.. index:: Extended pixel edge response (CTI-EPER), Pixel bounce; Calibration observations .. _calobs-vis-cti-eper: **CALPRODUCT-VS-018:** VIS Extended Pixel Edge Response (CTI-EPER) ================================================================== .. include:: definitions.rst Parent requirement ------------------ Not in the `CalCD-B `_ Scheduling ---------- Twice during PV phase. At the beginning, when radiation damage is still very low, and a second time after 4 weeks. Time-critical ------------- Yes. Must be taken close in time with | :ref:`calobs-vis-ptc` | :ref:`calobs-vis-cti-cil`. To this end, the required flat fields have been included in the observations delivering the data for | :ref:`calobs-vis-ptc` | :ref:`calobs-vis-prnu`. Objective --------- Radiation damage will increase the degree of Charge Transfer Inefficiency (CTI) in the VIS CCDs. The CTI is corrected for in the VIS science images using a calibrated CTI model. The CTI model is built based on two data sets: :ref:`calobs-vis-cti-cil` Charge Injection Lines (CIL) :ref:`calobs-vis-cti-eper` Measurements of the Extended Pixel Edge Response (EPER, or *pixel bounce*) in flat field images. This ``CalProduct`` provides the flat fields for the EPER, which can be measured in the overscan. EPER is a memory effect in the analog detection chain. It increases the value of a pixel following a very bright pixel, by typical relative amounts of a few 1e-4. We require flats with flux levels below and above the nominal charge injection (CI) range of 1000-6000 ADU, as well as inside the CI range to check consistency with the CI data. If the flats are taken in pairs, they can also be used for :ref:`calobs-vis-ptc` measurements. Fulfilled by ------------ :ref:`calobs-block-pv-vis-ptc-cti-eper` :ref:`calobs-block-f-wide-survey` Processing input ---------------- :command:`Ground Processing Input Data`: :command:`Data Description`: Processing method ----------------- The charge injection and flat field images are bias and NL corrected. A CTI model is then fitted using the VIS_CTI PE, which is performed a nonlinear search using the nested sampling algorithm MultiNest. Processing output ----------------- Updated parameters on the CTI model for that given observation data (.json of best fit parameters is stored via the DataModel), per detector quadrant. Additional information and metadata on the model-fit (e.g. nested sampling samples, summary statistics). Required accuracy, constraints ------------------------------ | `R-CAL-B-VS-4351 `_ | `R-CAL-B-VS-4354 `_ | `R-CAL-B-VS-4355 `_ | `R-CAL-B-VS-4356 `_ | `R-CAL-B-VS-4358 `_ The CTI model used to correct CTI in the VIS detectors must correct CTI to better than :math:`4.0\times10^{-3}` (TBC). Processing tools ---------------- VIS_CTI PE Validation ---------- Compare the hard edge response curves with the on ground characterization data. A change by TBD % requires an update of the pixel bounce correction (at the basis of a detector quadrant). Recovery -------- Update the pixel bounce correction. Quality control parameters -------------------------- DataModel: ``_ Contact person -------------- VIS instrument scientist .. raw:: latex \clearpage