You can follow along in T1500377. Or said more correctly, you'll *need* this document open to help follow along.

We have found actuation strength of the QUAD is varying slowly as a function of time, and we're quite confident that slow, time-dependent, charge on the test mass is changing the test mass actuation strength alone (relative to the upper stages of the quad). As such, and before the PUM / TST crossover is in the gravitational wave band, we need to be able to independently track the ESD strength as a function of time, with respect to the overall actuation strength. We have explored several different methods for potentially compensating the variation that we see (see T1500422 for discussion), and the method that results in the least amount of systematic error in the correction needs new front-end infrastructure. This update is to install said infrastructure.

Some of the details:
(1) New channels stored into science frames for storing each stage of actuation independently. These new channels are stored at 4096 [Hz].
$(IFO):CAL-DELTAL_CTRL_TST_DQ
$(IFO):CAL-DELTAL_CTRL_PUM_DQ
$(IFO):CAL-DELTAL_CTRL_UIM_DQ
Just like the $(IFO):CAL-DELTAL_CTRL_DQ channel, these channels will be whitened to avoid single-precision errors. Note that, for now, I've installed the same whitening filters, i.e.
      zpk([1;1;1],[500;500;500],1,"n)
but we'll need to do the same kind of study that Shivaraj did to make sure they're not limited by precision errors.

Also, in doing so, I've *reduced* the frame rate for $(IFO):CAL-DELTAL_CTRL_DQ also to 4096 [Hz].

Note, for offline analysis, this means one now needs to account for the DAQ down sampling filter between 16 [kHz] and 4 [kHz], which is detailed as the "x4" filter in T1400719.

(2) There are LOTS of new EPICs records to support the calculations defined in Eqs. 9, 11, 12, and 15 of T1500377. While we thought that the GDS calibration pipeline would be able to simply absorb EPICs records that housed the model parameter values (e.g. A_{0}^{tst}, A_{0}^{pu}, C_{0}, G_{0}) at each calibration line frequency, Maddie tested the new code, and found memory issues with all of the calucaltions that were necessary.

As such, now, we not only store the "raw" parameters values, but from those values, the contant / reference / "_{0}" terms in the above mentioned equations are calculated in the front-end. It's a mess of complex analysis, but the new EPICs records, and there associated relation to the above equations are mentioned below. 
for f_{tst} @ ~ 35 [Hz]
    User entered EPICs records:
    $(IFO):CAL-CS_TDEP_ESD_LINE1_REF_OLG_$(REAL,IMAG)             G_{0}                @ f_{tst}
    $(IFO):CAL-CS_TDEP_ESD_LINE1_REF_D_$(REAL,IMAG)               D                    @ f_{tst}
    $(IFO):CAL-CS_TDEP_ESD_LINE1_REF_C_$(REAL,IMAG)               C_{0}                @ f_{tst}
    $(IFO):CAL-CS_TDEP_ESD_LINE1_REF_C_NOCAVPOLE_$(REAL,IMAG)     C_{res}              @ f_{tst}
    $(IFO):CAL-CS_TDEP_ESD_LINE1_REF_A_TST_$(REAL,IMAG)           A_{0}^{tst}          @ f_{tst}
    $(IFO):CAL-CS_TDEP_ESD_LINE1_REF_A_PUM_$(REAL,IMAG)           A_{0}^{pum}          @ f_{tst}
    $(IFO):CAL-CS_TDEP_ESD_LINE1_REF_A_UIM_$(REAL,IMAG)           A_{0}^{uim}          @ f_{tst}
    $(IFO):CAL-CS_TDEP_ESD_LINE1_REF_A_TOP_$(REAL,IMAG)           A_{0}^{top}          @ f_{tst}

    Calcalulated records:
    $(IFO):CAL-CS_TDEP_ESD_LINE1_REF_A_TST_MODSQ                  |A_{0}^{tst}|^2      @ f_{tst}
    $(IFO):CAL-CS_TDEP_ESD_LINE1_REF_A_TST_NEGINV_$(REAL,IMAG)    - (1 / A_{0}^{tst})  @ f_{tst}    for Eq. 9
    $(IFO):CAL-CS_TDEP_ESD_LINE1_REF_A_USUM_$(REAL,IMAG)           A_{0}^{pu}          @ f_{tst}    = A_{0}^{top} + A_{0}^{uim} + A_{0}^{pum}
    $(IFO):CAL-CS_TDEP_ESD_LINE1_REF_A_USUM_MODSQ                 |A_{0}^{pu}|^2       @ f_{tst}
    $(IFO):CAL-CS_TDEP_ESD_LINE1_REF_A_USUM_INV_$(REAL,IMAG)      (1 /  A_{0}^{pu})    @ f_{tst}         
    $(IFO):CAL-CS_TDEP_ESD_LINE1_REF_C_MODSQ                      |C_{0}|^2            @ f_{tst}
    $(IFO):CAL-CS_TDEP_ESD_LINE1_REF_INVCLG_$(REAL,IMAG)          (1 + G_{0}) / C_{0}  @ f_{tst}    for Eq. 9

