Assuming that a subwoofer is located in an open area on the ground in at the center point of a flat wall that's shaped like a semi-circle of radius R, at what frequency does the boundary effect become appreciable? Is there any way to sim this effect using HornResp, Akabak or other readily available software tool?
If you assume the subwoofer is on the same plane as the wall (baffle flush with the wall) it's a simple diffraction calculation. The Edge or Bagby's Baffle Diffraction and Boundary Simulator can sim this easily (Bagby's can only do rectangular baffles though). Save the diffraction profile as a .frd and sum it with the subwoofer's simulated response curve also saved as .frd. Akabak can also sim diffraction.
If the subwoofer is any great distance in front of the wall you have to worry about Allison effect bounce back cancellation (similar effect as floor bounce), this is going to be a bit more tricky to simulate (I'm not sure how you would do it for a finite sized wall - Bagby's will simulate distance from a rear wall but it assumes the rear wall is infinite size) but you can calculate the distance from the baffle to the wall and easily calculate and expect a notch at that frequency.
A couple of years ago people were using a very short (0.01 cm) final segment in Hornresp with a large area to simulate large baffle faces, which is a kludged attempt to simulate diffraction effects of a large baffle face. But I'm pretty sure this resulted in garbage sims as Hornresp was never intended to simulate diffraction.
Much better to simulate a large baffle or boundary wall as a diffraction event with a real diffraction simulator.
If the subwoofer is any great distance in front of the wall you have to worry about Allison effect bounce back cancellation (similar effect as floor bounce), this is going to be a bit more tricky to simulate (I'm not sure how you would do it for a finite sized wall - Bagby's will simulate distance from a rear wall but it assumes the rear wall is infinite size) but you can calculate the distance from the baffle to the wall and easily calculate and expect a notch at that frequency.
A couple of years ago people were using a very short (0.01 cm) final segment in Hornresp with a large area to simulate large baffle faces, which is a kludged attempt to simulate diffraction effects of a large baffle face. But I'm pretty sure this resulted in garbage sims as Hornresp was never intended to simulate diffraction.
Much better to simulate a large baffle or boundary wall as a diffraction event with a real diffraction simulator.
Last edited:
Good article, even on lead walls! The math is simple: 1/(Distance Between Source To Wall times Pi)=Dip Frequency, at which ruffly -6dB gain due to destructive interference occurs.
If you put two drivers side by side you should get the same as with a single driver alongside a reflective wall. Sound that would come from the imaginary woofer through where the wall is, would be directed the same (but mirrored) way onto the wall by the single woofer and reflected back along this same path. The effect is clean within 1/4 wl.
When there is talk about wall reinforcement of a horn beside a wall, it is suggested that the result is like a horn of twice the throat, mouth and driver dimensions, but of course it is more like two horns side by side. This is near the same thing only at low frequencies.
When there is talk about wall reinforcement of a horn beside a wall, it is suggested that the result is like a horn of twice the throat, mouth and driver dimensions, but of course it is more like two horns side by side. This is near the same thing only at low frequencies.
Ok, based on the responses so far, perhaps I have to give a bit more information. I've included a diagram that shows the situation I'm referring to. All I want to find out is the point at which the boundary effect would become appreciable. I'm assuming that this is directly influenced by the radius of the wall (R in the picture).
Yes, I know this is theoretical, and IRL there will likely be losses, but at least the theoretical answer gives a starting point.
...and yes, there's something driving this question
Attachments
Yes, of course the size of the wall directly influences the size of the waves that bounce off it and the ones that are large enough to go around it.
Like I said before, a baffle diffraction simulator will give you the answers you need in about 5 minutes flat. "The Edge" program can simulate the round wall (or any shape) and Bagby's Diffraction simulator can only do rectangular.
The distance of the sub to the wall is negligible in the picture so there will be an Allison effect null but it will be fairly high in frequency, nothing to worry about.
It has to be a pretty big wall to affect subwoofer frequencies much.
Here's a quick sim with Bagby's spreadsheet. Wall is 120 x 120 inches (rectangular), sound source is 12 inches (round) at the bottom middle of the wall and assumed flush with the wall (although it won't make much difference below 100 hz if the sound source is in front of the wall 12 inches). Diffraction modeled as a gain (not a loss). Observation point makes a difference too, arbitrarily set at 20 ft mic distance from sound source.
This is JUST the diffraction profile, it has nothing to do with the frequency response of the sound source, but you can sum this with the subwoofer's frequency response.
Here's a quick sim with Bagby's spreadsheet. Wall is 120 x 120 inches (rectangular), sound source is 12 inches (round) at the bottom middle of the wall and assumed flush with the wall (although it won't make much difference below 100 hz if the sound source is in front of the wall 12 inches). Diffraction modeled as a gain (not a loss). Observation point makes a difference too, arbitrarily set at 20 ft mic distance from sound source.
This is JUST the diffraction profile, it has nothing to do with the frequency response of the sound source, but you can sum this with the subwoofer's frequency response.
An externally hosted image should be here but it was not working when we last tested it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.