A few comments on prevous posts and some speculations:
From the EMC2 site pictures, I would guess the WB-D demensions as ~ 6 meters in diameter, assuming the man standing beside it is a little under 2 meters tall. IF the magrid size is ~ 1/2 the diameter of the chamber, this would be consistant with predictions for scaling (1.5 meter radius for breakeven). I speculate that the WB-D designation may suggest a new naming scheme. The D may refer to duterium fuel. Othe fuel combinations may use designations like -H for helium3, -B for boron 11, possibly even -T for a fallback deuterium- tritium reactor.
The coils will cool slowly in a vacuum by radiation alone. Compared to WB6, the magrid still has to be supported by outside posts. If each grid is supported by its own posts (like the drawing in the paper modeling the Polywell last fall) instead of one set of posts supporting the entire nub connected magrid, there may be even more area for conductive cooling. Also, the size/ surface area of the coils themselfs contribute Then there are conciderations about the number of windings per coil, wire gauge, thermal mass, post material/ thermal conductivity, etc. The posts in WB 6 were presumably ceramic, so any conductive cooling was probably mostly from the contained feedthrough wires. These could probably be larger and more numerous (a set for each coil) in WB8. So the cooldown time between shots could be similar to WB6 (or even shorter) despite the increased magnetic fields generated.
I had assumed that WB8 would be liquid nitrogen cooled because the targeted field strength (8X) was consistant with the electrical conductivity gain foe copper wire at that temperature. Certainly the brute force approach of more and/ or thicker windings with increased current in possibly larger coils could do this on it's own.
As for test duration, I understand that in WB6 the magnetic field was first turned on , then the gas puffers and electron guns were fired. The magnetic current may have been run for over a second (I seem to recall several seconds being mentioned) for the pratical 1/4 millisecond tests.. If the subsequent setups had a tighter control on the timing of the various parameters, similar or perhaps longer tests might be doable, even with shorter magnet current times (with corresponding heating).
To error is human... and I'm very human.