for f_{pcal} @ ~ 35 [Hz] (which are currently running on PCALY)
    User entered EPICs records:
    $(IFO):CAL-CS_TDEP_PCALY_LINE1_REF_OLG_$(REAL,IMAG)           G_{0}                @ f_{pcal}
    $(IFO):CAL-CS_TDEP_PCALY_LINE1_REF_D_$(REAL,IMAG)             D                    @ f_{pcal}
    $(IFO):CAL-CS_TDEP_PCALY_LINE1_REF_C_$(REAL,IMAG)             C_{0}                @ f_{pcal}
    $(IFO):CAL-CS_TDEP_PCALY_LINE1_REF_C_NOCAVPOLE_$(REAL,IMAG)   C_{res}              @ f_{pcal}
    $(IFO):CAL-CS_TDEP_PCALY_LINE1_REF_A_TST_$(REAL,IMAG)         A_{0}^{tst}          @ f_{pcal}    
    $(IFO):CAL-CS_TDEP_PCALY_LINE1_REF_A_PUM_$(REAL,IMAG)         A_{0}^{pum}          @ f_{pcal}
    $(IFO):CAL-CS_TDEP_PCALY_LINE1_REF_A_UIM_$(REAL,IMAG)         A_{0}^{uim}          @ f_{pcal}
    $(IFO):CAL-CS_TDEP_PCALY_LINE1_REF_A_TOP_$(REAL,IMAG)         A_{0}^{top}          @ f_{pcal}
    
    Calcalulated records:
    $(IFO):CAL-CS_TDEP_PCALY_LINE1_REF_A_USUM_$(REAL,IMAG)        A_{0}^{pu}           @ f_{pcal}    = A_{0}^{top} + A_{0}^{uim} + A_{0}^{pum}
    $(IFO):CAL-CS_TDEP_PCALY_LINE1_REF_G_MODSQ                    |G_{0}|^{2}          @ f_{pcal}
    $(IFO):CAL-CS_TDEP_PCALY_LINE1_REF_G_TWICEREAL                2 Re{G_{0}}          @ f_{pcal}
    $(IFO):CAL-CS_TDEP_PCALY_LINE1_REF_ONEPLUSG_MODSQ             |1+G_{0}|^2          @ f_{pcal}    
    $(IFO):CAL-CS_TDEP_PCALY_LINE1_REF_CLG0_$(REAL,IMAG)          C_{0} / (1 + G_{0})  @ f_{pcal}    for Eq. 9 and Eq. 11

for f_{ctrl} @ ~ 35 [Hz]
    User entered EPICs records:
    $(IFO):CAL-CS_TDEP_DARM_LINE1_REF_OLG_$(REAL,IMAG)            G_{0}                @ f_{ctrl}
    $(IFO):CAL-CS_TDEP_DARM_LINE1_REF_D_$(REAL,IMAG)              D                    @ f_{ctrl}
    $(IFO):CAL-CS_TDEP_DARM_LINE1_REF_C_$(REAL,IMAG)              C_{0}                @ f_{ctrl}
    $(IFO):CAL-CS_TDEP_DARM_LINE1_REF_C_NOCAVPOLE_$(REAL,IMAG)    C_{res}              @ f_{ctrl}
    $(IFO):CAL-CS_TDEP_DARM_LINE1_REF_A_TST_$(REAL,IMAG)          A_{0}^{tst}          @ f_{ctrl}    
    $(IFO):CAL-CS_TDEP_DARM_LINE1_REF_A_PUM_$(REAL,IMAG)          A_{0}^{pum}          @ f_{ctrl}    
    $(IFO):CAL-CS_TDEP_DARM_LINE1_REF_A_UIM_$(REAL,IMAG)          A_{0}^{uim}          @ f_{ctrl}    
    $(IFO):CAL-CS_TDEP_DARM_LINE1_REF_A_TOP_$(REAL,IMAG)          A_{0}^{top}          @ f_{ctrl}    

