AFMG Network Forum
AFMG Network Forum Index -> EASE 4 -> Strange roll-off in reverb time below 250Hz?
Post new topic  Reply to topic View previous topic :: View next topic 
Strange roll-off in reverb time below 250Hz?
PostPosted: Fri Jul 08, 2005 7:48 pm Reply with quote
dblount
Member
 
Joined: 28 Jun 2005
Posts: 7
Location: New York, NY




Hello again,

I've finally got this project where i need it, thanks to everyones help. Very Happy after looking over the drawing with our principle, the steep roll-off in reverb time below 250 seems strange. i think is it due to the large amount of GWB (~.55 at 125) in the hall...but in reality, this seems inaccurate.

the project is located at:

http://www-personal.umich.edu/~dblount/Burchfield%20with%20curtains.zip

any help?

thanks,

denis.
View user's profile Send private message Visit poster's website
PostPosted: Tue Jul 12, 2005 10:46 am Reply with quote
Tom Donohue
Member
 
Joined: 12 May 2005
Posts: 7




Use the 1/2" Gyp material instead of the 5/8". That LF data for the 5/8" does not appear to be correct.

Tom Donohue
View user's profile Send private message
PostPosted: Wed Jul 27, 2005 5:01 pm Reply with quote
Jim Mobley
Member
 
Joined: 27 Jul 2005
Posts: 134
Location: Foothill Ranch, California, USA




Tom Donohue wrote:
Use the 1/2" Gyp material instead of the 5/8". That LF data for the 5/8" does not appear to be correct.

Tom Donohue
This is one of the things that I point out in EASE classes; that simulations using the 1/2 inch GWB better track real-world measurements than those using the 5/8 inch. As the LF absorption of gypsum board is due to diaphragmatic action, (the panels flex, thus absorb energy) it makes sense that the thicker material would absorb less energy and thus have a lower absorption coefficient. However, a glance at the values in the database shows that at 125hz, the 5/8" (16 mm) material has more than twice the absorption than the 1/2" (12.5 mm) material.

Which is not to say that the database is wrong. On the contrary, these values for gypsum board/drywall are published in numerous reference works on the subject of sound system design and acoustics including "Sound System Engineering," by Davis & Davis and "Architectural Acoustics" by Egan.

Further complicating matters is that the LF absorption of all diaphragmatic absorbers is highly dependant on the construction of the wall or ceiling in question. Stud material and spacing, the presence of insulation and any backing board will all change the absorption of gypsum board walls at the low frequencies.

In California and other areas where buildings aren't supposed to fall down in earthquakes, it's common to use plywood or OSB (Oriented Strand Board) as shear reinforcement behind gypsum wallboard. This of course stiffens the wall construction and reduces the LF absorption. It's generally very difficult to ascertain (without drilling holes in the walls) how much of the room is built in this manner.

I've created a material that I call CA_GYPBRD that I use for most new construction in place of the 1/2" or 5/8" materials in the database. The coefficients are the same as the GYPBRD 1/2 material in the database, except that at frequencies from 100Hz to 200Hz I use a coefficient of 0.15. This seems to track quite well with measured results in rooms of mixed plain and shear reinforced gypsum board construction.

Keep in mind that for many sound system designs, much of this is irrelevant to the primary goal of the model, which is to get accurate predictions of intelligibility. In that case, it's the middle frequencies that matter, 500Hz-2kHz, and errors, even substantial ones, at 100-200Hz are inconsequential to the question of intelligibility. It's always nice to know if you have a problem at low frequencies and certainly we all want accuracy and good correlation between simulated and measured results, but for sound system work, chasing ultimate accuracy at all frequencies can easily turn into an expensive exercise with little real return. We're all supposed to be making money with this stuff.

As in all simulations, a through understanding of what's being simulated and how will go a long way towards proper interpretation of the results.

_________________
Best regards,

Jim Mobley
Sr. Application Engineer
Renkus-Heinz, Inc.
19201 Cook St.
Foothill Ranch, CA 92610 USA
TEL: +1-949-588-9997
FAX: +1-949-588-9514
www.renkus-heinz.com
mailto:jim@renkus-heinz.com
View user's profile Send private message Send e-mail Visit poster's website
AFMG Network Forum Index -> EASE 4
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
All times are GMT  
Page 1 of 1  

  
  
 Post new topic  Reply to topic  


Powered by phpBB © 2001-2003 phpBB Group
Theme created by Vjacheslav Trushkin
Variation by CodeWeavers and AFMG