- Andreas Psallidas made a status report on a GEANT4 simulation of the experimental setup which he and his colleagues work on at NTU.
The first code is already available in github https://github.com/apsallid/SiWEcal. It produces very useful visualization of the setup, shown on the slides.
The code runs on lxplus in CERN, and visualization was also tested on Windows 10 (with putty and xming). At the moment, we are discussing with LLR engineers the right dimensions and the placement of various setup elements in the beam line, so the exact dimensions will be put later. The technical drawings we collect at https://twiki.cern.ch/twiki/bin/view/Main/Physics_program, section "Set-up's".
GEANT4 simulation is a very useful initiative from NTU, the Monte Carlo will be very much needed in every physical analysis of the showers.
We'll need MC for about 10 different setup configurations (various absorber / rotation angles).

- Kostya Shpak has found first square events: about 4 events per a layer from about 50k physical events. This was a 100 GeV electron run with the beam shooting at the center of PCB, ie. in the corners of 4 sensors. His selection criteria may be further improved in the future, so, the observed rate of square events may be considered as a lower limit.
A small reminder on "square" events. The conductive guard ring at the sensor periphery smooths the electric field gradient, which reduces the dark currents and improves HV stability. "Square" events are produced when there is a big energy deposition at the guard ring area. Due to a capacitive coupling between the guard ring and the peripheral pixels, the latter may all be fired, this is called the "square" event. In CMS this effect may be avoided by connecting the guard ring to a fixed potential using a wire bonding. In CALICE design we used gluing for pixels instead of wire bonding, but the glue can not be used to connect about 0.5 mm wide guard ring (0.5 mm is too thin for the glue dot, there will be shorts with the adjacent pixels). So, the potential of the guard ring is floating, which is the reason of the "square" events. After segmenting the guard ring, the rate of "square" events is much reduced, however.
Kostya also has found a strong source of square events: it is when we turned our setup by 90 degrees  and shot the beam in DIF1. The energy depositions in the guard ring in this case sometimes were much larger, so he found already 170 square events.

- Vladik has continued his studies of pedestals. The effect of pedestal splitting depending on the following retriggers has been confirmed in all chips of the setup, currently only in SCA1 and in one muon run (361). The splitting is larger in the upper chip channels (>40). No pronounced X-Y patterns are observed. The splitting is 2-10 in DIF0,2 and much larger, 10-25 in DIF1. RMS of both pedestal peaks are similar. It seems that the chip has two states, and the state determines both the pedestal position and the following retriggers in BX+1,+2,... Pedestals in all channels shift synchronously, depending on the state. The correlation pedestal - retrigger is rather high (almost 100% in DIF0).

Kyushu continues studies of their baby sensors and also reported an spill number mismatches (up to about 5) between the baby sensors and the main setup.
Vladik has asked about the possibility to test at Kyushu two newly observed effects with the SKIROC test board:
- trigger shift from BX to BX+1 under high load in the chip;
- splitting of pedestals depending on the following retriggers.
This may be especially important before the coming submission of a new SKIROC production.

Next meeting is this week, on 14 Jan, same time, 10:30 CET.