from the soffit wings become (resulting in not one but two problematic 1st reflections, one of which I can't remedy with an absorber). Here's a quick example of what that looks like:
What do you guys think? Any suggestions you guys could give me would be immensely appreciated!

snailboyawayyy wrote:Greetings all,

This is my first post here but I've been lurking for some time now. Goes without saying but I'm so thankful this forum exists as I've learnt so much!

I've decided to build some soffit mounts in my small bedroom for my Mackie HR824 (mk1) speakers.

I plan on eventually treating the whole bedroom (you'll notice the murphy bed at the rear wall that I built specifically for this purpose) but I figured soffits was a good place to start (I can do a REW test straight after and that will determine what I do next).

The room is 390 x 221 x 290. Here are some pics from the Sketchup model I made:

Room 1 5.jpg

Room 1 1.jpg

Room 1 2.jpg

Room 1 3.jpg

Room 1 4.jpg

(The door by the bed is the entrance door & the door with the windows around it is the balcony door)

From my months of lurking on here & digistar.cl I've gathered the following:

- Listening position should be around 35-40% of the room length
- Speaker angle can be anywhere between 25-40° (sometimes even slightly outside that window)
- Speakers should be roughly 55% of the room width apart
- Speakers be pointed towards around 12-16 inches behind the head (i.e. pointed at the ears, not the eyes)
- The soffit baffle width should be 3-5x the width of the speaker woofer
- The speaker should be positioned around 3/5 of the way down the width of the baffle
- A proper RFZ should have 1st reflections no closer than a foot to the listening position

(Please correct me if any of these are wrong!)

With these in mind, I whipped up the following basic soffit shape (for now I only have questions regarding the design from above):

29.4°.jpg

I wanted to achieve the best possible RFZ, given that the soffit wings can't extend further than the balcony door. The numbers you see above produce the following 1st reflections:

Amray L (29.4°).jpg

Amray R (29.4°).jpg

And here are the rays closest to my listening position:

Closest rays (29.4°) 3.jpg

As you can see, the problematic reflections are caused by the parallel walls (they're around 6" away from the listening position i.e. well within a foot).
Unfortunately the balcony door opens inwards & towards the speakers so I'm unable to extend the soffit wing so its partly on the door.
So my first question is, could they be remedied by absorbers placed right where the soffit wings end (the left one would be attached to the balcony door), like so:

Closest rays (29.4°) 3 w panels.jpg

I tried decreasing the angle (i.e. moving the speakers back) to offset the parallel wall reflections, but the more I do that the closer the reflections from the soffit wings become (resulting in not one but two problematic 1st reflections, one of which I can't remedy with an absorber). Here's a quick example of what that looks like:

27.4° Closest Rays scaled.jpg

What do you guys think? Any suggestions you guys could give me would be immensely appreciated!

Instead of using soffit wings to deflect problematic reflections why not use broadband absorption instead?

Instead of using soffit wings to deflect problematic reflections why not use broadband absorption instead?

My understanding was that 1st reflections that occur at the front wall are better off reflected around the listener towards the rear wall, where they're weaker and easier to absorb?

In my case the wings can't be longer than they are (due to the balcony door), so I'm wondering if using absorbers where the wings meet the walls is a suitable alternative? This way I get to reflect most of the 1st reflections around the listener, with the remainder of the reflections (i.e. the ones at the bedroom's parallel walls) being absorbed by the panels

Here's the pic again:

snailboyawayyy wrote:

Instead of using soffit wings to deflect problematic reflections why not use broadband absorption instead?

My understanding was that 1st reflections that occur at the front wall are better off reflected around the listener towards the rear wall, where they're weaker and easier to absorb?

In my case the wings can't be longer than they are (due to the balcony door), so I'm wondering if using absorbers where the wings meet the walls is a suitable alternative? This way I get to reflect most of the 1st reflections around the listener, with the remainder of the reflections being absorbed by the panels

Here's the pic again:

Closest rays (29.4°) 3 w panels.jpg

It depends how low down the spectrum you'd like to absorb. First order reflections that dramatically affect stereo imaging are only from 1khz and up:

"Early reflections are defined as reflections from boundary surfaces or other surfaces in the room which reach the listening area within the first 15 ms after the arrival of the direct sound. The levels of these reflections should be at least 10 dB below the level of the direct sound for all frequencies in the range 1 kHz to 8 kHz." - EBU Tech 3276 Page 5.

