Is it a bad idea to place speakers near the ceiling (and why)?

[ethernode]

Hello everyone,

My room will be really challenging (see attached plan), and the most practical location would be above the windows (at 2,30m, ceiling is at 2,55m). My expected main listening position is the couch.

I'm pretty sure that it's a terrible idea to place speakers above windows like on this picture below, but i'd like to know why (acoustically speaking) ?

Practical pros:
- it would cover both table and couch areas
- it would not take floor space

Acoustical pros intuition (completely wrong i guess)
- it would avoid the windows as immediate reflection surface (there are no windows on the other side)
- i'm thinking the ceiling would not be a first reflection point as much as if it was floorstanding

Why i guess it's very wrong
- the tweeters would not be at ear level
- ?

Note that:
- the speaker are 2x 1st gen Kef LS50s (coaxial design so they can be side-mounted, see amir's review thread for dispersion graphs)
- they will be high passed, probably quite high like 100-120 Hz (sub will do low frequency duty), so i don't expect bass performance from them

I'll probably end up using position "A" but still, i'd love to learn why/if this is such a bad idea.

Also, there are no closing doors in that room, i read here and there that sealed rooms are better (?), what is the downside of having multiple connected rooms on bass modes ? Will that increase interference, modes, or reduce power ?

Thanks !

 

[amicusterrae]

You have a shelf drawn above the windows, and if the speakers are placed between the shelves and the ceiling as shown, I think you will have three adjacent boundaries, positioned at right angles, to the speaker (the shelf, the ceiling, and the wall itself). Compared with mounting the speaker on a stand out in the room, away from a wall, energy that would have radiated behind and to the sides of the speaker is reflected forward. The result is increased sound power; I think the math is at least +9dB. In real rooms, though, the increase is typically not even across the whole frequency range, but I'd expect a significant bass hump, maybe as high as 100Hz. I also would expect a lot of acoustical interference if seated on the couch, for example. Our brains have evolved to handle reflections, but I think you would see a really irregular response curve if you took measurements.

Speakers designed to be mounted in a wall or on a wall should be designed to account for this effect and likely would be the best choice if you were to go with that location. Even so, it's a poor place to place to put the speakers, unless you can lower them from the ceiling enough to reduce that boundary's effect, and ideally eliminate that drawn shelf above the windows completely.

In the drawing, it looks like you have the speakers on their sides, which could be a problem if the tweeter has an acoustic waveguide designed for upright placement. I've seen some in-walls that have a pivoting waveguide to allow for easier installation either vertical or horizontally oriented.

Personally, I hate when music comes from overhead or in this potential case--so high above the plane of the ears. I think there's some research about how our brains are more likely to localize sound from high sources in rooms. In my experience, a listener eventually adapts to a center speaker, for example, placed right on top of or under a large TV, but the underneath placement tends to blend in better. But that may be more applicable to dialogue--ie, movie watching.

I am not quite sure what you mean with your last question, but I think you are going to have bass nulls and peaks following the standing waves in these connected rooms. It looks like a lot of volume, too. I would consider one or dedicated subs with EQ.

 

[DWPress]

If you point the speakers downward pointing to your listening position for directivity I think you'd be ok and better than another stereo pair firing off from the left or right corrupting the stereo image. Are you using these in a 5.1 setup or just reinforcing L/R channels? There will be some reflections from the ceiling of course but since they're high passed the boundary gain may or may not be an issue, if you are able to use DSP to adjust amplitude, response and time align you'd be fine.

I take it you have the wired versions of the LS50s so probably a lot of effort to just "try it, see if you like it" but regardless measuring the implementation in the room will tell you what you want to know.

My speakers "fly" from the ceiling and I have no problems with an "overhead" sound but my speakers are just a little bigger.

 

[ethernode]

Hi, thanks for your feedback and advice, thats very useful !

The speaker is a coaxial design and it exhibits similar horizontal and vertical dispersion (60°), and I do use a sub for low frequency, with EQ (minidsp).

I could very much remove the shelf (i was thinking of experimenting speaker spacing) and just keep the minimum for supporting the speakers, and I could probably install some absorbing material behind and on the back wall (which will be isolated with mineral wool for thermal reasons anyway) to reduce the back-originating energy.

I cannot lower them though, they will just fit between the top of the window and the ceiling, although I could consider tilting them down a bit if the listening position is not inside the listening window.

 

[ethernode]

If you point the speakers downward pointing to your listening position for directivity I think you'd be ok and better than another stereo pair firing off from the left or right corrupting the stereo image. Are you using these in a 5.1 setup or just reinforcing L/R channels?

Oh I'm sorry I didn't mention that all the other speakers in the picture are just alternative placement options, there is just a pair, just for music (not dozens !). Sorry that's confusing I'll edit my OP to clarify.

There will be some reflections from the ceiling of course but since they're high passed the boundary gain may or may not be an issue, if you are able to use DSP to adjust amplitude, response and time align you'd be fine.

I plan on DSP yes.

I take it you have the wired versions of the LS50s so probably a lot of effort to just "try it, see if you like it"

Precisely that's my worry, i need to plan for cabling and I basically wanted to know if this is worth a shot (keeping that scenario in sight) or plain stupid.

but regardless measuring the implementation in the room will tell you what you want to know.

My speakers "fly" from the ceiling and I have no problems with an "overhead" sound but my speakers are just a little bigger.

