I can’t be sure about the reflection but ain’t the reflection be only applicable to bass frequency and the mid to high frequency be more directional?
The primary cancellation will occur at approximately the frequency of wavelength 4x the distance from the front of the baffle to the wall (i.e. because at this frequency the reflection will be perfectly out of phase with the direct sound).
4 * 14.5cm = 58cm, which is the wavelength of frequency of approx. 600Hz.
A flat baffle will control directivity down to approximately 140/
w (in Hz, where
w = baffle width in metres).*
As you can see, as baffle width approaches 1.5 * baffle depth, directivity control approaches the frequency of the primary cancellation.
Following from this, in order to properly mitigate this cancellation, baffle width should be >1.5 * baffle depth.
In the case of this ML, baffle width and depth are approximately equal. You can see from the polar response plots that the baffle provides some degree of directivity control down to 600Hz, but not quite enough to prevent significant cancellation from occurring. If the baffle were 50% wider, or the box 33% shallower, directivity control would be (close to) sufficient to prevent significant cancellation.
*This is actually a special case of Keele's formula for rectangular CD horns, which states that: F = K/(a*w), where K = 25,306, a = coverage angle (degrees), and w = baffle width (metres).