... a 50mm rock wool panel will be more than sufficient for obliterating that.

However, most notable designers will aim for absorption down to at least 500hz, if not lower, at first reflection points. 125mm will be sufficient for this.

The reason for using soffit wings in a fully authentic RFZ design is to deflect the energy to a diffuser on the rear wall where it can be reflected back in to the room, lower in level and diffused, creating the appropriate initial time delay gap. This is supposed to create a nice "blooming" of the sound and helps to keep the room lively in stead of dead.

However, that is just one design criteria. There is no need to follow such an (IMO) outdated design. I do not see how or why ANY reflections, no matter how low in level or smeared, are beneficial to the engineer, personally I only want to hear the true, direct sound from the speakers with absolutely no colouration from the environment.

If a room is found to be uncomfortable due to the anechoic-esque environment then it is easily remedied by the clever placement of reflective surfaces which do not send destructive reflections back to the listener, just enough to provide "self noise cues" (talking, moving, etc) for the people inside of the room. But these reflective treatments never colour the sound coming out of the loud speakers, so that back at the listening position the listener hears ONLY the direct sound as well as feeling like they are in a safe, comfortable environment.

There is a lot more to psychoacoustics but that should give you enough of an idea for now.

The reason for using soffit wings in a fully authentic RFZ design is to deflect the energy to a diffuser on the rear wall where it can be reflected back in to the room, lower in level and diffused, creating the appropriate initial time delay gap. This is supposed to create a nice "blooming" of the sound and helps to keep the room lively in stead of dead.

Ah, I see. So you think I should have the panels extend to replace the soffit wings? Like so:
The balcony door becomes more an issue w/ this approach - do you think it'll be easier/cheaper to build than my design (that is, reflecting to the rear wall where it's absorbed (not diffused))? The reflections are getting absorbed either way...

However, that is just one design criteria. There is no need to follow such an (IMO) outdated design. I do not see how or why ANY reflections, no matter how low in level or smeared, are beneficial to the engineer, personally I only want to hear the true, direct sound from the speakers with absolutely no colouration from the environment.

I completely agree!

snailboyawayyy wrote:

The reason for using soffit wings in a fully authentic RFZ design is to deflect the energy to a diffuser on the rear wall where it can be reflected back in to the room, lower in level and diffused, creating the appropriate initial time delay gap. This is supposed to create a nice "blooming" of the sound and helps to keep the room lively in stead of dead.

Ah, I see. So you think I should have the panels extend to replace the soffit wings? Like so:

Extended panels.jpg

The balcony door becomes more an issue w/ this approach - do you think it'll be easier/cheaper to build than my design (that is, reflecting to the rear wall where it's absorbed (not diffused))? The reflections are getting absorbed either way...

However, that is just one design criteria. There is no need to follow such an (IMO) outdated design. I do not see how or why ANY reflections, no matter how low in level or smeared, are beneficial to the engineer, personally I only want to hear the true, direct sound from the speakers with absolutely no colouration from the environment.

I completely agree!

I think your design with the soffit wings and the rest of the wall absorptive will work well, or if you go without soffit wings then you could have a panel that slides across the door when it's not in use... or just a gobo.

I think your design with the soffit wings and the rest of the wall absorptive will work well, or if you go without soffit wings then you could have a panel that slides across the door when it's not in use... or just a gobo.

I decided to go ahead and remove the wings. Will figure out some kind of panel-based alternative later on.

Here's the frame design so far (left soffit w/ king studs, no inner studs, baffle, hangers etc. just yet):



Pls tell me if you see something wrong with how I've arranged the plates + king studs, this is my first time designing a frame

garethmetcalf pointed me to a build he made 3 years ago using the same speakers and similar room dimensions. I've been using certain aspects of his design:
http://www.johnlsayers.com/phpBB2/viewt ... =1&t=20895

I have a few Qs for anyone who can answer them

1.) In Gareth's design (and others I've come across) there's an opening at the top. Is there a reason for this? Shouldn't the baffle just extend to the ceiling if possible? Gareth's design has a hole in the baffle so it can't be an air circulation thing. Is it because the bass trap at the top is less effective if its sealed off? If that's the case, why not fully open the bass trap at the bottom instead of having it partially covered?
Here's his:


2.) Some soffit designs I've seen use hangers at both the bottom and top of the soffit, but others (like Gareth's) fill all cavities around and above the speaker w/ insulation. Which is better?
Here's an example of the former:


3.) That 2nd platform above the speaker - does it matter where exactly it is? Could it be immediately above the speaker if I wanted (i.e. if I go with hangers at the top then they could be larger)?

