Let 2-CCD (S3+FI) observations with less than 300 counts in the sequence go to -109 C

acis_thermal_check/acis_obs.py

@@ -1,5 +1,6 @@
 import numpy as np
 from acis_thermal_check.utils import mylog
+from kadi.commands.states import decode_power
 
 
 def who_in_fp(simpos=80655):
@@ -86,21 +87,11 @@ def fetch_ocat_data(obsid_list):
     if got_table:
         tab = ascii.read(resp.text, header_start=0, data_start=2)
         tab.sort("OBSID")
-        # We figure out the CCD count from the table by finding out
-        # which ccds were on, optional, or dropped, and then
-        # subtracting off the dropped chip count entry in the table
-        ccd_count = np.zeros(tab["S3"].size, dtype='int')
-        for a, r in zip(["I", "S"], [range(4), range(6)]):
-            for i in r:
-                ccd = np.ma.filled(tab[f"{a}{i}"].data)
-                ccd_count += (ccd == "Y").astype('int')
-                ccd_count += (ccd == "D").astype('int')
-                ccd_count += np.char.startswith(ccd, "O").astype('int')
-        ccd_count -= tab["DROPPED_CHIP_CNT"].data.astype('int')
         # Now we have to find all of the obsids in each sequence and then
         # compute the complete exposure for each sequence
-        seq_nums = list(tab["SEQ_NUM"].data.astype("str"))
-        seq_num_list = ",".join([seq_num for seq_num in seq_nums if seq_num != " "])
+        seq_nums = np.unique([str(sn) for sn in tab["SEQ_NUM"].data.astype("str") 
+                              if sn != " "])
+        seq_num_list = ",".join(seq_nums)
         obsids = tab["OBSID"].data.astype("int")
         cnt_rate = tab["EST_CNT_RATE"].data.astype("float64")
         params = {"seqNum": seq_num_list}
@@ -121,15 +112,18 @@ def fetch_ocat_data(obsid_list):
             return None
         tab_seq = ascii.read(resp.text, header_start=0, data_start=2)
         app_exp = np.zeros_like(cnt_rate)
-        for row in tab_seq:
-            i = seq_nums.index(str(row["SEQ_NUM"]))
-            app_exp[i] += np.float64(row["APP_EXP"])
+        seq_nums = tab_seq["SEQ_NUM"].data.astype("str")
+        for i, row in enumerate(tab):
+            sn = str(row["SEQ_NUM"])
+            if sn == " ":
+                continue
+            j = np.where(sn == seq_nums)[0]
+            app_exp[i] += np.float64(tab_seq["APP_EXP"][j]).sum()
         app_exp *= 1000.0
         table_dict = {"obsid": np.array(obsids),
                       "grating": tab["GRAT"].data,
-                      "ccd_count": ccd_count,
-                      "S3": np.ma.filled(tab["S3"].data),
-                      "num_counts": cnt_rate*app_exp}
+                      "cnt_rate": cnt_rate,
+                      "app_exp": app_exp}
     else:
         # We weren't able to get a valid table for some reason, so
         # we cannot check for -109 data, but we proceed with the
@@ -210,12 +204,18 @@ def find_obsid_intervals(cmd_states, load_start):
             datestart = eachstate['datestart']
             tstart = eachstate['tstart']
 
+        # Process the power command which turns things on
+        if pow_cmd.startswith("WSPOW") and pow_cmd != 'WSPOW00000' \
+            and firstpow:
+            ccds = decode_power(pow_cmd)['ccds'].replace(" ", ",")[:-1]
+
         # Process the first XTZ0000005 line you see
         if pow_cmd in ['XTZ0000005', 'XCZ0000005'] and \
                 (xtztime is None and firstpow):
             xtztime = eachstate['tstart']
             # MUST fix the instrument now
             instrument = who_in_fp(eachstate['simpos'])
+            ccd_count = eachstate["ccd_count"]
 
