ACDC Analysis

The is the main coordination page for analysis of ACDC Data. Please feel free to edit/update this page:

• Adding your name next to the analysis you are performing [want to perform]
• Edit the datasets suggested here with the list of runs you find to be more appropriate
• Embelish the description of the analysis for clarity and completeness
• Adding additional analysis topics

Charaterisation of detector perfromance vs. bias voltage

Description:	To study the signal/noise performance, charge sharing as a function of bias voltage. To check for variation of behaviour across the sensor.
Analysts:	Eugenia Puccio, ...	Status:	WRITING CODE
Links:	Web/TWiki page	Presentation	Documentation

| Suggested datasets: | M18R, normal incidence: 1223-1312
M18P, normal incidence: 820-827
M19R, normal incidence: 859 - 873
M19P, normal incidence:1604-1619
M20R, normal incidence: 1550-1574
M20P, normal incidence: 1577-1603
M18R, beam at -10 degree: 1495-1547 |||

Characterisation of detector efficiency

Description:	It is interesting to study the efficiency map across the sensor surface. There may be regions which are not fully depleted even when the sensor as a whole appears depleted. These regions can appear as circles, due to the wafer manufacture, or can appear along the strips. There is also an inefficient region in the phi sensor between the inner and outer sectors which it would be good to characterise as the exact inefficiency is not known. (There are of course similar inefficient regions in the R sensors but in the testbeam we read out by quarters so we are not sensitive to these). This analysis needs the ability to fit a track in 5 sensors and extrapolate to a position on the testsensor. A good alignment file has to be available for the data under study. Probably the function to translate NZS to ZS data is essential, so as to be able to use the tracking. The study could be extended to look at clustering, if the raw values of the ADCs around the intercept region can be compared to the clusters which were found by the clustering algorithm(s). And the study is always more interesting when done as a function of bias voltage. Finally, it would be very interesting to study the detailed efficiency, when different algorithms to apply common mode are used, linking beetle channels, or inner and outer regions for instance, and taking account of the cross talk from the Beetle header. See presentation on phi sensor noise performance in here.
Analysts:		Status:	WRITING CODE
Links:	Web/TWiki page	Presentation	Documentation

| Suggested datasets: | M18R, normal incidence: 1223-1312
M18P, normal incidence: 820-827
M19R, normal incidence: 859 - 873
M19P, normal incidence:1604-1619
M20R, normal incidence: 1550-1574
M20P, normal incidence: 1577-1603
M18R, beam at -10 degree: 1495-1547 |||

Characterisation of Signal and Noise

Analysts:	Karol Hennessy	Status:	Ongoing
Links:	Web/TWiki page	Presentation	Documentation
Description:	To characterise and reduce the noise from the sensors through the Tell1. This will also be compared to the Tell1 system at Liverpool. Once this is completed, a signal to noise measurement will be made (probably using ACDC3 data)
Suggested datasets	Using random Non-zero suppressed data.

Characterisation of DAQ performance vs. Beetle parameters

Analysts:		Status:	-
Links:	Web/TWiki page	Presentation	Documentation
Description:	To characterise the detector perfomance under the varying Beetle configuarations used dunig the testbeam
Suggested datasets	Runs 886-1219, Runs 1313-1383, Runs 1421-1450.

Timing analysis

Analysts:	Kazu Akiba,...	Status:	WRITING CODE
Links:	Web/TWiki page	Presentation	Documentation
Description:	To study the pulse shape and characterise the rise-time and spill-over effects
Suggested datasets	Run 367-562, M19R not function. Run 708-725, M19R may not function Run 1673-1705, downstream scintillation counters included.

Noise and pedestal behaviour

Analysts:	Marina Artuso, JC Wang, Ray Mountain	Status:	-
Links:	Web/TWiki page	Presentation	Documentation
Description:	To study signal/noise and cross-talk at different places on the sensor (esp. small-pitch), and overspill. Investigate chronic evolution of common mode over the test-beam period
Suggested datasets :	We took random trigger runs withbeam, inter-spill, or no beam throughout the whole testbeam period.

