Home |
Search |
Today's Posts |
#1
Posted to rec.audio.pro
|
|||
|
|||
Design for speaker box to evenly distribute sound
Hi, I need to find a way to use a single speaker in a fairly dead room but I need to create a sound field in the room that is the same level at least 500 mm around and in all directions (or as close as I can achieve) near to the centre of the room.
The measurement standard only allows one speaker to be used and normally you would use a fairly reverberant space (RT 0.8 up to 1.6 sec) to achieve a steady field at the room centre far enough from the source. Only trouble is that I don't have access to such a room that is quiet enough but I do have access to an Audiology testing room but it is just too dead I am thinking of two, three or four sounding boxes connected to a common speaker by an open tube. So that there is effectively multiple sources of sound. Doesn't matter if the freq response is a bit non-linear as long as it is stable and the sources would need to be incoherent which I expect will occur if the connecting tubes are different lengths. Any ideas?? Never really delved into speaker design before so would really appreciate some ideas from an experienced designer. Regards Tom |
#2
Posted to rec.audio.pro
|
|||
|
|||
Design for speaker box to evenly distribute sound
On 1/12/2016 10:19 a.m., Tom Harper wrote:
Hi, I need to find a way to use a single speaker in a fairly dead room but I need to create a sound field in the room that is the same level at least 500 mm around and in all directions (or as close as I can achieve) near to the centre of the room. The measurement standard only allows one speaker to be used and normally you would use a fairly reverberant space (RT 0.8 up to 1.6 sec) to achieve a steady field at the room centre far enough from the source. Only trouble is that I don't have access to such a room that is quiet enough but I do have access to an Audiology testing room but it is just too dead I am thinking of two, three or four sounding boxes connected to a common speaker by an open tube. So that there is effectively multiple sources of sound. Doesn't matter if the freq response is a bit non-linear as long as it is stable and the sources would need to be incoherent which I expect will occur if the connecting tubes are different lengths. Any ideas?? Never really delved into speaker design before so would really appreciate some ideas from an experienced designer. Regards Tom Single speaker that points up vertically, with an 'acoustic lens' parabolic reflector thingy that bounces it out 360 ° in the horizontal plane. Available commercially. May we enquire for what purpose ? geoff |
#3
Posted to rec.audio.pro
|
|||
|
|||
Design for speaker box to evenly distribute sound
On Wednesday, November 30, 2016 at 4:19:26 PM UTC-5, Tom Harper wrote:
Hi, I need to find a way to use a single speaker in a fairly dead room but I need to create a sound field in the room that is the same level at least 500 mm around and in all directions (or as close as I can achieve) near to the centre of the room. The measurement standard only allows one speaker to be used and normally you would use a fairly reverberant space (RT 0.8 up to 1.6 sec) to achieve a steady field at the room centre far enough from the source. Only trouble is that I don't have access to such a room that is quiet enough but I do have access to an Audiology testing room but it is just too dead I am thinking of two, three or four sounding boxes connected to a common speaker by an open tube. So that there is effectively multiple sources of sound. Doesn't matter if the freq response is a bit non-linear as long as it is stable and the sources would need to be incoherent which I expect will occur if the connecting tubes are different lengths. Any ideas?? Never really delved into speaker design before so would really appreciate some ideas from an experienced designer. Regards Tom by 500 mm do you mean 500 meters? by "single speaker" do you mean single speaker box with possibly multiple speakers in the one box or do you mean actually only one speaker driver in the box? in all directions do you mean 360 deg around azimuth, or do you mean above and below also? m |
#4
Posted to rec.audio.pro
|
|||
|
|||
Design for speaker box to evenly distribute sound
Tom Harper wrote:
Hi, I need to find a way to use a single speaker in a fairly dead room but = I need to create a sound field in the room that is the same level at least = 500 mm around and in all directions (or as close as I can achieve) near to = the centre of the room. What does "the same level" mean? +/- 10dB? What frequency range do you care about? The measurement standard only allows one speaker to be used and normally yo= u would use a fairly reverberant space (RT 0.8 up to 1.6 sec) to achieve a = steady field at the room centre far enough from the source. Only trouble is= that I don't have access to such a room that is quiet enough but I do have= access to an Audiology testing room but it is just too dead =20 What measurement standard? What are you trying to measure and why? Reverb rooms exist but they are comparatively narrowband; they can give uniform response throughout the room over about four octaves. This is good enough for many speech measurement applications. They are expensive but theh are available. I am thinking of two, three or four sounding boxes connected to a common sp= eaker by an open tube. So that there is effectively multiple sources of sou= nd. Doesn't matter if the freq response is a bit non-linear as long as it i= s stable and the sources would need to be incoherent which I expect will oc= cur if the connecting tubes are different lengths. This does little to deal with room modes. The speaker is easy, it's the room that is hard. Any ideas?? Never really delved into speaker design before so would really = appreciate some ideas from an experienced designer. What are you trying to measure? Why? How accurately? --scott -- "C'est un Nagra. C'est suisse, et tres, tres precis." |
#5
Posted to rec.audio.pro
|
|||
|
|||
Design for speaker box to evenly distribute sound
Conducting unoccluded hearing testing to test hearing protectors
|
#6
Posted to rec.audio.pro
|
|||
|
|||
Design for speaker box to evenly distribute sound
In article ,
Tom Harper wrote: Conducting unoccluded hearing testing to test hearing protectors Ahh, so zwislocki coupler on the hearing protector and you likely care only about 500-5000 Hz anyway, right? And likely you can just excite the room and get uneven response in the room and not worry about it since you're looking at the difference between two curves and any room response issues will drop out when you subtract the two. Don't think about waves or sound fields, think about pressures. --scott -- "C'est un Nagra. C'est suisse, et tres, tres precis." |
#7
Posted to rec.audio.pro
|
|||
|
|||
Design for speaker box to evenly distribute sound
On Thursday, 1 December 2016 08:57:09 UTC+11, wrote:
On Wednesday, November 30, 2016 at 4:19:26 PM UTC-5, Tom Harper wrote: Hi, I need to find a way to use a single speaker in a fairly dead room but I need to create a sound field in the room that is the same level at least 500 mm around and in all directions (or as close as I can achieve) near to the centre of the room. The measurement standard only allows one speaker to be used and normally you would use a fairly reverberant space (RT 0.8 up to 1.6 sec) to achieve a steady field at the room centre far enough from the source. Only trouble is that I don't have access to such a room that is quiet enough but I do have access to an Audiology testing room but it is just too dead I am thinking of two, three or four sounding boxes connected to a common speaker by an open tube. So that there is effectively multiple sources of sound. Doesn't matter if the freq response is a bit non-linear as long as it is stable and the sources would need to be incoherent which I expect will occur if the connecting tubes are different lengths. Any ideas?? Never really delved into speaker design before so would really appreciate some ideas from an experienced designer. Regards Tom by 500 mm do you mean 500 meters? by "single speaker" do you mean single speaker box with possibly multiple speakers in the one box or do you mean actually only one speaker driver in the box? in all directions do you mean 360 deg around azimuth, or do you mean above and below also? m 500 mm it is the space around the head of the person being tested and yes by one speaker I mean one driver. The standard used in testing does allow for multiple speakers but I would have to change all of the measurement set up to do so The all directions bit is 360 deg azimuth in three planes at 15 deg increments +/- 6 dB around the 360 in all three planes. Thanks |
#8
Posted to rec.audio.pro
|
|||
|
|||
Design for speaker box to evenly distribute sound
On Thursday, 1 December 2016 09:00:29 UTC+11, Scott Dorsey wrote:
