A mark V helmet and the suit comprise a closed compressible volume that is conectcted to the tender hose (no nice rubber dam around your neck). If we assume an incompetent check valve, and a tender hose which is cut above the surface of the water, a pressure gardient will develop across the surface of this closed compressible system equal 1atm X Depth in feet/33. Imagine you have a weather balloon full of toothpaste with a seven mile long non-compressible tube sticking out of it. Now submerge that balloon seven miles under the surface untill just the end of the tube is sticking out of the water. The toothpaste will be extruded up the rigid tube to equalize the pressure gradient. In fact, if the balloon is flexible enough, it will be pushed up the tube untill it ruptures. Just like when you stick a straw in a glass of water, the water tends to fill the straw up to the water line. Now back to our Mark V diver... If his suit does not rupture, he will extrude into the tube or the tube will become blocked. In essence, we have an old fashioned vertical column barometer, but instead of mercury, we have diver sucking up into the column. Any failure of the suits watertight integrity will just cause a flooding of water, which if fast enough would prevent compression of the suit into the helmet. If the cut was below the surface, water would rush in to equalize the pressures, so no compression of the suit/helmet volume. The inner diameter of the tube has no effect other than the speed at which the equaliztion can occur and the ease of getting blocked or extrusion. It would not matter if the helmet and suit were full of water IF the hose were full of air. If the hose is full of water no compression of the suit will take place (it would be just like the diver disconnecting the hose from his rig). The main question would be, how much pressure it would take to overcome the mechanical resistace and viscosity of the divers tissues resisting being forced into the helmet and tube like so much Play-Doh through a Fun-Factory.