It was my mistake in the original post to imply I was going to coat the entire piece in epoxy. If I will use epoxy at all it will be to glue the panel assemblies to the plywood backing. Only the rails and stiles will be glued, the individual panels will be allowed to float.
Thanks for your input. Let me explain my design choices. To your first concern with the front design elements lining up with the side panel elements, I can see why most people would immediately favor that. It is what seems to make sense. You and others concerned with with this aspect of my design need to understand that my intention in this project is to achieve as best as possible the illusion that this is an 18th or early 19th century Spanish Colonial armoire. In researching those pieces I began to notice that side panel designs were often either very simple compared to the front design or they were a repetition of the front EXCEPT they did not mirror the design in alignment. Quite often the design did not take into account the fact that on the door facade there are in effect two top rails, those being the structural rail of the carcass and the top rail of the doors. They simply used the same design elements and stretched them the over the entire height of the piece. See below. Concerning you final comment...I'm not sure why you are advocating flipping the side panel design horizontally. As for my reasoning, I prefer not to have a similar panels adjacent to each other in respect to the association of the side panel assembly where it meets the front panel assemblies. In what I understand you are suggesting, when viewing the enclosure from an angle, you would be viewing both the side and end panels. In the case of the bottom row of panels and moving from the rear around to the front that would yield a pattern of One Tall panel, then two stacked panels, then (moving around the corner) two more stacked panels then one tall and so on. I prefer to maintain the pattern of alternating arrangements. To each his own!
Attachment 371248Attachment 371249Attachment 371250