4.) Should the soffits be attached to the surrounding walls? Or should the whole thing be loose?

Thanks for the help guys!

the openings top and bottom are for the absorption cavities. the small slots above (and below sometimes) is the air venting for the speaker.

depending on the size of things - the cavities could use hangers or be filled (prefer semi-rigid insulation on the from behind cloth, and pack/fill with low density insulation the rest of the cavity.

as a rule of thumb, the minimum width/height for a baffle plate is 3x the diameter of the LF driver. that said, there are methods for more exact calculation of the baffle step but you're restricted in terms of space so, sometimes you just have to make do and leverage the width your given.

another consideration - if you need more width, or the listening position needs adjustment to make things work, you don't have to put the speakers on a purely 30° (or 60° depending on your perspective) angle. it's not uncommon to see 26-34°. or 22.5°. or even 15° to create the triangle. certainly we know this happens with stand mounted and console mounted speakers all the time.

hiya gullfo! Thanks for the reply

the openings top and bottom are for the absorption cavities.

Yes, for the bass traps right? Then why not fully open the absorption cavity at the bottom as well as the top? In all the soffit designs I've seen, it's partly covered with a plywood sheet with insulation on either side. Why not have the bass hangers at the bottom completely exposed to the room?
If it's to absorb mid-high reflections from the mixing desk, won't the hangers do that anyway? And if I have a very small desk, should I just ignore the plywood sheet and open the hangers to the room?

Examples of what I'm talking about:

depending on the size of things - the cavities could use hangers or be filled (prefer semi-rigid insulation on the from behind cloth, and pack/fill with low density insulation the rest of the cavity.

Which is better? Filling with insulation is easier, but maybe the hangers are worth the extra effort?

as a rule of thumb, the minimum width/height for a baffle plate is 3x the diameter of the LF driver. that said, there are methods for more exact calculation of the baffle step but you're restricted in terms of space so, sometimes you just have to make do and leverage the width your given.

Mine is now 4x the width of the woofer. I understood that 3x was the minimum, but that I should make it as large as possible?

another consideration - if you need more width, or the listening position needs adjustment to make things work, you don't have to put the speakers on a purely 30° (or 60° depending on your perspective) angle. it's not uncommon to see 26-34°. or 22.5°. or even 15° to create the triangle

Yes, I was aware that the equilateral triangle wasn't necessary, I just happened to land on 29.4°

Sorry if I'm being dim! I really appreciate the help!

the plywood on the bottom partially covering - it was to extend the baffle plate down (most times the LF driver is on the booth, and for correct height of the acoustic center of the speaker). it's inset a bit and covered so it provides some absorption for reflections off the desk/console, and still provides the baffle-step needed.

hanger or insulation fill - depends - if you have the space go with the broadband hanger approach, if you don't, then use the insulation fill. remembering the entire cavity is lined with semi-rigid one way or another

the plywood on the bottom partially covering - it was to extend the baffle plate down (most times the LF driver is on the booth, and for correct height of the acoustic center of the speaker). it's inset a bit and covered so it provides some absorption for reflections off the desk/console, and still provides the baffle-step needed.

In this thread:
https://www.johnlsayers.com/phpBB2/view ... &start=195
Stuart says the the speaker should be offset around 1/4 of the baffle height. Does this include the inset part you're describing? Or is it just the front panel he's referring to as the 'baffle'?

