Questions surrounding speaker placement

Plans and things, layout, style, where do I put my near-fields etc.

Moderators: Aaronw, kendale, John Sayers

glymur
Posts: 5
Joined: Wed Jan 04, 2017 11:46 am
Location: Lansing, MI

Questions surrounding speaker placement

Post by glymur »

Hello,

Thank you for this forum, this is a really great resource.

I recently purchased a new house near Lansing, MI USA and have the space to build my dream studio. I'm am still working on gathering data/layouts and Sketchup files to begin my build thread on this forum, but I have some questions that have been bugging me for the past week or so and will help me with those plans. They revolve around monitor placement.

Here are the measurements of my room as it currently configured:

Length: 20’
Width: 13’3”
Height: 7’3 3/4”

I have read through ITU-1116.3 and still have some questions on monitor placement. Let me recap a bit with what I know and then I'll detail my exact questions below (as that seems to align with the forum rules).

According to everything I have read (including many many threads on this forum) I believe the following is true:

- Speaker height is at "seated listener's ear height" on an azimuthal plane from the acoustical centre of each speaker.
- Speakers should be at least 1m (should the room allow it) from the front and sides of the room (again, according to the acoustical center).
- The "base width" (B) between the speakers (in a stereophonic setup) should be 2-3m apart.
- The reference listening position, according to ITU-1116.3, should be approximately 1.7 to 2 * B(base width) meters from the speakers. So for example, if our base width is 2m, our reference listening position is between 3.4m and 4m.
- The reference listening point is defined by the listening angle of 60º of the triangle created with the speakers.
- The angle of sound from the speakers that create the listening angle should be just past the listening position (since humans have two ears), or not to exceed 0.7m around the reference listening point.

Where I get lost is with the generally accepted “ideal” listening position of 38% and these "truths."

What I seem to be missing is a formula for determining a variance in these numbers dependent upon the room measurements.

Taking my room measurements above, there is a conflict between 38% (approximately 7.6ft) and 1.7*B (approximately 11.15ft). Those numbers are well outside the recommended listening area of 0.7m radius around the reference listening position.

Thus my questions are as follows:

- How do I adjust my listening position for my room measurements?
- What calculations am I missing here that should be obvious?
- Is there a formula I should be following to determine listening position and monitor placement?
- Am I an idiot and completely missing this in ITU-1116.3? :P

I appreciate any input here as the dimensions of my room will change as I start building. I have another room that I intend to build out into a control room that I will need to figure eventually figure out placement as well. Right now I want to get some early measurements just to learn a bit more about REW and understand where I am starting from.

Thank you for your time,

- Curt Micol
Soundman2020
Site Admin
Posts: 11938
Joined: Thu Aug 21, 2008 10:17 am
Location: Santiago, Chile
Contact:

Re: Questions surrounding speaker placement

Post by Soundman2020 »

Hi Curt. Please read the forum rules for posting (click here). You seem to be missing a couple of things! :)

That said, are you sure your room complies with the other requirements for a critical listening room? There are several equations in BS.1116 regarding the relationship between length, width and height. With a low ceiling and long room, are you sure that yours meets those? Regardless of whether or not they do, it's still gong to be problematic with a low ceiling.

But getting to your actual questions:

The ITU guidelines and other rule-of-thumb figures that you mention are just that: guidelines and rules of thumb. In a perfect room with a perfect ratio, perfect volume, perfectly treated, and with perfect speakers, then that would be the perfect setup. But since there is no such thing, those are just starting points.

For example, Wes Lachot's famous "38% rule" is not a rule at all! He never proposed it as such, and simply commented that it was the theoretical best location for the listening position, in relation to room modes... assuming that your room is perfectly rectangular, with the best possible ratio, blah, blah, blah. But in real rooms, I have seen successful setups with the mix position anywhere between around 28% and 45%. Because modal response is not the only factor in choosing the mix position! It is one of many. Wes later regretted ever having published that figure, since it still haunts him... people try desperately to nudge their chair fractions of an inch this way or that to get the "Perfect" 38%.... without ever considering that just leaning forward in their chair is going to reduce that by several percentage points, or sliding the chair back a bit to stretch your legs is going to increase it by several percentage points. Obviously, nobody mixes with their head held rigidly in a neck brace, never moving while they mix: engineers are all over the place! Yet they seem to turn out pretty good mixes anyway. So take the 38% "rule" with a grain of salt. A very large grain of salt. It's a helpful starting point for determining mix position, nothing more. It only take into account modal response, and in a well treated, reasonably large room, modal response is not even an issue. SBIR is often a much bigger issue. Plus, if the room is not rectangular, then the 38% rule is out the window anyway!

