You're throwing two phenomena together here.
First, we have the baffle step "6 dB loss" at the transition from 2pi room (sound is reflected by the baffle) to 4pi room (wavelength of the sound is so large that the baffle is negligible in comparison). This is a more or less smooth transition in the frequency response. The baffle step does not cause resonance.
Without other disturbing effects this can be shown with a spherical enclosure.
https://trueaudio.com/st_diff1.htm
View attachment 60327
Then there is an effect of the secondary sound sources created by the edges of the cabinet, which interferes with the original signal.
https://heissmann-acoustics.de/kantendiffraktion-sekundaerschallquellen-treiberanordnun/
View attachment 60329
Here, the path from the sound source center (chassis center) to the edge of the cabinet and the distance to the listener play a role. This interference leads to an rippled frequency response.
View attachment 60330View attachment 60333
What is the effect of combining both effects for the Buchardt bass driver? This results in the curve shown in
Post#68 (here only for the woofer):
View attachment 60336
The lateral edge reflections are too high in the frequency range for the 550 Hz resonance of the Buchardt. Only the upper edge would theoretically be a possible cause but it hardly contributes in the simulation and the Q of the resonance would be much lower.
Edge diffraction has only a minor influence overall and cannot lead to such a pronounced high Q resonance@550Hz shown in Amirs measurement of the Buchardt.
I know that could have been said in just one sentence