I made a 3x3 turn coil which also triggered the overcurrent limit on the MGNZ board.
I then swapped the power supply to the 1200w one, and upped the current limit to 19A.
It now idles at about 11.1A, and draws around 16.8A max with a case.
Now it's causing the Arduino to intermittently reset (sigh) so I might try running that off a separate power supply
View attachment 1622724
How are you going with this?
I'm finalising my build currently with some improvements.
I had similar issues with higher current than I'd like and trying different coils with longer than I'd like times (full 8sec on the MGNZ setup might JUST get my 6.5 brass a dull red in the dark).
I think I finally found the goldilocks coil for my setup.
Sorry to those who might have seen this on another forum. I've been reading everything I can on these machines.

Specs:
3mm copper, insulation pre-installed (yes I have fibres in my fingers, thanks for asking) and wound around a 25mm mandrel (3D printed), so probably ~26mm ID.
2 Layers
Inside Layer - 4 turns
Outside Layer - 3.5 turns
I did make a couple of other changes at the same time. Switched from 12AWG to 10AWG powering the ZVS. I don't think this made any difference because the 10AWG did not get warm at all.
The one thing that may have made a difference is that I had the coil legs going through a metal case (holes drilled big) with a plastic cover (with smaller holes to stop shorting) over it. The metal may have been interfering with the induction. I've since cut a large hole that I will cover with a 3D printed piece.
My quick test showed this one giving a dull glow between 3-4secs with the ZVS unit sitting separately (not in the case of the machine) and no water cooling connected.
I'll post more info once it's all back together with the new parts.
Edit: I'm still using the 650w 48v power supply. When it's back together I'll try a decent run of cases to see if it keeps up.
I do have a 48V 2000W power supply on the way (which is WAY more expensive), but I'll only use it if I have to.