the performance is pretty good, but not in line with what the specs (R50) might suggest. I have Aerogel panels from Glacier Bay on 4 of 6 sides, the remaining two have 6 or 8 inches of extruded polystyrene foam. Outside of the Aerogel panels is at least 1 inch of PS foam. When we built the box, I put thermocouples between each of the Aerogel panels and the PS foam, leading out to an access panel. This was done to measure the efficiency, but also to diagnose any panel which might have lost the vacuum.

Since the Aerogel is in theory R50, and the PS foam around R7, you would expect to see most of the temperature rise between the interior and the outside surface of the Aerogel, with a small additional rise across the PS foam. However the difference in all panels is less than predicted. It could be that the Aerogel is not performing to R50, or it could be that the edge effects are dominating the insulation. The R50 number assumes a continuous surface, but of course these panels are finite and have edges. The edges on Glacier Bay panels are done with 1/4 inch plastic, R value about 2. If you calculate the R value of the panel including edge effects, you find that you can ignore the center of the panel (R50 can be assumed to be infinite) because nearly all the heat gain is through the edges. And that reduces the effective insulation by a significant amount.

The refrigerator is still quite efficient, about 32 hours holdover followed by a 40 minute run time if you don't open the box much. The freezer consumes more, only about 12 hours holdover (was predicted to be 24) and about 40 minutes run time (was predicted to be 80). I don't quite understand what is happening in the freezer, it behaves as though the plate has only half the capacity, and I have never been able to figure out why.

Another very large factor is how much stuff you put in the fridge. This was called to my attention by Glacier Bay in the design process, and they turned out to be spot on. They said calculate the heat loss of the box, then multiply by 3 for typical use. This is due to opening the box, and putting warm drinks inside. If you calculate the heat added with warm drinks, it is quite significant in an efficient fridge.

Bottom line is that Aerogel panels are not cheap, do not perform as well as the specs in the real world - but do only take 1 inch of space out of the box.