Henrik Ronnow:
I assume the UB-stuff is with the intention that one can eventually drive out of the AB-plane using the goniometers? For some experiments - e.g. the incommensurate stuff Oksana, Sabine and I are working on, this would be extremely useful.
This is implemented with the Q3 variable. The only difference to the QH variable is, that it does really move out of the scattering plane, whereas a drive on QH gives an error when a vector outside the scattering plane is given. What remains open, is, how to resolve the redundancy on drives outside the scattering plane. For EN=0, you have 4 angles to drive, but only 3 components of the Q vector. We have to specify, how to resolve this redunant information. One possibility would be, that the vector in the AB-plane, which is perpendicular to the scattering vector, stays in the scattering plane.
"Henrik:" One solution to redundancy problem would be to choose solution that has minimum total tilt - sqrt(Gu^2+Gl^2) would do.
Henrik Ronnow:
Related to this is a command to tilt the current scattering vector. Normally we always rotate the sample or the cryostat to have the crystallographic axes along the goniometer axes. But this is silly since instead of scanning sgu and sgl independently one can just make a coupled scan that exactly tilts the current scattering vector. Much of the time, this is a small convenience thing, but sometimes, the sample does not have refelctions at perpendicular angles, in which case such a feature would allow more accurate alignment.
Me also, I thought about a virtual motor simulating a goniometer with arbitrary axis. I think about a syntax to implement this.
Henrik Ronnow:
The other thing we need to think about is how to deal with the multiplexing. For RITA-II its only 7-9 channels, but for EIGER-CAMERA it will be 60 degree continuous scattering and 5 of those. We obviously do not need to sort that out in detail before when CAMREA is closer to ready, but we should make sure that the syntax and underlying code can be generalised to this kind of analyser-detector systems.
I did not specify such things. The aim is, to have the same syntax for all stuff common on the 3 instruments. I have also not specified in detail the polarisation analysis, which is only used on TASP. But if something is found in the syntax, which might give trouble when extended to some instrument specific stuff, then we like to have a feedback.
Bertrand Roessli: thanks Christoph to have sent me the link. The following is just the result of some suggestions that I can think of after discussion with Christof and Markus. Sorry if they are redudant.
-It is written that En is set to zero in case the command 'Sc Qh h k l' does not contain the energy transfer. I think this is quite dangerous if somebody forgets to do it and the instrument drives back automatically to the elastic position.
-It seems there will be no switches in future apparently. Please remember that TASP has polarisation analysis.
-The option with the UB matrix allows scans out of the scattering plane. Some security should be installed to prevent that a typo in a scan or in a job drives the goniometers (and the cryostat that is on it). I am not sure that setting the soft limits is a good option. I like Markus' proposition with the 'sc/dr Q3' command very much.