This one is hard. Not only just the shake, but also, the internal stress and warp of the model itself. Using better filament would also help. Try it with a CF filament and I guess it would be fine. Or maybe try print with a higher chamber temperature.
E.g., a large vertical hole on a vertical part with small contact, the top part overhang can be warping upwards a LOT and cause a nozzle knock, or too high nozzle pressure when printing over here. That pressure might push your part falling.
Then, here, a floating overhang that is completely floating on top of nothing, and it’s too big.
I don’t even think this would print too well here. With overhangs like this, no wonder why everybody is blaming the model maker. Even if you have perfect bed adhesion, warps on overhang would make it easily break off just because of the nozzle pressure.
A huge difference on H2D against other printers is that it’s using a… servo motor for extruder. This means that when it faces higher than normal backpressure (e.g., warp blocked the nozzle), this extruder will push harder to make sure it can reach the extrusion assigned from the GCODE. Meanwhile, for other nozzles, if there’s a warp that blocked the nozzle slightly, the extruder would not increase its power, it usually end up with a skip instead.
This difference would lead to less “failure”, but a more pronounced under extrusion after overhang corners.
You should probably show us the photos, especially where it starts to fail.
Also, get a large USB drive and enable auto monitor video recording (not timelapse, but a full recording). So that you can see the exact moment of what happened when printing.
One more thing you can try is to retension your bed. Default tension on the bed might be too low, causing the plate surface to shake around slightly with the movements.
I first noticed this issue when I was printing a building model, with some pillars that is shaped like a T shape. Without proper support, the dash - above the pillar would warp upwards, and eventually even though Z hop is working fine, the nozzle would eventually press the warped T part down hard enough, and cause it to break.