Do you think I should go ahead and include the plywood sheet part? And leave a foot-high opening at the bottom? (Considering my desk is tiny & backless)
Could I do without it, or is it essential to the soffit working properly?

hanger or insulation fill - depends - if you have the space go with the broadband hanger approach, if you don't, then use the insulation fill. remembering the entire cavity is lined with semi-rigid one way or another

Thanks for the info so far (and the v prompt replies)! My room is v small so I think I'm gonna go ahead and do absorption cavities at both the top and bottom of the soffits, and use acoustic panels in both

There are other important points that should be taken into account, such as making sure that the speaker is offset from the center of the baffle, to prevent symmetrical artifacts forming around it and creating intense patterns of comb filtering. So there's a trade-off here between the need to have as much area as possible around the speaker, and also to not have the speaker in the center. A good rule of thumb is to have it offset to about 2/5 of the width and around 1/4 of the height of the baffle. But just like the infamous "38% rule", those are not absolute perfect positions, written in stone: just good starting points.

what Stuart is referring to is the slight asymmetry on the baffle plate to offset a possible reinforcement of resonances due to modal actions of the plate (just like a room, a plate has modal behavior). however, i think the context is also WRT to a very large baffle plate compared to the speaker, and without reading more, it may be the reinforcement behind the plate is not suggested to include significant bracing to stiffen the panel to the resonance frequencies are pitched high enough to be readily muted by the insulation which damps the panel.

also, consider that the speaker is decoupled from the plate as well as the internal structure so only the close proximity of the acoustical energy from the speaker will excite the modes- which will be much less available energy than a direct coupling would provide. think of the baffle plate as more of a wave guide (because it's decoupled) in this practice rather than a sound source (which direct coupling would do).

so to answer your question - i would add the lower plate and make it absorptive. leave 12" opening on the bottom. if using power speakers, make sure you include a venting path to allow convection flow - i like to use duct board for this as it makes things neat and reliable. see attached. quick note - the bottom vent could be moved up to the bottom of the exterior plate if that helps with hangers.

The designs you've shown so far are really only baffle extensions, or partial flush mount, whatever you want to call it. A sort of hybrid of baffle extension and bass trap.

These designs may work well in a small room since you can maximise bass trapping by using the space behind the baffle extension for absorption.

However, I think it's relevant to point out that traditionally the front wall is continuous from side to side, floor to ceiling. It connects to all other walls, ceiling and floor. It is sealed, massive and completely reflective. This means the entire front wall (and arguably, all of the solid boundaries of the room) becomes the baffle extension. This maximises the baffle step response frequency, and the speakers are literally no longer inside the room at all. Therefore, no bass trapping is used in the space around the speakers, since as far as the room is concerned, there isn't any space around the speakers... the speakers become the front wall, or the front wall becomes the speakers...and the speakers are no longer projecting in to full space (behind the cabinets) and all sound is projecting into half space. This approach means you would need proper trapping on all other walls, especially the rear wall since it will take full responsibility for taming the length mode(s).

If it were me in your situation (and much to the disproval of Stuart) I would just do a "soft flush" and cover the treatment in slats. This is where you enclose the speakers fully in treatment, cover the frame in fabric and then apply slats over the entire face of the treatment. This won't provide a baffle extension, but will maximise your trapping in the room as well as provide some nice reflective surfaces for self noise cues. It's also a lot easier to build, very straight forward.

Regarding your framing, where you have two timbers joining at an angle, the angle should be the same for each. In your sketchup look at the top plates, the top plate for the centre section is cut straight resulting in the adjoining members to be cut at a steeper angle and therefore they become "wider" than the central timber. Do you see what I mean?

also, consider that the speaker is decoupled from the plate as well as the internal structure so only the close proximity of the acoustical energy from the speaker will excite the modes- which will be much less available energy than a direct coupling would provide. think of the baffle plate as more of a wave guide (because it's decoupled) in this practice rather than a sound source (which direct coupling would do).

So if I was following John's design (tight fit in the housing box, v strongly attached to the surrounding frame) then the plate's modes would be more of an issue, as its a direct coupling?
As of yet I'm unsure if I'm gonna go down this road, or try the sorbothane route (i.e. decouple the speaker from its housing box), but you're saying if I decouple the speaker from the plate & frame, then symmetry on the plate should be less of an issue?

