next up previous contents
Next: About this document ... Up: pdbi-cookbook Previous: Offline Calibration; the rules   Contents


Appendix: Pipeline or AoD First Look

The First Look report created by the pipeline is much longer than what is available under the CLIC First Look widget (Sect. [*]). This report provides detailed and critical information of system performance to the AoD. It shows hence various technical aspects which may be difficult to interpret by non-trained astronomers. Here we briefly describe the content of these plots.

Figure: Pipeline or AoD First Look widget

Image pdbi-cookbook-27


Summary: The first page of the First Look report shows the correlator setup, for the Narrow-band and Widex correlator units8, and the number of correlations obtained on each source. Warnings are often included, and are mainly addressed to the AoDs for technical assessment.


Meteorogical data: See details in Sect. [*]. The verification of the wind speed is particularly important for AoDs if tracking problems are found in the ``Elevation and Azimuth'' plot.


Elevation and Azimuth: See description in Sect. [*]. Colored lines are included to mark flagged scans or records (seconds). Deep blue lines refer to flags due to antenna shadowing, red to correlator unit flags, solid pale blue lines to DATA (correlator related) flags, dashed pale blue lines to tracking problems, green to phase lock losses, yellow to doppler or time issues, and pink to outliers in the system temperature, pointing problems or user or pipeline flagged data. More details about the flags can be found in ``Flag data List'' paragraph presented below.


Pointing and Focus corrections: See description in Sect. [*]. Colored lines are included if pointing or focus corrections change by more than 30% the size of the primary beam or wavelength, respectively.


Antenna Tracking Errors: See description in Sect. [*]. Problems are often related to strong wind. It may happen, though it remains very rare, that large tracking errors indicate a technical problem. This monitoring is particularly important to foresee interventions by the technical staff in NOEMA antennas.


22GHz monitors present the results obtained in time from the 22GHz receiver in each antenna. The first three plots show the counts obtained in each of the three channels of the 22GHz receivers. They are combined to produce the `triple' values, which are used to model the atmosphere, disentangle between water vapor and droplets, and predict the atmospheric phase fluctuations by which the interferometric data are affected. This is then used to reduce of phase decorrelation within each scan. Derived water vapor amounts are displayed in the `Water Monitoring' plots. The given ambient temperatures are obtained by sensors placed close to the 22GHz receivers. Peltier Temperatures are directly related to the receiver performance, and should remain within the plot limits.


CALI scans vs time plots show the time evolution of the CALI autocorrelation scans, which can consist of two (on the hot load and sky respectively) or three (on the cold load, hot load and sky) subscans. Measurements on the cold load are typically performed every $\sim$ 50 min, and are used to derive and monitor receiver temperatures. Differences between antennas are normally linked to receiver attenuations, and differences between units are often linked to correlator tweaks. In correct weather conditions, the sky autocorrelations show a constant value because the signal variations due to airmass change is compensated by correlator tweaking. This tweaking effect is however visible in the autocorrelations on the loads. (Note that Widex tweak levels change in large steps.) Strong variations in the input signal often result in changes in the tweak values, and accordingly in the CALI autocorrelations.


IFPB scans vs time plots present the time evolution of the amplitudes of the IFPB correlations obtained on the noise source (see Sect. [*]). Absolute amplitude values change from unit to unit, depending on the IF phase delay. As the noise diode provides a constant input, IFPB amplitudes should remain constant along a track. Sharp variations in the tweak levels (for instance due to bad weather) can however result in changing amplitudes of the IFPB scans, due to tweaking adjustment effects.


Tweak levels vs time plots show the tweak levels for each of the Narrow correlator units, per correlator sub-band. Values remaining constant (in correct weather conditions) in time correspond to the IFPB scans, tweak values change with the airmass for the other acquisitions.


Monitoring the time, UTC, NTP and PPS. The monitoring of the time synchronization signals is important as it is essential to stop fringes, ensure a good pointing, etc. This plot helps us to monitor differences in time between the UTC (from the maser-synchronized GPS) and NTP ([GPS-]synchronized NTP) times, as well as time offsets between hardware clocks (PPS used by the correlator from maser-synchronized GPS time) and software synchronized events (NTP).


Receiver Temperatures in the IF plots present the receiver temperature computed along the IF bandwidth from one of the first CALI scans including a cold load acquisition (often called `cal-cold' scan). Different colors are used to identify the different correlator inputs. These plots shows better than any other the presence of parasites and tuning features. Particularly, marks are included to identify the known system parasites (from the IF processor at 6300, 4500 MHz, and at 3 and 4 times the LO1ref local oscillator frequencies). Dotted lines represent the values stored in the data header, computed by the online RDI software.


Receiver Temperatures vs Time plots display the evolution of the mean receiver temperature values averaged over the correlator-input8bands. They should not change by more than a few K in projects with stable tunings. Tunings affected by strong parasites may show changing receiver temperatures. An intervention from the frontend group may happen as a consequence of the information extracted from these plots.


Dewar Temperature plots show the temperatures measured at various stages in the cryostat.


Observing List summarizes the sequence of obtained scans. CALI and IFPB acquisitions are ignored in this list.


Flagged data List shows all the flagged records and scans. It should be consistent with the color marks in the 'Elevation-Azimuth' plots.


Total Power vs time: See description in Sect. [*]. Values are presented per each calibration unit (which correspond to correlator inputs8): Two Narrow Quarters and four Widex units. Moderate differences (of a few K) in the values from the different units are due to the different covered frequency ranges.


Cable Phase plots present the phase delays (in degrees) produced at the Master Frequency level by the cables in their movements, mostly due to antenna tracking, for the used and unused bands. Colored lines mark changes that could result in phase variations larger than 30$^{\rm o}$. Significant changes in the LO1ref local oscillator frequencies are also displayed in grey lines.


System Temperatures vs time: See description in Sect. [*]. Plots are shown per calibration unit.


Water Vapor vs time: See first description in Sect. [*]. A plot is created per calibration unit. A differential plot is created to compare the results obtained from all the antennas. Pale blue lines in the WVR H$_2$O plots mark the update of the 22GHz receiver calibration by the online software.


RF phases plots compare the phases obtained per correlator input in the (IF1) frequency band. Remaining phase delays can be identified in these plots. One plot is created for the first correlation scan at the project start, another for the last correlation scan at the project end. Note that if the (frequency averaged) phases from H and V polarization receivers are not equal, ``let phcal `*' '' should be entered after Select (see Sects. [*] and [*]).


Amplitudes for Narrow and Widex correlator inputs: Comparison of the signal level obtained from all the correlator inputs8(or calibration units), which -in principle- should be almost identical. Uncorrected delays result in amplitude differences among the correlator units, which become particularly important between the Widex and the Narrow-band correlator units. Delays can be identified in the RF phases and the RF calibration plots, and the data can be corrected for them by following the instructions in Sect. [*].


next up previous contents
Next: About this document ... Up: pdbi-cookbook Previous: Offline Calibration; the rules   Contents
Gildas manager 2018-09-24