TMS Meeting Minutes, July 23, 2020
Attendees: Andrew, Breann, Heejoo, John, Trenton, Yang (Zoom meeting)
Meeting Discussion Summary:
- Heejoo and Yang discussed the results from the TCP test from Tuesday, July 21, completed with with the engineering team from eTalon.
- By changing the configuration file prior to measurement, all channels may be obtained simultaneously over TCP.
- The eTalon engineers also presented a demo for shutter control, but the power meter was turned off, so we were unable to check the shutter performance. Six to eight months ago, the shutter was tested with a long exposure using LUCI and no light leakage was found.
- eTalon engineers asked what functionality we want them to develop, and they stated that they will put out a new version at the end of this year or the start of next year. We would like to overcome the problem of having to restart the program every time a process fails.
- Yang plans to continue writing code to enable distance measurements to LBC and to complete the Jacobian calculation with the Ciddor correction.
- Andrew would like to compare three data streams: raw data, eTalon corrected data, and data with our own Ciddor correction applied.
- Trenton completed analysis on the error in the Ciddor index calculation and determined the difference from the NIST standard is no larger than 1EE-08 (over a 10 meter distance, this is 100 nm/K).
- Andrew insists that this amount of error is negligible for our purposes, and that we need not worry about these deviations.
- Yang suggests sending data to Trenton to check the MATLAB implementation against the Python implementation.
- When comparing the table in the Ciddor paper to the NIST calculations (see appendices, https://emtoolbox.nist.gov/Wavelength/Documentation.asp#AppendixA) there is some difference in the calculations which Trenton took into consideration when writing his MATLAB code.
- We agree to use the NIST website as the gold standard. (https://emtoolbox.nist.gov/Wavelength/Ciddor.asp)
- Yang and Trenton will compare the Python and MATLAB implementations.
- Three data sets were collected over the past week.
- Olga took on-sky data of a star with LBCr and Heejoo recorded all TMS data (telescope pose, elevation, and angle) while Olga tried to find collimation and pointing.
- John and Heejoo took TMS data over elevation change from zenith with 10 degree steps (zenith to 30 degrees then to zenith). The same data was then taken with the [collimation] model applied.
- The collimation model is necessary because, over elevation, the LBC hub rotates. The model provides an aberration-neutral rotation, establishing the repeatable rotation of the hub and best centering the LBC. The range-balancing function built in to the PSF subsystem produces the model by calculating the best rotation to maximize the available decenter range.
- Sit and stare data was also taken.
- Summer shutdown planning was discussed and a few proposals were put forth.
- Andrew proposes the use of a Zerodur rod (provided by Chang Jin Oh) with two flat mirrors to make an eTalon for the eTalon, boucning a laser back and forth for to find the eTalon error plus the air error. The goal is to eliminate the effect of temperature-dependent structural steel changes and isolate the effect of air temperature. Andrew suggests using wireless transducers (0.1 K sensitive) used by LBTO. He suggests placing this setup in the aux control room.
- We propose a sit-and-stare measurement at zenith over a diurnal cycle to characterize the air and structural temperatures and determine which telemetry to use from the different thermocouples. It is critical for us to determine which telemetry channels are useful for TMS and which are not.
- Yang will continue to work on developing fully functional software for TMS during summer shutdown.
- We would like to install TMS on the SX side, if at all possible. For this effort, we must develop a proposal detailing the necessary tasks and who is needed to complete the tasks and see if we are able to get to work on this. It has been suggested that Mike Gardner could complete the laser tracker survey, if available.
- At some point, we would like to fix the problem with the fiber splitter box so that we can make use of the 12-channel multi-core fiber. We will need to follow up with the eTalon engineers to solve this issue.
- Yang will send Trenton data to compare the results of the MATLAB implementation of the Ciddor calculation to the Python implementation.
- Andrew will write a proposal for the Zerodur measurement setup.
- Trenton will assist Heejoo with data processing for last week's data.