         # Process the first AA00000000 line you see
         if pow_cmd == 'AA00000000' and firstpow:
@@ -234,9 +234,11 @@ def find_obsid_intervals(cmd_states, load_start):
                               "datestop": datestop,
                               "tstart": tstart,
                               "tstop": tstop,
+                              "ccds": ccds,
                               "start_science": xtztime,
                               "obsid": eachstate['obsid'],
-                              "instrument": instrument}
+                              "instrument": instrument,
+                              "ccd_count": ccd_count}
                 obsid_interval_list.append(obsid_dict)
 
             # now clear out the data values
@@ -295,15 +297,30 @@ def acis_filter(obsid_interval_list):
     acis_i = []
     cold_ecs = []
 
+    mylog.debug("OBSID\tCNT_RATE\tAPP_EXP\tNUM_CTS\tGRATING\tCCDS")
     for eachobs in obsid_interval_list:
+        # First we check that we got ocat data using "grating"
+        hot_acis = False
         if "grating" in eachobs:
-            hetg = eachobs["grating"] == "HETG"
-            s3_only = eachobs["S3"] == "Y" and eachobs["ccd_count"] == 1
-            hot_acis = hetg or (eachobs["num_counts"] < 300.0 and s3_only)
-        else:
-            hot_acis = False
+            eachobs["num_counts"] = int(eachobs["cnt_rate"] * eachobs["app_exp"])
+            # First check to see if this is an S3 observation
+            mylog.debug(f"{eachobs['obsid']}\t{eachobs['cnt_rate']}\t"
+                        f"{eachobs['app_exp']*1.0e-3}\t{eachobs['num_counts']}\t"
+                        f"{eachobs['grating']}\t{eachobs['ccds']}")
+            if eachobs["ccd_count"] <= 2:
+                # S3 with low counts
+                low_ct_s3 = eachobs["num_counts"] < 300 and "S3" in eachobs["ccds"]
+                # Is there another chip on? Make sure it's not S1
+                if eachobs["ccd_count"] == 2:
+                    low_ct_s3 &= "S1" not in eachobs["ccds"]
+            else:
+                # All higher ccd counts are invalid
+                low_ct_s3 = False
+            # Also check grating status
+            hot_acis = (eachobs["grating"] == "HETG") or low_ct_s3
+        eachobs['hot_acis'] = hot_acis
         if hot_acis:
-            acis_hot.append(eachobs) 
+            acis_hot.append(eachobs)
         else:
             if eachobs["instrument"] == "ACIS-S":
                 acis_s.append(eachobs)

acis_thermal_check/apps/acisfp_check.py

@@ -137,13 +137,6 @@ def make_prediction_plots(self, outdir, states, temps, load_start):
         those from the commanded states data structure called "states" 
         """
 
-        # extract the OBSID's from the commanded states. NOTE: this contains all
-        # observations including ECS runs and HRC observations
-        observation_intervals = find_obsid_intervals(states, load_start)
-
-        # Filter out any HRC science observations BUT keep ACIS ECS observations
-        self.acis_and_ecs_obs = hrc_science_obs_filter(observation_intervals)
-
         # create an empty dictionary called plots to contain the returned
         # figures, axes 1 and axes 2 of the PredictPlot class
         plots = {}
@@ -192,15 +185,18 @@ def make_prediction_plots(self, outdir, states, temps, load_start):
                         textypos[i], fontsize[i], plot_start)
 
             # These next lines are dummies so we can get the obsids in the legend
+            plots[name].ax.errorbar([0.0, 0.0], [1.0, 1.0], xerr=1.0,
+                                    lw=2, xlolims=True, color='blue',
+                                    capsize=4, capthick=2, label='Cold ECS')
             plots[name].ax.errorbar([0.0, 0.0], [1.0, 1.0], xerr=1.0,
                                     lw=2, xlolims=True, color='red',
                                     capsize=4, capthick=2, label='ACIS-I')
             plots[name].ax.errorbar([0.0, 0.0], [1.0, 1.0], xerr=1.0,
                                     lw=2, xlolims=True, color='green',
                                     capsize=4, capthick=2, label='ACIS-S')
             plots[name].ax.errorbar([0.0, 0.0], [1.0, 1.0], xerr=1.0,
-                                    lw=2, xlolims=True, color='blue',
-                                    capsize=4, capthick=2, label='ECS')
+                                    lw=2, xlolims=True, color='saddlebrown',
+                                    capsize=4, capthick=2, label='Hot ACIS-S')
 