Tom Harper wrote: Hi, I need to find a way to use a single speaker in a fairly dead room but = I need to create a sound field in the room that is the same level at least = 500 mm around and in all directions (or as close as I can achieve) near to = the centre of the room. What does "the same level" mean? +/- 10dB? What frequency range do you care about? The measurement standard only allows one speaker to be used and normally yo= u would use a fairly reverberant space (RT 0.8 up to 1.6 sec) to achieve a = steady field at the room centre far enough from the source. Only trouble is= that I don't have access to such a room that is quiet enough but I do have= access to an Audiology testing room but it is just too dead =20 What measurement standard? What are you trying to measure and why? Reverb rooms exist but they are comparatively narrowband; they can give uniform response throughout the room over about four octaves. This is good enough for many speech measurement applications. They are expensive but theh are available. I am thinking of two, three or four sounding boxes connected to a common sp= eaker by an open tube. So that there is effectively multiple sources of sou= nd. Doesn't matter if the freq response is a bit non-linear as long as it i= s stable and the sources would need to be incoherent which I expect will oc= cur if the connecting tubes are different lengths. This does little to deal with room modes. The speaker is easy, it's the room that is hard. Any ideas?? Never really delved into speaker design before so would really = appreciate some ideas from an experienced designer. What are you trying to measure? Why? How accurately? --scott -- "C'est un Nagra. C'est suisse, et tres, tres precis." The standard is AS/NZS 1270 for hearing protectors, the variation is 6 dB for the directional measurements and 3 dB (i think) for the spatial variation. Hearing you about room modes but it is octave measurements so that rounds it out a bit freq of interest is about 125 to 8kHz so I am not too worried about nodes. Thanks, this is keeping me thinking |
#9
Posted to rec.audio.pro
|
|||
|
|||
Design for speaker box to evenly distribute sound
Tom Harper wrote:
Hi, I need to find a way to use a single speaker in a fairly dead room but I need to create a sound field in the room that is the same level at least 500 mm around and in all directions (or as close as I can achieve) near to the centre of the room. The measurement standard only allows one speaker to be used and normally you would use a fairly reverberant space (RT 0.8 up to 1.6 sec) to achieve a steady field at the room centre far enough from the source. Only trouble is that I don't have access to such a room that is quiet enough but I do have access to an Audiology testing room but it is just too dead ** The test method REQUIRES a reverberant room. Any room that has bare walls and floor will do fine. Its the reverberation that creates a non-directional sound field. I am thinking of two, three or four sounding boxes connected to a common speaker by an open tube. ** That idea is bonkers. Find a suitable room, use a small single cone speaker and an octave Eq to get the response even. The "quiet enough" condition sounds fake to me - you can do the test at some time when the background is low OR just use more level into the speaker. ..... Phil |
#10
Posted to rec.audio.pro
|
|||
|
|||
Design for speaker box to evenly distribute sound
Tom Harper wrote:
You haven't answered either one of these questions, and you need to answer them to your own satisfaction. The standard you referenced does answer them but you need to read it and know why. The standard is AS/NZS 1270 for hearing protectors, the variation is 6 dB for the directional measurements and 3 dB (i think) for the spatial variation. This standard requires a reverb chamber... a room that is highly reverberant and which has a single speaker that _excites the room_. The sound hitting the device is the sound from the room, not so much the sound from the speaker. This is a specially designed chamber that needs to be built to meet certain standards. The relevant one used to be ANSI S1.2-1962 although this is likely superseded by a newer standard. A good discussion of such rooms and some good guidelines for construction can be found in the General Radio Handbook of Noise Measurement, which I highly recommend. It is a dated book but the physics have not changed. Hearing you about room modes but it is octave measurements so that rounds it out a bit freq of interest is about 125 to 8kHz so I am not too worried about nodes. Room modes are a serious problem, but the purpose of the reverb chamber is to blend them all together into one big mush. It's much easier to make as many as possible than to eliminate them all. --scott -- "C'est un Nagra. C'est suisse, et tres, tres precis." |
Reply |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Forum | |||
Sound Effects Libraries for Foley & Sound Design | Pro Audio | |||
Sound Effects Libraries for Foley & Sound Design | Pro Audio | |||
Sound Effects Libraries for Foley & Sound Design | Pro Audio | |||
We distribute them, then we superbly abandon Simon and Laura's maximum demand. | Car Audio | |||
speaker design | Tech |