View attachment 268598

Awesome picture, thanks for sharing, that's actually the first time I see someone implement this ! Yes, they are "just a little bigger"

Looking forward for other opinions and examples!

 

[gnarly]

Who said it's a bad idea?
I've had some very good results doing so, particularly when the speakers are "hung upside down" with the woofers closest to ceiling.

 

[ethernode]

Who said it's a bad idea?

Well, nobody seems to talk about that here, except for one or two rare cases where that option was pretty much rejected.

I've had some very good results doing so, particularly when the speakers are "hung upside down" with the woofers closest to ceiling.

Thanks !

Another question: if I kept the shelf, would it change something to the boundary issue if there were books on the free space between the speakers mentioned by @amicusterrae ?

 

[DVDdoug]

In the real world speaker placement is usually a compromise. To some people "soundstage" is important and they won't like hearing the sound coming from a high angle like that.

I have a TALL "speaker stack" and the (front) tweeters are near the ceiling. My rear speakers are not as large but they are supposed to be floor-standers and they are hanging by chains on the back wall above my couch, almost touching the ceiling.

 

[DWPress]

there is just a pair, just for music

In that case I'd go with position A which would be a lot easier and more conventional. Looks like you can cross them a lot lower than 100-120Hz though for better localization - something to work out when you get to the measurements phase.

With the ports on the rear placement and boundaries will change frequency response quite a bit in the room including crowding them with other objects on a shelf.

 

[ozzy9832001]

I think it comes down to taste, but you'll end up with a lot more room gain them if they were on up by the ceiling. Down on the floor, even with the windows, the windows handle sound differently then a wall. In most cases, the windows actually allow the bass to escape, so you'd probably want a subwoofer closer to the couch to sure it up.

Crossing over too high will localize the sub and give it a strange sound unless it's directly behind or in front of you.

Sealed is different then just closing a door. Sealed would be a door that is basically a wall you close. Same thickness, same treatment, etc. It forms a seal when closed. Honestly, it's unnerving to be in a room like that.

Have someone hold the speakers up there and give it a listen. I would think, just from what I know, that they will a) have very large room gain and room modes, especially the ceiling ones will be super excited, and b) will sound "off" because they are more top down.

 

[ethernode]

Down on the floor, even with the windows, the windows handle sound differently then a wall. In most cases, the windows actually allow the bass to escape, so you'd probably want a subwoofer closer to the couch to sure it up.

Thanks everyone for your insight.

From what I start to gather, near the ceiling it's going to be more complex and probably not worth it precabling for this scenario, even if just measurements can tell. I'm a bit torn...

Window-wise, would you say that position C has more potential than position A ? Is it best having the windows as back wall or front wall ?

There is no position that is really symmetrical in this room, unless i have missed something (e.g. in position A I'm afraid that the side opening will create imbalance).

 

[ozzy9832001]

Thanks everyone for your insight.

From what I start to gather, near the ceiling it's going to be more complex and probably not worth it precabling for this scenario, even if just measurements can tell. I'm a bit torn...

Window-wise, would you say that position C has more potential than position A ? Is it best having the windows as back wall or front wall ?

There is no position that is really symmetrical in this room, unless i have missed something (e.g. in position A I'm afraid that the side opening will create imbalance).

I'd say position A or E have the most potential if you flip the couch to face them.

 

[amicusterrae]

Another question: if I kept the shelf, would it change something to the boundary issue if there were books on the free space between the speakers mentioned by @amicusterrae ?

It's not going to eliminate the problem, but it probably will help a little.

 

[amicusterrae]

I could probably install some absorbing material behind and on the back wall (which will be isolated with mineral wool for thermal reasons anyway) to reduce the back-originating energy

Whether this is viable will depend on what frequencies you need to absorb (and how much room you have). An couple inches of fiberglass board, for example, isn't going to make a difference below 200 Hz.

 

[fpitas]

 

[ethernode]

View attachment 268846

Nice ! How high are these ? What's the vertical dispersion angle like with these horns ? No problem with ceiling reflections ?

 

[ethernode]

Whether this is viable will depend on what frequencies you need to absorb (and how much room you have). An couple inches of fiberglass board, for example, isn't going to make a difference below 200 Hz.

I was thinking of the 140mm / 5-6 inches of mineral wool (thermal isolation) behind the wallboard and above the ceiling. How much is needed to effect 100-200 Hz ?

Found this online, not sure it applies

 

[fpitas]

Nice ! How high are these ? What's the vertical dispersion angle like with these horns ? No problem with ceiling reflections ?

Lol those are ancient Altec 9846A studio monitors. The 511 horn is 90 x 40 degree dispersion at -6dB beamwidth.

 

[amicusterrae]

I was thinking of the 140mm / 5-6 inches of mineral wool (thermal isolation) behind the wallboard and above the ceiling. How much is needed to effect 100-200 Hz ?

Found this online, not sure it applies

View attachment 268892

Not sure what the source is but this is consistent with what I've seen for similar materials. Notice how the effectiveness drops sharply. Added thickness essentially moves the curve down (simplistically, lower frequencies have longer wavelengths). But, we don't know at what specific frequencies you would potentially benefit from absorption without in room measurements, and . . . you'd have to put the material on the outside of the walls--that is, on the interior surface--not behind the wallboard

 

[ethernode]

and . . . you'd have to put the material on the outside of the walls--that is, on the interior surface--not behind the wallboard

Lol, that's clearly not happening I'm afraid ! I was assuming that low frequency would pass the wallboard -- at least partially.