PECVD: Microwave Plasma convergence?

Hello,

I am still fighting for my models to converge, where do I have to look for the trouble?

The start is the 2D-waveguide model from the Library. I just inserted an extra substrate into the plasma with some bias voltage, like we do for PECVD processes.

whenever this substrate gets too big, or is located too close to the microwave window, my model will stop..

I need to know why, so that I can look for a solution? What do I need to follow? (the model took 10 hours to stop with convergence error)(if the substrate is put more to the downstream region, it will do the job within 2 hours or less)

is there a problem of hf-wave scattering with the overdense plasma? is the physical description not good enough? but this should already be a problem without the substrate, but it is not.. (the library model starts from overdense initial values, I start from low values ne:10e14)
but:
why are the waves in the library model not expulsed from the plasma region? in my plasmaline model (model exchange) they do? something to do with TE or TM modes?

Or is the problem simpler: whenever my DC-current increases strongly the model would stop.. is it only a "power source" problem? How do I make the right coupling between the circuit and the plasma?
( I want to be able to pulse my DC bias later..)

Yes, many questions..

Thanks for your help,

Lukas
Nagoya-u