             # Make the legend on the temperature plot
             plots[name].ax.legend(bbox_to_anchor=(0.15, 0.99),
@@ -230,7 +226,7 @@ def make_prediction_plots(self, outdir, states, temps, load_start):
 
         return plots
 
-    def make_prediction_viols(self, temps, load_start):
+    def make_prediction_viols(self, temps, states, load_start):
         """
         Find limit violations where predicted temperature is above the
         red minus margin.
@@ -252,9 +248,16 @@ def make_prediction_viols(self, temps, load_start):
                  - science_viols
 
         """
+        # extract the OBSID's from the commanded states. NOTE: this contains all
+        # observations including ECS runs and HRC observations
+        observation_intervals = find_obsid_intervals(states, load_start)
+
+        # Filter out any HRC science observations BUT keep ACIS ECS observations
+        self.acis_and_ecs_obs = hrc_science_obs_filter(observation_intervals)
+
         times = self.predict_model.times
 
-        mylog.info(f"\nMAKE VIOLS Checking for limit violations in "
+        mylog.info(f"MAKE VIOLS Checking for limit violations in "
                    f"{len(self.acis_and_ecs_obs)} total science observations")
 
         viols = {}
@@ -278,7 +281,7 @@ def make_prediction_viols(self, temps, load_start):
         acis_hot_limit = self.limits["acis_hot"].value
         cold_ecs_limit = self.limits["cold_ecs"].value
 
-        mylog.info(f'\n\nACIS-I Science ({acis_i_limit} C) violations')
+        mylog.info(f'ACIS-I Science ({acis_i_limit} C) violations')
 
         # Create the violation data structure.
         acis_i_viols = self.search_obsids_for_viols("ACIS-I",
@@ -293,7 +296,7 @@ def make_prediction_viols(self, temps, load_start):
         # science run. These are load killers
         # ------------------------------------------------------------
         #
-        mylog.info(f'\n\nACIS-S Science ({acis_s_limit} C) violations')
+        mylog.info(f'ACIS-S Science ({acis_s_limit} C) violations')
 
         acis_s_viols = self.search_obsids_for_viols("ACIS-S",
             acis_s_limit, ACIS_S_obs, temp, times, load_start)
@@ -306,7 +309,7 @@ def make_prediction_viols(self, temps, load_start):
         # science run which can run hot. These are load killers
         # ------------------------------------------------------------
         #
-        mylog.info(f'\n\nACIS-S Science ({acis_hot_limit} C) violations')
+        mylog.info(f'ACIS-S Science ({acis_hot_limit} C) violations')
 
         acis_hot_viols = self.search_obsids_for_viols("Hot ACIS-S",
             acis_hot_limit, ACIS_hot_obs, temp, times, load_start)
@@ -317,7 +320,7 @@ def make_prediction_viols(self, temps, load_start):
         # ------------------------------------------------------------
         # Science Orbit ECS -119.5 violations; -119.5 violation check
         # ------------------------------------------------------------
-        mylog.info(f'\n\nScience Orbit ECS ({cold_ecs_limit} C) violations')
+        mylog.info(f'Science Orbit ECS ({cold_ecs_limit} C) violations')
 
         ecs_viols = self.search_obsids_for_viols("Science Orbit ECS",
             cold_ecs_limit, sci_ecs_obs, temp, times, load_start)
@@ -328,7 +331,7 @@ def make_prediction_viols(self, temps, load_start):
 
