Looking forward to giving a listen tonight..
I did an experiment with a mic arrangement used for a time by Onno Scholze at Phillips. DPA 4006A with only 32 cm capsule spacing, angled out 90 deg. On paper, this shouldn't work, [..snip]
Because on paper the spacing is overly narrow? I suspect it will produce some binaural cues and have a solid center but not sound as open or wide as it would with increased spacing. At that spacing a center baffle is likely to be helpful. The 90° inclusive angle should help provide a bit more stereo interest and openness up top.
I just modeled it in Image Assistant. SRA is somewhat strongly dependent on distance from the source. If placed relatively close, ~2m away, total SRA is 119° (75% SRA = 71°). If relatively far ~10m away, total SRA is 143° (75% SRA = 79°). A difference in SRA of 24°. Perhaps of greater importance than image distribution, DFC (Diffuse field correlation, lower is better) remains high (>0.9) up to around 100hz, then drops off between there and 500Hz where it reaches zero, remaining fully decorrelated above that point.
Not sure if I'll do this again or go back to my usual starting point of 42 to 52 cm spacing, straight ahead or with a slight opening angle.
Modeled that too for comparison- At 42cm, SRA is less strongly dependent on distance from the source than at 32cm. If placed relatively close, ~2m away, total SRA is 86° (75% SRA = 54°). If relatively far ~10m away, total SRA is 93° (75% SRA = 58°). A difference of 7°. DFC improves slightly, remaining highly correlated (>0.9) up to around 70hz, then drops off between there and 400Hz where it reaches zero and remains fully decorrelated above.
And at 52cm with the array ~2m from the source, total SRA is 67° (75% SRA = 44°). 10m away, total SRA is 72° (75% SRA = 46°). A difference of just 5°. DFC is improved somewhat further, remaining high (>0.9) up to around 60hz, dropping off between there and 200Hz where it reaches zero, remaining fully decorrelated above.
I don't think any of those configs are wide enough to have any problem with hole in the middle. Preference for one over the other is likely to come down to 3 things: Image distribution (SRA), the portrayal of reverberance (DFC), and possibly, the alteration of frequency response ripple in the upper bass from the constructive/destructive interaction of wavelength as it relates to the spacing between microphones.
The DFC corner frequencies are somewhat high for all of them IMO, but that can only be improved further by increased spacing. I suspect I would likely prefer your wider 52 cm spacing because of the difference in reverberant portrayal, unless weaker voices in the center of the ensemble needed help from the wider SRA, image-distribution-wise.