TMS Meeting Minutes, December 16, 2021

Attendees: Andrew, Breann, Christian, Heejoo, John, Olga, Trenton, Yang (Zoom Meeting)

TMS Operation:
  • Commissioning night on 20211213
    • There was some confusion on the quality of the data
      • The data is fine, there are NaN's shown on the Sx side b/c TMS was not used on this side
      • The pose data on the Dx side looks just fine, so we are all good, and TMS appears to be working properly
      • There were lots of "none" data points from the EAMT (not sure if the error came from the machine or from the software
      • One side used active TMS and the other side used hybrid mode, with FPIA controlling collimation
    • The temperature change was rather interesting on this night, with one mirror cold and rising and one more spiking hot and then cooling
      • The different temperature trends on the left and right side confuse some of the comparison between the left and right
      • Andrew contends that it will always be the case with the two mirrors, where they will differ in temperature
    • Though we expect LBCr to be better than LBCb (longer wavelength == smaller FWHM), but sometimes we see the behavior flip
      • On LBCr, the averaged FWHM across the focal plane has a larger deviation than across the LBCb focal plane
    • John notes that the temperature gradient on the two mirrors was not very different
    • There was a brief confusion on which channels are being called, but we sorted it out
      • It is possible that the wrong channel.py file is located on the mountain robs machine and TMS might poll the wrong channels
      • Yang plans to check if the channels are up to date or not
    • The Sx side channels were not being polled
    • John suggests writing in a simple way to adjust the logic in John's code to allow running TMS on only one side
    • John and Yang to update their code to accomodate the sort of operation mode encountered on 20211213, active TMS on one side, no TMS on the other
    • Yang planning to move all of the options to a config file that can be edited to affect all part of the MetrologySupport code, rather than having to go in to each piece separately, likely after the break
      • As of right now, you have to go into many separate pieces of code to adjust the operational parameters
  • It appears we will need more engineering time to validate TMS for commissioning

H/W:
  • ´╗┐Nowadays, we have been turning on and off the laser b/c our shutter control is not very reliable, possibly b/c of the USB cable
  • Heejoo bought three different cables for testing; the primary difference is the gauge or the allowable bandwidth of the USB shutter control cable, as per Etalon's suggestion
  • Andrew suggests going back to Mark at Etalon to figure out which parameters of the USB cable are most important and get the best one for the job (shielding and length are likely some of the most important features)
  • To test the cables, Heejoo has been using a passive data collection mode
  • During testing, Heejoo has been opening the shutter repeatedly for testing to take measurements and ensure the USB cable is working
  • Heejoo swapped the old USB cable with a new one with LeRoy's help, but he suggests we may need to go back to the original for commissioning tests
  • Once the mountain network is back online, Heejoo plans to run another test with the new cable
  • Heejoo is uncertain as to whether or not this error is actually caused by the USB cable or if it's something else
  • However, using the old method of leaving the shutter open while running TMS, the system works fine, but we would really like to bring back the shutter functionality so that we can leave the laser on for extended lifetime
  • Yang wonders if it is possible that part of the H/W issues could actually be coming from issues with incorrect channel numbering in the software (channel.py or channel_group.py)
  • There is a possibility that the Etalon software is handling bad data poorly, b/c the software should not just crash upon a shutter communication error requiring a full system restart
  • Andrew to write a letter to Mark and Kai at Etalon to see if it's possible that there could be a solution on their end, if they could look into error handling this kind of fatal error
  • It's possible there could also be an issue with the way that MATLAB handles COM ports during long-term operation (Heejoo has experienced this with multi-camera operation within MATLAB)
  • We will consider disabiling unused COM ports

Andrew on Measuring Aspheres w/ Laser Trackers:
  • Has been working on fitting aspheres with Zernikes for asphere measurement
  • Andrew has a nice piece of code that enables fitting of Zernikes during asphere measurement, which could be used for fitting the thermal data on the faceplate and backplate of the mirrors
  • Notes that we have better data from active optics in Gregorian mode (but we don't have TMS running, so this doesn't solve our problems)
  • Andrew and Trenton to look on extracting data on focus from FPIA from any night we have run TMS from active or passive nights
  • We should have plenty of data for temperature telemetry to get Zernike for front and back, take differences, and look at how this relates to historical focus over time
  • Could be broadened to include more data on spherical
  • Buddy Martin has expressed a strong interest in this, and we could keep him in the loop (and Breann)
  • The analysis will have to be smart enough to know how many sets of temperature data we are including in a given night
  • Andrew to share the Mathematica notebook to run this data
  • With existing data, we should be able to make good progress on this
  • We have a few cases where we have been doing active TMS to predict the appropriate focus correction to correct against the current simple thermfoc correction developed by Andrew
  • There's a good chance we will need to update the way that the thermal focus analysis drops bad thermocouples (add the ability to add them back in if the channel comes back online)

LBT Primary CAD Modelling:
  • Heejoo and Trenton met with Brandon Challifoux to discuss the method for thermal deformation and gravity sag analysis of the LBT primary
  • Planning to use ANSYS, and try out different constraint models for the actuator stiffnesses
  • Finding clever ways to calculate the stiffness matrix of the mirror by simplifying geometry with shells and 2D plane stresses
  • Brandon will also be working alongside us on this task, as he wants to learn how to use ANSYS

Engineering Time:
  • 20211226 UT: probably going to be used up by PEPSI
  • 21211227 UT: similarly, likely going to be used up by PEPSI

Resuming Meetings:
  • We will resume TMS meetings on January 6th
  • Heejoo and Andrew will be available to run TMS over the break

-- TrentonBrendel - 16 Dec 2021
Topic revision: r1 - 17 Dec 2021, TrentonBrendel
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Foswiki? Send feedback