         # Store all obsids which can go to -109 C
         for obs in ACIS_hot_obs:
-            self.acis_hot_obs.append(obs["obsid"])
+            self.acis_hot_obs.append(obs)
 
         return viols
 
@@ -416,6 +419,7 @@ def draw_obsids(obs_list, plots, msid, ypos, endcapstart, endcapstop,
 
         obsid = eachobservation['obsid']
         in_fp = eachobservation['instrument']
+        hot_acis = eachobservation['hot_acis']
 
         if obsid > 60000:
             # ECS observations during the science orbit are colored blue
@@ -426,7 +430,7 @@ def draw_obsids(obs_list, plots, msid, ypos, endcapstart, endcapstop,
             if in_fp == "ACIS-I":
                 color = 'red'
             else:
-                color = 'green'
+                color = 'saddlebrown' if hot_acis else 'green'
 
         obsid_txt = str(obsid)
         # If this is an ECS measurement in the science orbit mark

acis_thermal_check/main.py

@@ -339,12 +339,13 @@ def make_week_predict(self, tstart, tstop, tlm, T_init, model_spec,
         plt.rc("grid", linewidth=1.5)
 
         temps = {self.name: model.comp[self.msid].mvals}
-        # make_prediction_plots runs the validation of the model 
-        # against previous telemetry
-        plots = self.make_prediction_plots(outdir, states, temps, tstart)
 
         # make_prediction_viols determines the violations and prints them out
-        viols = self.make_prediction_viols(temps, tstart)
+        viols = self.make_prediction_viols(temps, states, tstart)
+
+        # make_prediction_plots runs the validation of the model
+        # against previous telemetry
+        plots = self.make_prediction_plots(outdir, states, temps, tstart)
 
         # write_states writes the commanded states to states.dat
         self.write_states(outdir, states)
@@ -504,7 +505,7 @@ def _make_prediction_viols(self, times, temp, load_start, limit, lim_name,
 
         return viols
 
-    def make_prediction_viols(self, temps, load_start):
+    def make_prediction_viols(self, temps, states, load_start):
         """
         Find limit violations where predicted temperature is above the
         specified limits.
@@ -513,6 +514,8 @@ def make_prediction_viols(self, temps, load_start):
         ----------
         temps : dict of NumPy arrays
             NumPy arrays corresponding to the modeled temperatures
+        states : NumPy record array
+            Commanded states
         load_start : float
             The start time of the load, used so that we only report
             violations for times later than this time for the model

acis_thermal_check/templates/index_template.rst

@@ -34,9 +34,15 @@ States                 `<states.dat>`_
 {% if proc.msid == "FPTEMP" %}
 "Hot" ACIS Observations (-109 C limit)
 --------------------------------------
-{% for obsid in acis_hot_obs %}
-| {{ obsid }}  
+{% if acis_hot_obs|length > 0 %}
+=====  =================  ==================  =======
+Obsid  CCDs               # of counts in seq  Grating
+=====  =================  ==================  =======
+{% for eachobs in acis_hot_obs %}
+{{ eachobs.obsid }}  {{"{0: <17}".format(eachobs.ccds)}}  {{"{0: <18}".format(eachobs.num_counts)}}          {{eachobs.grating}}
 {% endfor %}
+=====  =================  ==================  =======
+{% endif %}
 {% endif %}
 
 {% for key in viols.keys() %}

acis_thermal_check/tests/acisfp/answers/FEB2122A_hot_acis.json

@@ -0,0 +1,26 @@
+[
+    [
+        "24244",
+        "S0,S1,S2,S3,S4,S5",
+        "765000",
+        "HETG"
+    ],
+    [
+        "25556",
+        "S3",
+        "3",
+        "NONE"
+    ],
+    [
+        "26335",
+        "S1,S2,S3,S4",
+        "765000",
+        "HETG"
+    ],
+    [
+        "26334",
+        "S3",
+        "29",
+        "NONE"
+    ]
+]

acis_thermal_check/tests/acisfp/answers/FEB2822A_hot_acis.json

@@ -0,0 +1,38 @@
+[
+    [
+        "25333",
+        "S3",
+        "5",
+        "NONE"
+    ],
+    [
+        "23653",
+        "S2,S3",
+        "260",
+        "NONE"
+    ],
+    [
+        "24705",
+        "S2,S3",
+        "500000",
+        "HETG"
+    ],
+    [
+        "26344",
+        "S2,S3",
+        "500000",
+        "HETG"
+    ],
+    [
+        "26345",
+        "S2,S3",
+        "500000",
+        "HETG"
+    ],
+    [
+        "25291",
+        "S3",
+        "61",
+        "NONE"
+    ]
+]