HFSS HFSS-IE convergence problem.

The attached HFSS file is for a half-wave dipole for 14.2 MHz. The dipole is only just over 1 m long, but is emmersed in water, so the relative permttivity is around 91, so it should resonate around 14.2 MHz or so, though perhaps a bit higher, as there is not a huge amount of water.

The problem is, the mesh will not converge. It slowly converges after 5 or itterations, but then |S| diverges.

Can anyone suggeest how one might get around this?

Note this uses the integral equation solver (HFSS-IE), so one would need an HFSS-IE license to run it.

I've not tried running this on the normal method in HFSS, as I suspect it would take ages due to the size of the airbox needed.

I believe the MoM, which is what HFSS-IE uses, requires a "ficticious" surface current to model objects which are not fully conductive or insulators. Perhaps the HFSS implementation is not too good, or perhaps the method just breaks down, but I'm suspecting that might be an issue here.