NOTE: These instructions are based on OLD python based GUIs.

1. After the daytime staff have aligned the vis/IR dichroics and IR Mirrors to the MIRCx table, you can begin the lab alignment of MIRCx. This should be done at the start of a MIRCx run or whenever one of the MIRCx pickoff mirrors are taken out of the beam. Make sure the CHARA corner cubes are in all of the beams.

2. Confirm that the MIRCx pickoff mirrors are on the kinematic mounts. Also make sure that the IR alignment target on the edge of the MIRCx table is out of the beam. Take off the lens cover on the MIRCx camera.

3. Turn on the white light source (POS 4).

4. Follow the instructions in the MIRCx User Manual to start the mircx server and open the GUIs for mircx_ircam, mircx_parameters, and mircx_fiberexplorer.

5. Follow the instructions in the MIRCx User Manual to start the camera (Gain 5), take backgrounds, update/send the fiber positions using mircx_parameters, and initialize mircx_fiberexplorer.

6. Open B1-6, IR1-6, and WL shutters. Select visbeams 1&2, 3&4, or 5&6.

7. On Fiber Explorer, go to the “Fiber Mapping” tab. For a good first search for the white light source, set num_steps = 15 and XY step = 8 um. Select the check boxes only for the two visbeams for which you are currently searching. Click “Perform Map (xchan)” to begin the fiber explorer search for the two beams.

If you find the light on one or more beams, then click “Move to max” or “Move to Gauss peak” for the beams that have light. Do not be too alarmed if the light does not show up after moving the fiber - the 8 um step size is not too sensitive for proper centering. After moving the fibers, do a finer search using num_steps of 9 at a step size of 2-4 um to center the fibers better.

If you do not find the light, you might want to use a larger number of steps (like ~ 25 or so).

If you see light on the detector while searching, but not in the fiber maps, then this means that the xchan positions are not set correctly. Try closing all but one shutter and search for light on only one beam at a time using the “Perform Map” button. This reads out the data quadrant rather than the xchan quadrant. After you find the light you will need to read out the proper xchan location from the cross section profile (change column numbers to zoom in). Update the xchan positions (all beams should be shifted by same amount) using the MIRCx parameters gui (remember to Click “Send Values to RT”) and then click “Load xchan pos from mirkwood” button on Fiber Explorer.

Also try looking for light on a different visbeams pair, just to confirm whether there is light coming through at all. If not you don't see light on any beam, then check to make sure all the shutters are open and do a visual check in the lab to make sure corner cubes are in and no targets are blocking the beam path.

If you do not find light on a particular beam (but confirm that light is getting though to the other beams), you can try searching the entire range of the fiber mount. To do this, go to the Luminos Stage tab on Fiber Explorer. Select the relevant beam from the pull down menu. Write down the current x, y, z positions. Set the x,y position to the center of the range (64,000, 64,000) and click “Move to” button (full range goes from 0 to 128,000). Do not change the Z focus position. Then go to the Fiber Mapping tab and set the search range to the full range to num_steps = 64 at 8 micron step sizes. This search will take a very long time. If there is still no light, then a careful check has to be done by eye using the laser on the MIRCx table - EXPERTS ONLY - do not move any optics if you are not certain what they do. Contact the MIRCx team for advice.

8. The last step is to record the counts (“Get Intensity” button on Fiber Explorer) and approximate ratio between data section and xchan section for each beam one at a time. The current ratios for xchan/data are 2-3 (rows 15 and 45 in cross-section view). Check to make sure the fringe data light and the xchan light show up in the default quadrants. If there is a shift (e.g., the light does not cover the full number of pixels in the quadrant view mode), then the output stages might need to be adjusted - please talk with a MIRCx EXPERT before making any adjustments.