Open trentjbrendel opened 3 years ago
Planning for TMS Test
Yang and Heejoo are planning to run a TMS test on Monday at 9:00, pending availibility of the telescope
Andrew wants to probe both absolute and relative accuracy of TMS measurement
For the relative test:
Andrew suggests a sample of mirror motions (log scale of translation, rotation) via a script for mirror motions assoc. w/ ea. ch.
Dubbed the "Mirror Dance" Script
Reconfigure Jacobian as the mirror is cycling through the dance
For the absolute:
Yang has a function to define probability of channel being dropped
We are ready to go for the test!
Andrew suggests testing functionality of driving in large astig on the primary mirror at Zenith to see if we can measure this
Need channel for two diameters, as well as lengths around the edges
If we drive in 1 um of astig, should be able to measure astig by fitting a circle around the edges
Expect to be able to detect 5 um RMS astig
We want to see where the measurement floor is; 1 um, 2 um? How far can we push it?
In PSF GUI, can command up to around 10 um RMS astig across the mirror
Could try some other modes after, if there's time
To fit sin(2 theta), we should need about nine evenly distributed points (trusses)
Andrew thinks we can measure down to 200 nm RMS astig (PV is about 2-3 X this)
200 nm RMS is 500 nm PV astig (scaling is root 6)
Could even try to measure spherical or focus
Last time, we noted that we had a dropped timestamp (all channels failed due to expected internal consistency)
Currently, if we don't have enough lines to do a Jacobian, we do nothing and wait for the next measurement
This seems to be an uncommon problem
Heejoo found MPO-12 to 6x LC Duplex, Type AF, 12 Fibers OM3 Multimode FHD MPO Casette ($68.00)
Going to buy extra unit and bring with to route fibers to Sx side on next mountain visit
Once Sx is up and running, we are essentially ready to go (offsets take care of binocular mode)
Heejoo also looked for achromatic fiber collimator... no one makes one
Andrew designed the current achromatic fiber collimator with eTalon, manufactured in China w/ Chinese glass
The fiber collimators were sold to eTalon for $15 each
Now eTalon want $400 each, b/c they aren't available anywhere else
Heejoo to speak w/ Christian about buying these from eTalon directly
Sequential vs. parallel Sx + Dx measurements
John suggests using simultaneous to ensure they don't run into one another
Andrew thinks that using an optical switch for Gregorian to get 28 channels
Gregorian: two different things
Each side measured relative to the rotator, the stiffest location on the telescope (5+7+7+1)
Having a switch would allow us to get Sx then Dx, 7 seconds + 7 seconds for 14 seconds total
Power meter on TMS laser capable of measuring laser leakage remotely
Currently, Heejoo using separate website API to monitor, iMonit poll time is 10 minutes
Also, we have a license issue, so that's a thing...
Desire to look at power directly through LBT archive (includes all telemetry data)
Need help from software guys (Peter) to input iMonit data into data mining system for LBT
Andrew suggests having an alarm in the observatory software to have let the TO know if the laser is on when it shouldn't be
This would be a great safety interlock to avoid ruining an observation run
Once we have this, TO knows the laser system operation parameters, which may allow us to run more frequently
When the laser shutters work, they work really well (completely eliminate any light from system)
LUCI and LBTI are the primary instruments that can see our laser band
Heejoo suggests that we have an "always on" TMS system, that is only turned off when LUCI or LBTI are needed
Planning for TMS Test
Yang and Heejoo are planning to run a TMS test on Monday at 9:00, pending availibility of the telescope
Andrew wants to probe both absolute and relative accuracy of TMS measurement
For the relative test:
Andrew suggests a sample of mirror motions (log scale of translation, rotation) via a script for mirror motions assoc. w/ ea. ch.
Dubbed the "Mirror Dance" Script
Reconfigure Jacobian as the mirror is cycling through the dance
For the absolute:
Yang has a function to define probability of channel being dropped
We are ready to go for the test!
Andrew suggests testing functionality of driving in large astig on the primary mirror at Zenith to see if we can measure this
Need channel for two diameters, as well as lengths around the edges
If we drive in 1 um of astig, should be able to measure astig by fitting a circle around the edges
Expect to be able to detect 5 um RMS astig
We want to see where the measurement floor is; 1 um, 2 um? How far can we push it?
In PSF GUI, can command up to around 10 um RMS astig across the mirror
Could try some other modes after, if there's time
To fit sin(2 theta), we should need about nine evenly distributed points (trusses)
Andrew thinks we can measure down to 200 nm RMS astig (PV is about 2-3 X this)
200 nm RMS is 500 nm PV astig (scaling is root 6)
Could even try to measure spherical or focus
Last time, we noted that we had a dropped timestamp (all channels failed due to expected internal consistency)
Currently, if we don't have enough lines to do a Jacobian, we do nothing and wait for the next measurement
This seems to be an uncommon problem
Heejoo found MPO-12 to 6x LC Duplex, Type AF, 12 Fibers OM3 Multimode FHD MPO Casette ($68.00)
Going to buy extra unit and bring with to route fibers to Sx side on next mountain visit
Once Sx is up and running, we are essentially ready to go (offsets take care of binocular mode)
Heejoo also looked for achromatic fiber collimator... no one makes one
Andrew designed the current achromatic fiber collimator with eTalon, manufactured in China w/ Chinese glass
The fiber collimators were sold to eTalon for $15 each
Now eTalon want $400 each, b/c they aren't available anywhere else
Heejoo to speak w/ Christian about buying these from eTalon directly
Sequential vs. parallel Sx + Dx measurements
John suggests using simultaneous to ensure they don't run into one another
Andrew thinks that using an optical switch for Gregorian to get 28 channels
Gregorian: two different things
Each side measured relative to the rotator, the stiffest location on the telescope (5+7+7+1)
Having a switch would allow us to get Sx then Dx, 7 seconds + 7 seconds for 14 seconds total
Power meter on TMS laser capable of measuring laser leakage remotely
Currently, Heejoo using separate website API to monitor, iMonit poll time is 10 minutes
Also, we have a license issue, so that's a thing...
Desire to look at power directly through LBT archive (includes all telemetry data)
Need help from software guys (Peter) to input iMonit data into data mining system for LBT
Andrew suggests having an alarm in the observatory software to have let the TO know if the laser is on when it shouldn't be
This would be a great safety interlock to avoid ruining an observation run
Once we have this, TO knows the laser system operation parameters, which may allow us to run more frequently
When the laser shutters work, they work really well (completely eliminate any light from system)
LUCI and LBTI are the primary instruments that can see our laser band
Heejoo suggests that we have an "always on" TMS system, that is only turned off when LUCI or LBTI are needed