problem with results

[stefano rossi]

Hi everybody!

I have been using i-simpa for some time now and i have a problem with the results according to the spps code.

• geometry of the model:

• results classical theory 

• results with spps 

• question: why are the results so different?

i used the following spps settings;

• Meshing:

• properties 

Thanks for your help.

[Augusto Pavon]

Hello!

As far as I'm concerned, there are no set up problems that may be causing the results you get with SPPS calculation. Maybe you can try checking the Surface Receivers Constraint option, that enables a more detailed description of what happens at the point of the receiver (it split the surface receiver in more surface elements, as explained in other post in this forum). 

Anyway, I think it is important to analyze in which cases you may expect an accurate aproximation of RT when using the Classical Theory approach. The values you see in the columns TR Sabine and TR Eyring are calculated by the use of equations developed by these people, that have certain assumptions on the acoustical properties of the room. It is to say, you can expect a good approximation of TR by these equations, only if your room fulfills some conditions. You can check this paper which explains some of them (note that in the Conclusion Section it says that the 'real' RT may be between Sabine and Eyring's ones, but it is not always true, i would say it depends on the case). 

When the situations of the room are not the ones assumed in the development of the simple theoretical equations, the simulation approach takes notificably importance, because it is the best alternative. So, in case you wouldn't have done it yet, I recommend you to annalyze whether the conditions mentioned above are sufficiently fulfilled in order to take into account the results given by the Classical Theory calculation.

I hope it may help you. Let me know it. 

Augusto

[stefano rossi]

Hi Augusto, 

Thanks for the reply.

Why do you think there is so much difference between the T20 and the T30?

thank you

[JP]

Hello.

I'm agree with Augusto. In your case, your room is very disproportionate (very long enclosure), probably with non-uniform absorption (and perhaps with large absorption coefficient values ?). It is a very different configuration from the ones for which the statistical theory of reverberation (Sabine or Eyring) could be applied, that is : quasi cubic room, uniform and low absorption, i.e. the diffuse field assumption. It could explain why the statistical theory give different results with the SPPS ones (that could be more correct in this case).

Concerning your last question, RT20 vs RT30, one possible relevant reason is that you don't have a linear decrease of sound energy at the receiver, due to the long room configuration. Take a look of the Schroeder curves to verify this point. 

Best regards and thank you for using I-Simpa.

J. Picaut

[stefano rossi]

HI J. Picaut, 

thanks for the reply.

I tried to do the analysis of the same model with another commercial software.

With I simpa I get higher values ​​greater than the reverberation time. While with the commercial software the values ​​are more similar to those of sabine.

for example:

[JP]

Thank you.

This is very strange.
Could you send me your project file ?

Best regards.

J. Picaut

[stefano rossi]

Hi, 

thank you. 

I tried to attach the file here but the site rejects it.

can you give me your email?