i built the whole custom box with Glacier Bay vacuum panels. When you do the calculations, you can pretty much ignore the heat gain through the panel itself - but at its edges, the R value is low, and the end up being a lot of edges. In addition to the vacuum panels, there is at least 1" of extruded PS foam outside of that, and between the two on each panel, a thermistor run to a central location. This lets me measure the temperature between the vacuum panel and foam. It is typically running in the low 60s on a mid 70s day. That leads one to believe the R50 figure is not being realized in practice, or the edge effects are even worse than I thought.

That said, using Glacier Bay's now defunct eutectic plates, we have about 30-40 hour hold over on the refrigerator, and about 15-20 on the freezer if no one is opening it or putting warm stuff in. With guests aboard an hot weather, a lot of drinks are going in and out of the frig and the door openings and heat gain from the warm drinks cuts the holdover to less than 24 hours.

If I did it again, I would try to make more room for the box and just use 4 or 6 inches of XPS foam. In the lid, given the many edges (edge of door, edge of frame, edge of top surface) the R50 is going to end up being more like R15 or 20 at best, average.