Speakers next to window or super chunks?

[hazza26]

Hi all

My Dynaudio BM15a's + sub are currently on/under my work station, but I'm considering moving them closer to the wall on stands to improve the bass response. Desk is currently 90cm from the wall.

Would you a) put the speakers up against the window and move the desk closer, or b) move the speakers up against the super chunks and keep desk in (roughly) the same position?

Many thanks
Harry

speakers move.jpg

[DanDan]

Try both positions, and optimise the listening position in each case.
DD

[Soundman2020]

My Dynaudio BM15a's + sub are currently on/under my work station, but I'm considering moving them closer to the wall on stands to improve the bass response.

Definitely! That's a smart move.

Desk is currently 90cm from the wall.

OK, but what are the dimensions of the room, and where is the mix position (location of your head)?

Would you a) put the speakers up against the window and move the desk closer, or b) move the speakers up against the super chunks and keep desk in (roughly) the same position?

It's hard to say for sure without knowing the dimensions of the room, but most likely the best location for the speakers would be up against the glass at the point where the superchunks meet the glass. The best location of the mix position can then be determined from the room dimensions, and perhaps by using REW to optimize that. The location of the desk is determine by the location of the mix position: after you find the correct spot for placing your head in teh room, you then move the desk to be in a comfortable position for mixing.

Contrary to popular belief, the best location for the mix position is not determined by the location of the speakers, nor by any mythical triangles... There's a much better method for figuring that out.


- Stuart -

[hazza26]

Thanks so much for the replies. Yes will certainly be doing some trial & error with REW. Just trying to find out if there's a theoretical best position (I've read conflicting things about glass being reflective / not a problem).

Room dimensions are 6.1m x 4.2 x 2.3.

Not sure exactly where the mix position is (can check in the morning), but it's movable.

[Soundman2020]

Room dimensions are 6.1m x 4.2 x 2.3.

That's a fairly decent size for a room. It should be possible to get good results with that.

Just trying to find out if there's a theoretical best position (I've read conflicting things about glass being reflective / not a problem).

Glass is both! And also neither... Sorry to be cryptic, but that's the reason why you might be seeing conflicting things.

It works like this: As with any other building material, glass will reflect some amount of sound back into the room, and will allow some other amount of sound to pass through, which is a process called "transmission". How much it allows through, and what frequencies it allows it through at, depends on one thing alone: mass. Or rather "surface density", to be more precise. If you could cut out one square meter of the glass and weigh that, then that's the surface density of your glass. Or if you don't want to chop up your windows, you can calculate what that would be! The absolute density of glass is around 2500 kg/m3, so if you know the thickness of your glass (in meters), then just multiply 2500 by that number, and .. Bingo! you know the surface density of your glass. So, for example, if you have 6mm glass, then multiply 2500 by 0.006 meters = 15. Thus, the surface density is 15 kg/m2. Done! Or if you have 4mm glass then 2500 x 0.006 = 10 kg/m2. etc.

Once you know the surface density, you can use another equation to tell you how well that amount of mass will stop sound, for any given frequency. This is called the "mass law" equation, and it is also very simple:

TL(dB)= 20log(M) + 20log(f) -47.2

Where:
M is the surface density of the panel (mass per unit area (kg/m²) ), and
F is the center frequency of any one-third-octave measurement band

So, if you have 6mm glass, and you want to know how much sound it will stop for frequencies around 800 Hz, then it's dead easy:

TL(dB)= 20log(15) + 20log(800) -47.2
TL(dB)= 20x1.176 + 20x2.903 -47.2
TL(dB)= 23.52 + 58.1 -47.2
TL(dB)= 34.42 decibels.

So your window will reduce the transmission at 800 Hz by a total of about 35 dB. For obvious reasons, this is called "transmission loss", and that's what the "TL" part of the above equation stands for.

Now for the fun part: whatever the wall does NOT allow to go through, will be reflected back into the room! (Not all of it, as some is absorbed inside the wall, but most of it). So, if you happen to be playing an 800 Hz tone at a level of 85 dB inside your room, then the TL for your glass is 35 dB, so if you were to check the level outside the glass (on the other side), the level would be 85-35 = 50 dB. But the 35 dB that was stopped by the wall, stayed inside the room! So now you have an additional 35 dB of 800 Hz tone bouncing around inside your room, along with the original 85 dB sound that is still coming out of your speakers, so you have a tiny bit more than 85dB of that tone in your room. In fact, the difference is so tiny that for all practical purposes, the level is still 85 dB inside the room (the actual answer is 85.027 dB, in case you really wanted to know! ). So not a big deal at all!... at least, not for 800 Hz. But what about 18,000 Hz? Hmmm... for 18kHz, your glass will stop nearly 62 dB (so the level on the outside will be just 23 dB, which is not audible... but those 62 dB stayed inside, and added to the 85 that is already in there you get a whopping... 85.6 dB! Still not much at all.... except that it can be a lot, because it might all be focused in one single reflection, rather than being spread around in general... and in that case, it could be as much as 91 dB, and that's a problem!

