As can be seen in the photos of my listening room, there is very little sound absorbent material. A clap generates a very distinct slap echo. When the piano is played, it can be quite overpowering. The only saving grace is that the room is fairly large, at 6 metres wide and 18 or so metres deep, this lets me sit the speakers out into the room a bit. About half of the left wall furthest from the speakers comes in around two and a half metres. The corner of this wall can be seen in tlees.jpg. In this photo, the dining room table can be seen in the foreground, and closer to the camera again is a breakfast bar which is on one end of the kitchen. I've found that the line source nature of the ribbon speakers helps a lot, it keeps a lot of the sound off the floor and ceiling. A conventional 2 way tower type speaker using a Dynaudio 17w75xl and Esotec tweeter sounded bad in this room, even though it sounded wonderful in a less live environment. On the wall opposite the left hand speaker, I'm planning to build a floor to ceiling bookshelf. This should break up the reflection coming off this wall quite well.
For the wall behind the speakers, I'm planning to make some acoustic panels. These panels will cover the wall from each side of the piano to the corners of the room. I may also make a panel for each side wall beside the speakers.
Joseph Saluzzi did a great series of articles in Speaker Builder in 1992/93, called What Makes Your room HiFi?. In these articles he describes how to make some attractive cloth covered panels. The panels use 25 by 200mm pieces of timber around the outside. Another 2 pieces of 50 by 200mm timber divide the panel into 3 compartments. There are two pieces of 25 by 25 mm pine which are used to secure the whole assembly to the wall.
One of the three compartments house a Helmholtz resonator which was tuned for around 60Hz. This compartment has 6mm hardboard panels on the back and front. The front hardboard panel is drilled 6mm on about 90mm centres. The cavity is then filled with fibreglass or rockwool.
The other two compartments are filled with rockwool or fibreglass. If fibreglass is used, it can be covered in BAF (polyester) to keep the fibres in. These two compartments are then covered in chicken wire and a decorative layer of burlap covers the whole assembly.
There were two great articles on the Live End / Dead End concept in Audio, December 1986 and January 1987. These articles covered the modification of all room surfaces. It was highlighted in these articles just how much the low frequency absorbtion can be increased by spacing the panels from the wall. A 100mm cavity increased the absorbtion of a fiberglass panel from about 12% at 125Hz to around 58%. This is MUCH cheaper than thicker panels!
Knowing how to fix up the acoustics in my listening room is only half of the hurdle. It's only half my house, I'm not sure Danielle is too keen on the idea. One improvement which will get the nod is to put a bigger and thicker rug on the floor in front of the speakers. The kitchen area poses a bigger problem. A lot of sound is reflected out of there, because of all the hard surfaces. Acoustic panels in the kitchen will definitely not be on! L Luckily these reflections occur a long time after the first arrival from the speakers.
Sound absorbed vs frequency for 25Kg/M3 fiberglass.
|
Panel thickness/Frequency |
125 |
250 |
500 |
1000 |
2000 |
4000 |
50mm thick |
0.32 |
0.50 |
0.83 |
0.86 |
0.74 |
N/A |
75mm thick |
0.43 |
0.76 |
0.98 |
0.93 |
0.87 |
0.87 |
Sound absorbed vs frequency for 37.5Kg/M3 fiberglass.
|
Panel thickness/Frequency |
125 |
250 |
500 |
1000 |
2000 |
4000 |
50mm thick |
0.43 |
0.60 |
0.92 |
0.88 |
0.77 |
0.87 |
75mm thick |
0.68 |
0.80 |
0.99 |
0.97 |
0.82 |
0.79 |
Sound absorbed vs frequency for 50Kg/M3 fiberglass.
|
Panel thickness/Frequency |
125 |
250 |
500 |
1000 |
2000 |
4000 |
50mm thick |
0.55 |
0.69 |
0.96 |
0.94 |
0.87 |
N/A |
75mm thick |
0.72 |
0.97 |
0.99 |
0.97 |
0.90 |
0.81 |
Sound absorbed vs frequency for 35Kg/M3 polyester.
