Reply to thread

Message

<blockquote data-quote="Ivan Beaver" data-source="post: 40913" data-attributes="member: 30">Re: What does architect provide for acoustical prediction?Randy nailed it.&nbsp; However it is VERY important to realize what the preduiction can and can&#039;t provide.how high of a freq do you want to predict to?&nbsp; You need ot model to at least (at a minimum) 1/2 wavelength.&nbsp; So if you only want to go to 2Khz (not very high) you will need to model everything 3&quot; or larger. That is A LOT of detail that most people do not even attempt to do.The other thing is the data available for your wall materials.&nbsp; The specs that are available are generally assuming an incident angle of 90°.&nbsp; So when sound strikes the panel at a different angle (which is most of the case) the data is invalid.Just things you have to consider.&nbsp; The results you get are only as good as the data that is inputed-on all fronts.You also need to careful of the data of the loudspeakers used.&nbsp; There are a number of products that have values for 1/3rd oct 5 degree polars.&nbsp; However they are not actually measured that way.&nbsp; A lot of 1 oct 10 degree data is actually measured, and then the averages between each of next points are used as the 1/3rd 5° values.&nbsp; Which is not the actual data.And then some data is just &quot;smoothed&quot; to look better than it actually is.You really have to dig a bit deeper sometimes.</blockquote>

[QUOTE="Ivan Beaver, post: 40913, member: 30"] Re: What does architect provide for acoustical prediction? Randy nailed it. However it is VERY important to realize what the preduiction can and can't provide. how high of a freq do you want to predict to? You need ot model to at least (at a minimum) 1/2 wavelength. So if you only want to go to 2Khz (not very high) you will need to model everything 3" or larger. That is A LOT of detail that most people do not even attempt to do. The other thing is the data available for your wall materials. The specs that are available are generally assuming an incident angle of 90°. So when sound strikes the panel at a different angle (which is most of the case) the data is invalid. Just things you have to consider. The results you get are only as good as the data that is inputed-on all fronts. You also need to careful of the data of the loudspeakers used. There are a number of products that have values for 1/3rd oct 5 degree polars. However they are not actually measured that way. A lot of 1 oct 10 degree data is actually measured, and then the averages between each of next points are used as the 1/3rd 5° values. Which is not the actual data. And then some data is just "smoothed" to look better than it actually is. You really have to dig a bit deeper sometimes. [/QUOTE]

Verification