So, glass can be reflective for high frequencies, but not for low frequencies... unless it is very thick. And that's the same for all other building materials! You'll notice that there is nothing at all in the mass law equation to say what TYPE of mass we are talking about: it's just "mass". So if you had a piece of plywood with a surface density of 15 kg/m2, it will transmit and reflect the same levels as a piece of glass with 15kg/m2, and both of those would be the same as a piece of concrete with 15kg/m2, as well as a piece of drywall with 15kg/m2, or a steel plate with 15kg/m2. What matters is the MASS, and ONLY the mass. That defines how well the material will reflect or transmit. What would be different, is the thickness: steel is much more dense than glass (about three times more dens), so a steel plate at 15kg/m2, would be only 2mm thick, instead of the 6mm you need for glass... or the 26mm of plywood you would need to get the same 15 kg/m2. That's what changes: thickness. But all materials of the same surface density (even if one is much thicker or thinner than another) will reflect and transmit the same amounts of sound. Because Mass Law is a fixed, careved-in-stone law that is correct.

OK, so no you know how to figure out the transmission and reflection for various densities of materials: how does that help you here with your original question?

Simple: Thin glass will transmit most of the bass out of your room, while reflecting back some of the highs. But since your speakers are pointing AWAY from the glass (in your case), no highs will get to the glass! So there will be nothing to reflect. With thin glass, the bass just goes right out through it... mostly (some small amount gets reflected back). But if you have very thick glass there, then you could end up with substantial low frequency reflections doing nasty things in there, such as causing modal issues, or SBIR.

In other word: it all depends on the thickness of your glass.

So why might you want thick glass? Well, you might be happy that all your bass is rushing out through the glass and not causing problems in your room, but your neighbors might not be to happy about it! And your family too.... So you might need to isolate your room, to STOP the bass from getting out, so your neighbors don't lynch you at midnight with flaming pitchforks.... In that case, you might need to have a window with very high transmission loss, that stops most of the sound from getting out, and therefore reflects it all back into the room....

So that's what you are facing here: a whole bunch of possible issues that you should probably think about, before deciding on how to proceed. There might well be reflection issues with the glass, such as modal issues, or SBIR, or comb filtering, or flutter.... and/or there might be isolation issues that you need to deal with first.

Acoustics is a lot more complex than it looks at first glance! Sometimes there are no simple answers to what seems like a simple question, because there's so many factors to consider.

Now, having said all that, let me give you a partial answer to your question:

Soundman--Speaker-layout-tool-- hazza26-0001.jpg

Soundman--Speaker-layout-tool-- hazza26-0002.jpg

Soundman--Speaker-layout-tool-- hazza26-0003.jpg

Sorry about the image width: I wrote that tool a while back to help me with quick room prototyping, but it only puts out one specific image size.

So, that's the THEORETICAL layout you could start with, but it does make a lot of assumptions. It's worth trying, then use REW and the Walking Mic tests to optimize.

- Stuart .

[hazza26]

Stuart - WOW, incredibly grateful for your lengthy response! Loving the physics stuff

Ok well that clears up why I've been getting mixed messages about glass. My windows are double-glazed, 2 panes @ maybe 2mm thick with ~12mm gap between. No idea how that plays into the calculations but I'm up for getting my calculator out! No issues with neighbours.

Thanks so much for the suggested layout. Funnily enough the width of my window is exactly the same as your suggested speaker spacing (227cm). Your suggested listening position is actually FURTHER away from the wall than my current position (205-210cm). Your triangle of speakers / ears doesn't quite look equilateral, perhaps that's another myth I have taken on board but is it necessary to have them pointing slightly further down the room?

Changing my Dynaudios from nearfield to midfield is going to take a bit of getting used to, perhaps I should consider some additional smaller nearfields on my desk for a balanced perspective. Really looking forward to testing this all out through REW