|
Panel thickness/Frequency |
125 |
250 |
500 |
1000 |
2000 |
4000 |
25mm thick |
0.30 |
0.50 |
0.75 |
0.80 |
0.85 |
0.90 |
50mm thick |
0.45 |
0.85 |
1.00 |
0.95 |
0.95 |
1.00 |
Sound absorbed vs frequency for other building materials.
|
Material/Frequency |
125 |
250 |
500 |
1000 |
2000 |
4000 |
Concrete, unpainted |
0.36 |
0.44 |
0.31 |
0.29 |
0.39 |
0.25 |
Concrete, painted |
0.10 |
0.05 |
0.06 |
0.07 |
0.09 |
0.08 |
Brick |
0.03 |
0.03 |
0.03 |
0.04 |
0.05 |
0.07 |
Plaster on lath |
0.14 |
0.10 |
0.06 |
0.05 |
0.03 |
0.03 |
Plywood panelling |
0.28 |
0.22 |
0.17 |
0.09 |
0.10 |
0.11 |
Light drapes |
0.03 |
0.04 |
0.11 |
0.17 |
0.24 |
0.35 |
Heavy drapes |
0.14 |
0.35 |
0.55 |
0.72 |
0.70 |
0.65 |
Terrazzo |
0.01 |
0.01 |
0.02 |
0.02 |
0.02 |
0.02 |
Wood floor |
0.15 |
0.11 |
0.10 |
0.07 |
0.06 |
0.07 |
Carpet |
0.02 |
0.06 |
0.14 |
0.37 |
0.60 |
0.65 |
Carpet on underfelt |
0.08 |
0.24 |
0.57 |
0.69 |
0.71 |
0.73 |
Ac. Tile suspended |
0.76 |
0.93 |
0.83 |
0.99 |
0.99 |
0.94 |
Ac. Tile on concrete |
0.14 |
0.20 |
0.76 |
0.79 |
0.58 |
0.37 |
Gypsum board |
0.29 |
0.10 |
0.05 |
0.04 |
0.07 |
0.09 |
Glass, large panes |
0.05 |
0.03 |
0.02 |
0.02 |
0.03 |
0.02 |
Glass, windows |
0.10 |
0.05 |
0.04 |
0.03 |
0.03 |
0.03 |
Marble or tile |
0.01 |
0.01 |
0.01 |
0.01 |
0.02 |
0.02 |
Water surface, pool etc. |
0.01 |
0.01 |
0.01 |
0.01 |
0.02 |
0.03 |
25 Kg/M3 = 2 Lbs/Ft3
An interesting experiment to do if you think you have a room problem is to calculate the rooms Rt60 or reverberation time at different frequencies.
To work out the Rt60 of your listening room, first you have to make a list of all of the different surfaces and their area. Looking at the table will give the absorption coefficients for the different surfaces. If your particular surface is not listed, choose one you think is close. Enter these numbers into the formulas below to calculate the Rt60 at different frequencies. A value of between 0.35 and 0.6 seconds is desirable.
Where:
V = room volume
S = total surface area of room
a = average Sabine absorption coefficient
S1 = area of material 1
a1 = absorption coefficient of material 1
S2 = area of material 2
a2 = absorption coefficient of material 2
S3 = area of material 3
a3 = absorption coefficient of material 3
I ran through the equations for my room and got a nasty fright. The figures I came to were: 125Hz 0.54 Sec, 250Hz 1.25 Sec, 500Hz 1.86Sec, 1000Hz 2.2Sec, 2000Hz 1.6Sec and 4000Hz 1.34 Sec. Apart from the low frequency end, my room is shocking. A lot of absorption is required in the midrange and upper midrange. I think the large areas of Gyprock were responsible for the better figures at 125Hz.