Ditto for the other "rules". Take then as good starting points, not "written in stone by the hand of God Himself".

Having said that, there are a few "rules" that are really important. One of those is that the room muse be symmetrical, at least for the section between your head and the front wall. Another is that the room ratio should not be one of the bad ones (direct mathematical relationship between dimensions). Another is that the mix position should not be at the "terrible" spots of 25% or 50% of room depth. Another is that the distance between the rear wall and your ears should be as large as possible, in order that the reflections coming from behind fall outside the Haas time. Another: The speakers must all be the same distance from your ears, or if not then some must have their signal delayed so that all signals arrive at your ears at the same time, in phase. Etc.
.- Speakers should be at least 1m (should the room allow it) from the front and sides of the room
That depends on the frequency range that the speaker is producing. If you have a bass-managed system where the mains are not putting out anything below about 120 Hz., then you could get closer, as the SBIR dip would not happen: the sub would be filling that in, not the mains. But if your mains are full-range, going all the way down to say 30 Hz, then I'd increase that minimum distance to 3m. The first SBIR notch for a speaker-wall distance of 3m is 29 Hz. So that "1m" number is a good rule of thumb for typical small speakers with little bottom end, or speakers in a bass managed system. But like all such "rules", it is only a starting point, and other factors come into play as well.
- The "base width" (B) between the speakers (in a stereophonic setup) should be 2-3m apart.
Assuming the room is big enough! If you wanted your speakers 3m apart, and also at least 2m from the walls, that implies the room has to be at least 7m wide.... In order to meet the other conditions, it would have to be about 11m long, and nearly 3m high. Most rooms are not that big, so this "rule" can also be modified to fit reality.
- The reference listening position, according to ITU-1116.3, should be approximately 1.7 to 2 * B(base width) meters from the speakers. So for example, if our base width is 2m, our reference listening position is between 3.4m and 4m.
Yup. And 4m being 38% implies that the room is nearly 11m long... Once again: guideline. Modify as needed.
- The reference listening point is defined by the listening angle of 60º of the triangle created with the speakers.
Once again, 60° intercept is a good staring point, but not written in stone. There's no reason why it should be! Our ears are capable of picking up sound from any point in a complete sphere, and our brains are pretty good at using the acoustic clues in the room to figure it all out. When you have a chat with your friends, can you hear them just fine when they are right in front of you? OR do you have to turn your head exactly 30° until you hear them clearly? :) If somebody plays the guitar straight out to your right, 90° off, can you still hear it? :) Once again, 60° is a good starting point, and sort of an "average" of the best angles that our ears are good at, but your studio wont be disgusting if you happen to have an angle of 60.5°.... nor 70°, nor 50°. There's "optimum", and there's "real world". For example, when I'm designing a room, I might change the speaker angle based on the speakers themselves, if they happen to be very large, very small, or have a rather wide or narrow dispersion angle, etc. Or if the room is narrow, or wide, I might change the angle too. Or if there is a large console or deep desk, then I'd raise the speakers a bit, to help avoid issues with first reflections from the console surface, "shadowing" by the console, and uneven response in the mid range. Etc. There are many factors to consider.
- The angle of sound from the speakers that create the listening angle should be just past the listening position (since humans have two ears),
Yes, absolutely. This is one of my big pet peeves: The "standard" diagram you see for setting up speakers shows them pointing at the center of your head, with the axis passing through your eyeballs! Most people don't listen with their eyeballs... Most people keep their ears in the original location on their heads, a couple of inches off to the sides and further back from their eyeballs. And since speaker manufacturers take massive special care to get the on-axis response of their speakers as clean and pristine and flat and smooth as they possibly can, it makes sense to have that axis aimed at your ears, not your eyes. So the diagram you see all over the place is flat wrong. The axes should not meet in the middle of your head, but rather a distance behind your head, such that the axes point at your ear hole. In fact, there are other reasons why it makes sense to go even a bit further here, and have the axes pass by just beyond the tip of your ear lobes. For most rooms and most layouts, the point where the axes meet will be about 12" to 18" behind the center of your head, give or take a few.
What I seem to be missing is a formula for determining a variance in these numbers dependent upon the room measurements.
Me too! I'm also missing that formula! Because unfortunately, there is no such formula. There are rules of thumb, and there is experience. I have a method that I use that gives pretty good results for most rooms and most situations, but just like any other rule, it's only a good starting point. I might well modify the speaker and mix positions, angles, distances, etc. to take into account other factors.
- How do I adjust my listening position for my room measurements?
For that room, here's how I would do it:
Speaker-Layout-Template-Generic-- glymur--S01-V1.png
Speaker-Layout-Template-Generic-- glymur--S01-V2.png
Speaker-Layout-Template-Generic-- glymur--S01-V3.png
Speaker-Layout-Template-Generic-- glymur--S01-V4.png
Speaker-Layout-Template-Generic-- glymur--S01-V5.png
Speaker-Layout-Template-Generic-- glymur--S01-V6.png
Speaker-Layout-Template-Generic-- glymur--S01-V7.png
- Am I an idiot and completely missing this in ITU-1116.3?
Not at all! It's more complicated than it looks in a paper like that.... and also simpler! Set it up as above, then let your ears and REW be the judge. Listen carefully to many tracks that you know very well with that setup, and do some tests with REW. Then try moving the listening position (chair) backwards and/or forwards in small increments of a couple of inches, and adjusting the speaker angles to compensate, then do some more critical listening with the same trakcs, and some tests with REW. Try moving the speakers closer together and/or further apart in small increments of a an inch at a time, one again adjusting the speaker angles to compensate ... critical listening... REW tests. Try adjusting the speaker angle in small steps without adjusting anything else. If you have a sub, then try moving the hole thing back into the room a bit, but allowing for the SBIR and the speaker cross-over frequency. You might get lucky and find a better location, but I'm betting it wont be far from that.

