Testing

SetUp Simulated Device

There are some simulated device setups in the private share ...\Reflectometry\configs\<instrument>. To use these:

  1. Copy galil directory to your configuration directory

  2. Created a configuration with the correct number of simulated motors

    • some configs have components in that help with that. Your server may need a restart to pick them up after copying them over to your machine

  3. Switch to config so that the motors are running

  4. Run the setup_motors.bat in a epics terminal on your machine

  5. Copy the config.py to your refl directory in your config directory

  6. Restart the refl ioc

This should make the instrument act like a simulated version of the chosen instrument.

If you want to swap between multiple configurations you can also create a package (includes __init__.py) for each instrument and then use the config.py in the base directory to swap between them.

CRISP in Feb 2019.

Setup

Scan in sample and align beam in SECI.

Copy scan.py from LOQ to CRISP to allow in IBEX plotting. Sort out a crisp package and alter the init so that it loads it.

Issue I can not see how to reload my changed code while developing my scan except to restart the scripting window (which is non-ideal).

Issue The fitting does not work at the moment so graphs will not display correct fits.

Issue The errors do not work correctly they are just square root of value

Ideally, this code should be in crisp_inst.py but I haven’t put it there yet this is to load in the scanning library:

from instrument.crisp.scans import scan

Theta scan

After the final align of the beam you should perform a theta scan to ensure that the beam is aligned.

scan("THETA", -0.05, 0.05, count=21, frames=100, fit=Gaussian)

which produces Theta scan showing a peak at 0.0.

Sample alignment

After aligning with the laser:

  1. Perform a shadow scan (this is only possible if your sample absorbs/scatters neutrons silicon does not). I could not do this with the sample I has.

  2. Perform a shallow bounce scan. Set theta to a small angle (0.25) then tilt phi to find the peak in intensity. Start with reasonably wide slits otherwise you will not see anything. NB that the offset in CRISP defined in “Define offsets” is applied before sending it to the motor so when comparing to CRISP don’t forget to manipulate this correctly.

    1. scan("PHI", 0.21, 0.285, count=21, frames=100, fit=Gaussian)

    2. Phi scan showing a peak at 0.250.

    3. Move the table to be at the peak of this scan (this should be done by fitting)

  3. Perform a sample height scan, this should peak at the height and be symmetric.

    1. scan("HEIGHT", 0.22, 0.28, count=21, frames=200)

    2. Move the table to be at the height given by the peak in this scan

  4. Now repeat until happy that the results are not changing.

  5. Set the offset so that the user value is 0

  6. Now perform a phi scan (on CRISP with the silicon block you don’t see anything)

Troubleshooting

Make sure the spectrum 1 look like Make sure the spectrum 2 look like Large spike at low time of flight is prompt spike it should not dominate spectra 1 or 3.

Do a theta scan without the sample or super mirror in the way. Does it look correct if not stop and think.

CRISP in March 2019.

Did many scans comparing SECI and IBEX see ISIS_Experiment_Controls\Reflectometry\Ticket3462.

CRISP Blind Test