Although absorption is not every thing, (reduction of standing waves and a diffuse reflected sound field are also important) I think it is pretty obvious that I need to increase the absorption in my listening room. I calculated the effect of putting panels on the wall behind the speakers and also using a larger rug in front of the speakers. The new figures I got were: 125Hz 0.48 Sec, 250Hz 0.83 Sec, 500Hz 0.96Sec, 1000Hz 0.96Sec, 2000Hz 0.86Sec and 4000Hz 0.79 Sec. It can be seen that these modifications will improve the situation considerably. Even more absorption is still required in the mid range though.
One problem I can see regarding the lining of the back walls is what effect will it have on the character of the dipoles. A dipolar loudspeaker relies on its rear radiation to achieve its open and spacious sound. On a recent post to the Bass mailing list, Carl Huff (who had a similar problem) said:
Hey ribbon heads ....
Well I've had my BG75's up and running VERRRYYYY well for about a month now. I like them a lot. Since I work mostly at home and the BG75's are in my work area, I've spent a lot of hours listening to them. My NHT 1259's have just arrived (thank you Madisound), so I'll soon be adding those to my listening room as well. Until then I have adapted some other cabinets to service the bottom end of the spectrum.
If I have any complaint at all bout the BG's, it would be how they present the upper midrange. My best speakers previous to the BG's are a pair of aging ADS towers (recently factory modified). What I'm critical of is hard for me to describe. What I'm missing is that subtle edge to well recorded 'hi hats' and cymbals in closely miked jazz combos. My ADS towers revealed that well. The overall imaging of the BG's are fine. Mine are done as dipoles on a trapezoid baffle, per John Whitakers findings.
I don't think that what I'm describing is an electronics issue, as I'm using the same electronics to drive my BGs as I used to use on my ADS towers. Placement has a dramatic effect on imaging overall, but does not seem to address 'the edge' that I'm missing. I don't think that what I'm describing is a frequency response thing, as I've got a CROWN EQ2 parametric equaliser that I've put into the audio chain and have tried to tweak for the 'missing edge'.
Perhaps what I'm attempting to describe is the same thing as what others have attempted to describe when they referred to the BG's as unexciting.
Anyway, I was speculating that perhaps what I'm looking for is more of a 'point source' at that upper mid band. Perhaps the best imaging (if that is what I'm describing) is best realized by controlling the amount of reflected sound versus direct sound at those all important upper midband freqs.
Rudi Blondia wrote:
"There is indeed a school of thought (Linkwitz) that advocates the use of dipoles for the lows and mids only, so monopole highs. I did play around with it and I would definitely describe this as "uninteresting" You can easily try this out by carpeting the back wall. The highs are easier to absorb than lows."
Then Carl Huff followed:
Hmmm ... maybe you are onto something here. My listening room was originally a LEDE (Live End Dead End) room. But that was a long time ago, and changes have been made. However there still is an unusual amount of acoustic traps and sound absorbing materials there. Although my BGs are well away from the wall, they are in the old 'dead end'. The room has never been a problem for traditional 'front firing loudspeakers'. I had always assumed that the amount of reflected sound that was absorbed at that end of the room was linear across the audible band. Maybe it is not. I'll try introducing varying amounts of reflective surfaces within that old 'Dead End', and see how that changes the mid band.
I'm hoping that I don't experience a change in character in my ribbons like Carl did. I suppose the only way to find out is to experiment.
The results I obtained were a little different than I had expected. More amplifier power was needed to obtain the same perceived level. Ambience information on good recordings was much easier to hear, as it was not being smothered by all of the sound bouncing around my room. The tonal balance of my system also changed a little. I needed to do a little fine tuning of my active crossover output levels. The greatest difference, however, was the improvement in sound from the piano. In the untreated room, it could be quite overwhelming if played loudly. Now it sounds fantastic. My brother-in-law has a large grand piano in his untreated lounge room, and often comments that our small upright, which is worth around 1/7th of the grand, sounds a whole lot better. All in all, I am very happy with the results from this not too drastic change.