When activating the crab cavities, the calculation of the sigma matrix (SIG11 - SIG66) seems to be wrong,
e.g. beta_x * emittance_x != sig_11 , for small (i.e. LHC-like ex<1e-5, as compared to ex=1 - the latter seems ok) emittances.
It might be because activating the crab cavities causes twiss to run in 6D,
while the twiss for which this works runs in 4D.
It should be tested if "simply" switching the twiss to 6D also throws the values off.
Thanks to @tpersson who helped me investigate issue.
When activating the crab cavities, the calculation of the sigma matrix (SIG11 - SIG66) seems to be wrong, e.g. beta_x * emittance_x != sig_11 , for small (i.e. LHC-like ex<1e-5, as compared to ex=1 - the latter seems ok) emittances.
It might be because activating the crab cavities causes twiss to run in 6D, while the twiss for which this works runs in 4D. It should be tested if "simply" switching the twiss to 6D also throws the values off.
Thanks to @tpersson who helped me investigate issue.