Dear ET folks,
We (the Illinois Relativity group) have been following the ET mailing list with great interest since its inception and were planning to send representatives to the upcoming ET workshop. Sadly, we will be too busy preparing for the Midwest Relativity Meeting, which we are hosting the following weekend.
We noticed that there is some interest in implementing a vector potential-based GRMHD algorithm in ET. Were we able to attend the upcoming ET workshop to discuss this topic with you, we would have shared some recent improvements to our vector potential-based GRMHD algorithm, in the hope that the ET community might benefit.
As many of you know, we currently evolve the vector potential to perform GRMHD simulations with divergence-free magnetic fields. We found some major problems with our original implementation when evolving magnetized binary inspirals, such as BHNS inspirals. We discovered that the problems were due to a zero-speed EM gauge mode, and after switching to another (Lorenz) EM gauge, we find huge improvements.
We recently posted a paper summarizing our findings at: http://arxiv.org/abs/1110.4633
All the best,
Zach Etienne Yuk Tung Liu Vasilis Paschalidis Stu Shapiro