Of course, if you REALLY wanted your room to have spectacular response, then the single best thing you can do for it, is to flush-mount your speakers. (a.k.a. "soffit mount"). That would be the absolute finest thing you cold do.
I appreciate any input here as the dimensions of my room will change as I start building. I have another room that I intend to build out into a control room that I will need to figure
So you are going to have TWO control rooms in your studio? The one you are working on now, and also this "another room ... control room"?


- Stuart -
glymur
Posts: 5
Joined: Wed Jan 04, 2017 11:46 am
Location: Lansing, MI

Re: Questions surrounding speaker placement

Post by glymur »

Soundman2020 wrote:Hi Curt. Please read the forum rules for posting (click here). You seem to be missing a couple of things! :)
Sorry, I thought I had saved my location, but apparently it didn't. Updated. I'll read through the list again though to ensure I fulfill all the requirements.
That said, are you sure your room complies with the other requirements for a critical listening room? There are several equations in BS.1116 regarding the relationship between length, width and height. With a low ceiling and long room, are you sure that yours meets those? Regardless of whether or not they do, it's still gong to be problematic with a low ceiling.
It doesn't, not in it's current form. The long story is that I have an entire side of a basement to build into my dream setup. One portion of it, the measurements I gave you, is currently setup as a "home theatre", and the rest is unfinished basement. I have been studying studio design and construction and want to understand what it is I am learning. My thought then is that I'll take the room that is further along, measure it as a "starting point" for reference once I start building out the full basement into a live room and control room.