    Calcalulated records:
    $(IFO):CAL-CS_TDEP_DARM_LINE1_REF_A_TST_MODSQ                 |A_{0}^{tst}|^2      @ f_{ctrl}
    $(IFO):CAL-CS_TDEP_DARM_LINE1_REF_A_TST_NEGINV_$(REAL,IMAG)   - (1 / A_{0})        @ f_{ctrl}
    $(IFO):CAL-CS_TDEP_DARM_LINE1_REF_A_USUM_$(REAL,IMAG,MODSQ)   A_{0}^{pu}           @ f_{ctrl}  = A_{0}^{top} + A_{0}^{uim} + A_{0}^{pum}
    $(IFO):CAL-CS_TDEP_DARM_LINE1_REF_A_USUM_INV_$(REAL,IMAG)     1 / A_{0}            @ f_{ctrl}  This is the constant part of Eq. 12
    $(IFO):CAL-CS_TDEP_DARM_LINE1_REF_C_MODSQ                     |C_{0}|^{2}          @ f_{ctrl}
    $(IFO):CAL-CS_TDEP_DARM_LINE1_REF_INVCLG_$(REAL,IMAG)         (1 + G_{0}) / C_{0}  @ f_{ctrl}  

For Eq 9,
    Calculated Record:
    $(IFO):CAL-CS_TDEP_REF_INVA_CLGRATIO_TST_$(REAL,IMAG)         - (1 / A_{0}^{tst})|f_tst *  C_{0} / (1 + G_{0})|f_pcal * (1 + G_{0}) / C_{0}|f_tst
                                                                  This is the constant part of Eq. 9
For Eq 11,
    Calculated Record:
    $(IFO):CAL-CS_TDEP_REF_CLGRATIO_CTRL_$(REAL,IMAG)             C_{0} / (1 + G_{0})|f_pcal * (1 + G_{0}) / C_{0} |f_ctrl
                                                                  This is the constant part of Eq. 11

for f_{pcal2} @ ~330 [Hz] (which are currently running on PCALY)
    User entered EPICs records:
    $(IFO):CAL-CS_TDEP_PCALY_LINE2_REF_OLG_$(REAL,IMAG)           G_{0}                @ f_{pcal}
    $(IFO):CAL-CS_TDEP_PCALY_LINE2_REF_D_$(REAL,IMAG)             D                    @ f_{pcal}     For Eq. 15 
    $(IFO):CAL-CS_TDEP_PCALY_LINE2_REF_C_$(REAL,IMAG)             C_{0}                @ f_{pcal}
    $(IFO):CAL-CS_TDEP_PCALY_LINE2_REF_C_NOCAVPOLE_$(REAL,IMAG)   C_{res}              @ f_{pcal}     For Eq. 15 
    $(IFO):CAL-CS_TDEP_PCALY_LINE2_REF_A_TST_$(REAL,IMAG)         A_{0}^{tst}          @ f_{pcal}     For Eq. 15 
    $(IFO):CAL-CS_TDEP_PCALY_LINE2_REF_A_PUM_$(REAL,IMAG)         A_{0}^{pum}          @ f_{pcal}
    $(IFO):CAL-CS_TDEP_PCALY_LINE2_REF_A_UIM_$(REAL,IMAG)         A_{0}^{uim}          @ f_{pcal}
    $(IFO):CAL-CS_TDEP_PCALY_LINE2_REF_A_TOP_$(REAL,IMAG)         A_{0}^{top}          @ f_{pcal}
    
    Calcalulated records:
    $(IFO):CAL-CS_TDEP_PCALY_LINE2_REF_A_USUM_$(REAL,IMAG)        A_{0}^{pu}           @ f_{pcal}    = A_{0}^{top} + A_{0}^{uim} + A_{0}^{pum},  For Eq. 15 
    $(IFO):CAL-CS_TDEP_PCALY_LINE2_REF_G_MODSQ                    |G_{0}|^{2}          @ f_{pcal}
    $(IFO):CAL-CS_TDEP_PCALY_LINE2_REF_G_TWICEREAL                2 Re{G_{0}}          @ f_{pcal}
    $(IFO):CAL-CS_TDEP_PCALY_LINE2_REF_ONEPLUSG_MODSQ             |1+G_{0}|^2          @ f_{pcal}    
    $(IFO):CAL-CS_TDEP_PCALY_LINE2_REF_CLG0_$(REAL,IMAG)          C_{0} / (1 + G_{0})  @ f_{pcal}    for Eq. 9 and Eq. 11


I've not yet installed all of the above user defined EPICs records, bare with us while we get the MEDM infrastructure updated, and then we'll fill in the values needed. Note, the only difference from what I installed on Friday (see LHO aLOG 20291), is what was quoted for the "ESD_LINE" there is now called "DARM_LINE," and I need to install new values for the *actual* "ESD_LINE".

Also, I've committed all model changes to the userapps svn repo, and I have captured all of the new new settings (thus far) in the SDF system, initializing new channels, and removing old channels.