Thanks for the pics! I'm sure they'll prove v useful down the line

However, I think it's relevant to point out that traditionally the front wall is continuous from side to side, floor to ceiling. It connects to all other walls, ceiling and floor. It is sealed, massive and completely reflective. This means the entire front wall (and arguably, all of the solid boundaries of the room) becomes the baffle extension. This maximises the baffle step response frequency, and the speakers are literally no longer inside the room at all. Therefore, no bass trapping is used in the space around the speakers, since as far as the room is concerned, there isn't any space around the speakers... the speakers become the front wall, or the front wall becomes the speakers...and the speakers are no longer projecting in to full space (behind the cabinets) and all sound is projecting into half space

Thanks for the input Paulus! What you're saying does make sense, but if I completely seal the baffle so its closed at the bottom, ceiling and sides, won't I need to bass trap the 'new corners' of the room (i.e. the corners of the soffit), effectively using up way more space?

This approach means you would need proper trapping on all other walls, especially the rear wall since it will take full responsibility for taming the length mode(s).

So as you can see from my first post, I'm rather limited in available space for treatment at the rear wall corners, due to the murphy bed and the small, wall cupboard at the top, left corner of the rear wall. Obv I'll treat them as much as I can, but the only corners that I can bass trap to my hearts content are the ones on the front wall

Regarding your framing, where you have two timbers joining at an angle, the angle should be the same for each. In your sketchup look at the top plates, the top plate for the centre section is cut straight resulting in the adjoining members to be cut at a steeper angle and therefore they become "wider" than the central timber. Do you see what I mean?

Ah, right. Would this be better?

snailboyawayyy wrote:

also, consider that the speaker is decoupled from the plate as well as the internal structure so only the close proximity of the acoustical energy from the speaker will excite the modes- which will be much less available energy than a direct coupling would provide. think of the baffle plate as more of a wave guide (because it's decoupled) in this practice rather than a sound source (which direct coupling would do).

So if I was following John's design (tight fit in the housing box, v strongly attached to the surrounding frame) then the plate's modes would be more of an issue, as its a direct coupling?
As of yet I'm unsure if I'm gonna go down this road, or try the sorbothane route (i.e. decouple the speaker from its housing box), but you're saying if I decouple the speaker from the plate & frame, then symmetry on the plate should be less of an issue?

Thanks for the pics! I'm sure they'll prove v useful down the line

However, I think it's relevant to point out that traditionally the front wall is continuous from side to side, floor to ceiling. It connects to all other walls, ceiling and floor. It is sealed, massive and completely reflective. This means the entire front wall (and arguably, all of the solid boundaries of the room) becomes the baffle extension. This maximises the baffle step response frequency, and the speakers are literally no longer inside the room at all. Therefore, no bass trapping is used in the space around the speakers, since as far as the room is concerned, there isn't any space around the speakers... the speakers become the front wall, or the front wall becomes the speakers...and the speakers are no longer projecting in to full space (behind the cabinets) and all sound is projecting into half space

Thanks for the input Paulus! What you're saying does make sense, but if I completely seal the baffle so its closed at the bottom, ceiling and sides, won't I need to bass trap the 'new corners' of the room (i.e. the corners of the soffit), effectively using up way more space?

This approach means you would need proper trapping on all other walls, especially the rear wall since it will take full responsibility for taming the length mode(s).

So as you can see from my first post, I'm rather limited in available space for treatment at the rear wall corners, due to the murphy bed and the small, wall cupboard at the top, left corner of the rear wall. Obv I'll treat them as much as I can, but the only corners that I can bass trap to my hearts content are the ones on the front wall

Regarding your framing, where you have two timbers joining at an angle, the angle should be the same for each. In your sketchup look at the top plates, the top plate for the centre section is cut straight resulting in the adjoining members to be cut at a steeper angle and therefore they become "wider" than the central timber. Do you see what I mean?

Ah, right. Would this be better?

new ceiling plate.jpg

As I said, you do not need to treat the new corners formed by the new front wall, you just need to treat all the other surfaces except the floor.

As I said, this continuous front wall is not the best approach in your case, I am just pointing it out for the sake of completeness, so that people reading this can understand the principles involved of an ideal infinite baffle design.

I wouldn't worry too much about decoupling, it's not going to make a massive audible difference in this design in my opinion unless you have rather flimsy speaker cabinets that vibrate.

What I mean regarding your framing is this...