The one thing that was tripping me up was monitor placement, hence this post.
Soundman2020 wrote:The ITU guidelines and other rule-of-thumb figures that you mention are just that: guidelines and rules of thumb. In a perfect room with a perfect ratio, perfect volume, perfectly treated, and with perfect speakers, then that would be the perfect setup. But since there is no such thing, those are just starting points.
That was my suspicion, but I did not have the expertise to say that with confidence. This confirmation is exactly what I needed.
Having said that, there are a few "rules" that are really important. One of those is that the room muse be symmetrical, at least for the section between your head and the front wall. Another is that the room ratio should not be one of the bad ones (direct mathematical relationship between dimensions). Another is that the mix position should not be at the "terrible" spots of 25% or 50% of room depth. Another is that the distance between the rear wall and your ears should be as large as possible, in order that the reflections coming from behind fall outside the Haas time. Another: The speakers must all be the same distance from your ears, or if not then some must have their signal delayed so that all signals arrive at your ears at the same time, in phase. Etc.
Excellent, thanks for that list.
.- Speakers should be at least 1m (should the room allow it) from the front and sides of the room
That depends on the frequency range that the speaker is producing. If you have a bass-managed system where the mains are not putting out anything below about 120 Hz., then you could get closer, as the SBIR dip would not happen: the sub would be filling that in, not the mains. But if your mains are full-range, going all the way down to say 30 Hz, then I'd increase that minimum distance to 3m. The first SBIR notch for a speaker-wall distance of 3m is 29 Hz. So that "1m" number is a good rule of thumb for typical small speakers with little bottom end, or speakers in a bass managed system. But like all such "rules", it is only a starting point, and other factors come into play as well.
I’ll work on understanding SBIR better and read through the specs on my speakers to understand the range of frequencies I am dealing with.
For most rooms and most layouts, the point where the axes meet will be about 12" to 18" behind the center of your head, give or take a few.
Great! That makes a lot of sense. Also, I appreciate your explanation. :)
I'm also missing that formula! Because unfortunately, there is no such formula.
It is good for me to understand that, very much appreciate the explanation above.
For that room, here's how I would do it:
Speaker-Layout-Template-Generic-- glymur--S01-V1.png
Speaker-Layout-Template-Generic-- glymur--S01-V2.png
Speaker-Layout-Template-Generic-- glymur--S01-V3.png
Speaker-Layout-Template-Generic-- glymur--S01-V4.png
Speaker-Layout-Template-Generic-- glymur--S01-V5.png
Speaker-Layout-Template-Generic-- glymur--S01-V6.png
Speaker-Layout-Template-Generic-- glymur--S01-V7.png
Wow, thank you for laying that all out. I really appreciate you spending the time to provide those images.
Set it up as above, then let your ears and REW be the judge. Listen carefully to many tracks that you know very well with that setup, and do some tests with REW. Then try moving the listening position (chair) backwards and/or forwards in small increments of a couple of inches, and adjusting the speaker angles to compensate, then do some more critical listening with the same trakcs, and some tests with REW. Try moving the speakers closer together and/or further apart in small increments of a an inch at a time, one again adjusting the speaker angles to compensate ... critical listening... REW tests. Try adjusting the speaker angle in small steps without adjusting anything else. If you have a sub, then try moving the hole thing back into the room a bit, but allowing for the SBIR and the speaker cross-over frequency. You might get lucky and find a better location, but I'm betting it wont be far from that.
I had planned on doing that, but wanted to understand if there was a “best” position and then adjust from there. Your response sets me up perfectly to experiment and understand what it is I am doing here. Again, thank you for the extensive information.
Of course, if you REALLY wanted your room to have spectacular response, then the single best thing you can do for it, is to flush-mount your speakers. (a.k.a. "soffit mount"). That would be the absolute finest thing you cold do.
Yep, I do plan on doing this…eventually
So you are going to have TWO control rooms in your studio? The one you are working on now, and also this "another room ... control room”?
I don’t have all my plans laid out yet, but with the setup I have now, I do plan on the control room being the other room. Right now it is just a bare basement. That will take sometime to accomplish. I have a lot to do to make the “home theatre” into a proper live room first though. The following has to be done first:

- Remove the carpeting
- Remove the drop ceiling acoustic panels

Those two things need to happen regardless of the layout I come up with. I also need to finish the basic layout of what I have in Sketchup so I can start my design/build thread on here. Though I am also considering hiring professionals to help me. I love learning, and there is no better way to learn than from someone who is very good at what they do.

Ultimately, I want a space where I can be creative without compromise. I want to build my dream setup.
Post Reply