Since I'm looking at a problem at a listening distance anyway, I played around with microphone placement. Following plot shows the quarter round rectangular symmetric baffle on a RD75. The yellow curve was with the microphone at listening height "standing" up. The white curve was with the microphone dropped 6 inches down. The distance between horizontal lines is again 1.5 dB.
Wow, lowering the microphone has more side-effects then changing baffle shape ;-). Let's take a look what happens when we compare symmetric trapezoidal with rectangular baffle under these circumstances. The white plot is the same in both graphs. The yellow plot below belongs to the same driver in the same baffle given a trapezoidal shape by exchanging a 3 inch wedge. Except for the dip at 1.5 kHz the yellow plot has a more desired characteristic: peaks and dips are less pronounced in magnitude.
Don't get confused with the large dip at 200 Hz. Room interaction at 5 meter distance of my driver. No trace of this at 3 meter distance as shown in plots taken during previous weeks. This dip has nothing to do with the drivers nor baffles, the Carver plots show the same dip. More work to clean up the listening space. (Hi honey, I just ordered the lastest fashion Sono-tec wallpaper...)