#20 Surface pressure magnitude mismatch between point source and vibrating element.

[Jay Deoe]

Hi
I am having problems matching the results I get using a vibrating element (with setting the material of an element inside the ear as "Left ear") and those from a point source placed in the ear very close, like 2mm away from the original vibrating element. I can inderstand the phase difference but why the magnitude of surface pressures is almost 1000 larger on all elements when simulated with the point source?

I saw in your output2HRTF_main.m script you normalize the results by the area of the element. This made me thought the difference might be just the area of the element, but that was not the case as the the area of the vibrating element is ~1e-8 in my case (I used the openflipper plugin to remesh my mesh). If normalizing by the area of the element cannot correct the mismatch can be the normalization part of the output2HRTF_main.m script be wrong? Certainly there's something here that I am missing. I will be grateful if you can enlighten me here.

Also on a different but related note, is there a reason you have chosen a default value of 0.1 as opposed to one for the intensity of the sources in your mesh2input blender plugin?

Thanks
John

[Fabian Brinkmann]

Hi John,

from my point of view the normalization seems works fine - HRTFs show a low frequency magnitude of close to 1 (0 dB) as expected. It might be that numCalc uses different constants for the definition of the the element and point source velocity.

Best, Fabian

[Fabian Brinkmann]

Hi John,

I found some time to look into that. The thing that was missing was the a/(4pir) normalization with a=0.1 being the hard coded point source amplitude and r the radius of the evaltion grid. This is now implemented in Output2HRTF_main.m (currently only in the dev branch)