Abstract
This post is a detailed report of klog#25395.
Because SDF set points of two unused switches (K1:CAL-PCAL_EY_CLINE_A_SW and K1:CAL-PCAL_EY_CLINE_B_SW) were not well managed,
I modified SDF set points on observation.snap after discussing with Miyoki-san and Miyakawa-san.
ON/OFF situation of these switches are not affected the Interferometer situation at all.
These modifications should be accepted also on safe.snap and down.snap in the next maintenance day.

Details
Because of the historical reasons, there are a lot of wasted switches on the signal path of calibration lines.
In current implementation, ON/OFF situation of all calibration lines are managed by the oscillator gain such as K1:CAL-PCAL_EY_CLINE_A_OSC_CLKGAIN.
An oscillator gain is changed proper value in ENABLING_LINE and changed 0 in DISABLING_LINE by Pcal guardians.

In addition, we have a lot of unnecessary switches on the signal path of calibration lines such as K1:CAL-PCAL_EY_CLINE_A_SW.
These switches are opened (ON) in TURN_ON_EXCITE_SW state and closed (OFF) in REDUCE_P state due to obsoleted implementation of Pcal guardian.
(These switches are no longer needed. They should be removed from the real-time model after O4a.)

Before starting O4a, we missed the fact that set points of these switches were ON in observation.snap of k1caley
even though calibration lines are not injected from YPcal.

When EQ occurred and ETMY went to the ISOLATED state by the safety structure of guardian implementation,
YPcal guardian also went to low power mode for the safety.
At this time, these unnecessary switches were meaninglessly closed (OFF) in REDUCE_P state
and they weren't opened (ON) at coming back to READY state because YPcal lines were not used.

As the results of discussion with Miyoki-san and Miyakawa-san,
we concluded that switches on the signal path of unused calibration lines should be closed (OFF).
So we decided to accepted the set points of K1:CAL-PCAL_EY_CLINE_A_SW and K1:CAL-PCAL_EY_CLINE_B_SW as OFF as shown in Fig.1.
(This figure is same as one in klog#25395.)

Regardless of the state of these switches, the fact reminas that the calibration lines from YPcal are absence.