Nearfield measurement - box diffraction issue

[Marius Tanasescu]

All I want to do is make a full frequency response measurement ("quasi - anechoic" indoors) of my loudspeaker. So I make my far field and near field measurements. But before I splice them together, I need to fix the nearfield measurement. Since this type of measurement acts like a half space measurement, I need to compensate the lower frequency depending on the size of the baffle. Is there any 2π/4π equalizer function for given baffle dimensions ?
In ARTA there is an option called "LF box diffraction" where it lowers the lower frequencies depending on the size of the baffle. I can't find anything similar in REW.
Is there anyway to do this (maybe using excel or something)?

 

[Matthew J Poes]

All I want to do is make a full frequency response measurement ("quasi - anechoic" indoors) of my loudspeaker. So I make my far field and near field measurements. But before I splice them together, I need to fix the nearfield measurement. Since this type of measurement acts like a half space measurement, I need to compensate the lower frequency depending on the size of the baffle. Is there any 2π/4π equalizer function for given baffle dimensions ?
In ARTA there is an option called "LF box diffraction" where it lowers the lower frequencies depending on the size of the baffle. I can't find anything similar in REW.
Is there anyway to do this (maybe using excel or something)?

John can say for sure. I don’t believe there is an automated feature for this. You could use an excel work book that does this and create an frd file that you load into REW. Then multiple that response by the speaker response to compensate. I have an excel tool that does this but I would need to see what it’s called.

 

[AudiocRaver]

Zing, you have jabbed ab a real frustration for me, getting good full-range frequency response measurements inside. Or outside for that matter. My go-to method these days, short of building an expensive lift-tower outside, is to take a Moving Mic Method measurement as a reference (outdoors if possible), then use the REW math function to splice together pieces of close-up driver measurements to get the most accurate composite, suitable for publication.

 

[AudioCAT]

I am also concerned about this issue. After all, when using near field measurement to replace far field data, it requires box diffraction or low frequency drop curve to participate in the calculation. However, it is difficult to find a program that can accurately generate this curve, so that the far field curve cannot be stitched together.

This is important for designing or adjusting X-Over

 

[AudiocRaver]

Indeed. Doncha just love audio, traveling in an infinite number of "path of least resistance" directions at any moment, and we are left trying to figure out how to measure the result. For driver / cabinet / crossover design tweaking you are looking at extremely fine details. Al I can say about cabinet diffraction is that the best inexpensive speakers I have heard have paid attention to diffraction details in their design.

How to measure? No suggestions at this time. How to predict? Gotta be next to impossible, other than in the time domain, using impulse diagrams.

 

[Matthew J Poes]

Indeed. Doncha just love audio, traveling in an infinite number of "path of least resistance" directions at any moment, and we are left trying to figure out how to measure the result. For driver / cabinet / crossover design tweaking you are looking at extremely fine details. Al I can say about cabinet diffraction is that the best inexpensive speakers I have heard have paid attention to diffraction details in their design.

How to measure? No suggestions at this time. How to predict? Gotta be next to impossible, other than in the time domain, using impulse diagrams.

Hey Wayne, I think he is referring to Baffle step specifically.
I think I addressed this somewhere already. There are excel tools for modifying response files to account for baffle step. I use two different ones myself.

 

[AudiocRaver]

Thanks, Matt, sounds like you have it covered.