Evaluation of resolution performance

Analysts:		Status:	-
Links:	Web/TWiki page	Presentation	Documentation
Description:	To measure the resolution of the detectors as a function of pitch. The basic idea is to plot the residuals for tracks fitted through the 3 sensors and from this extract the sensor resolution. It can be shown that for three equally spaced sensors, with tracks passing through regions of the same pitch, which is the testbeam case, the expected residuals have a width equal to the detector resolution divided by sqrt(2/3) for the middle sensor, and by sqrt(1/6) for the external sensors. So considering for instance the 40 micron regions, we expect to see residuals of around 5 micron in the central sensor in the finest pitch region. The requirements for the basic analysis are: a data set with all detectors biased to the nominal settings, and the provision of a good alignment file for that data set. The mean of the residuals as a function of phi and R as well as the width, should of course be checked. It is suggested that NZS data is used for the study of the phi sensors, so the function to translate NZS to ZS data in VETRA is needed, so the tracking can be used. For the R sensors there should be no problem in using NZS or ZS data, as even though the reordering was not done, adjacent Beetle channels still represent adjacent strips. The next suggested step in the analysis is to run the simulation, and compare the results with a simulated testbeam setup directly with the data. This should provide input on the simulation tuning. The analysis can be later refined to include the angled data, so that the detector precision can also be broken down as a function of angle. For this it is suggested firstly to concentrate on the R sensors, as the rotation of the box causes a rotation relative to the strips which is greatly reduced in the case of the phi strips, which run almost perpendicular to the beam direction. It is suggested to refer to LHCb notes 2000-103 and 2000-99 where similar analyses were done. The function to calculate the effective track angle relative to a chosen strip with a particular orientation was written by David Petrie but not yet committed but he could be contacted directly.
Suggested datasets :	-

Evaluation of phi sensor cross talk

Analysts:		Status:	-
Links:	Web/TWiki page	Presentation	Documentation
Description:	The phi sensor, with its very strange strip ordering, provides a nice testbed for studying cross talk. The normal intersymbol cross talk seen in the data applies to adjacent (forward) Beetle channels. In the phi sensor there are also other classes of clusters, i.e. adjacent strips in the outer region which are not adjacent in Beetle channels, adjacent strips in the inner region which are not adjacent in Beetle channels, strips in the inner region and the corresponding strip in the outer region which the routing line of the inner strip passes over, but are not adjacent Beetle channels. Study of these different classes allows one to disentangle cross talk in the detector from cross talk in the Beetle. This effect has been partially studied with testbeam data, see http://indico.cern.ch/conferenceDisplay.py?confId=a053225 and there was a thorough study done with testpulse data see http://indico.cern.ch/materialDisplay.py?contribId=7&materialId=slides&confId=5148. The effect should now be studied with our new copious testbeam data. The principal tool needed is the ability to fit the track through 5 sensors and extrapolate to the phi sensor under study. Then one can pick those events where the track intersects a strip, and look to see if there is signal in the adjacent strip. Strips should be chosen where the adjacent strip is not an adjacent Beetle channel. Also the alignment mean should be well below the strip pitch. The easiest way to ensure this is to look in the wide pitch regions of the inner and outer region. The second study would be to choose tracks which hit an inner strip and look for pickup in the outer strip which underlies the readout line of the inner strip. Again, for those combinations where the Beetle channels are not adjacent. Then to do the reverse, i.e. look for intersections with outer strips and look for pickup in the routing line. The list of strips and the corresponding software and beetle numbering is needed. It is provided as an attachment on the bottom of this page.
Suggested datasets :	NZS data, with the Vetra translation to ZS, so as to have the tracking available. 0 degrees, all 6 sensors functional. Study as a function of bias voltage in the test sensor very welcome.-

Evaluation of threshold settings

Analysts:		Status:	-
Links:	Web/TWiki page	Presentation	Documentation
Description:	This is really an analysis mainly done in the simulation, using the testbeam data as a cross check. The idea is to find the best settings of the thresholds to accept clusters so as to accept tracks from b hadrons with 100% efficiency and reject spillover tracks. The signature of a spillover track is that the signal will be on average lower than for a real track. So if you demand that a certain fraction of clusters on a track pass a second threshold, you can reject spillover tracks. A bit exists in the cluster to signal this second threshold. https://edms.cern.ch/file/637676/2/ClFormat.pdf The idea would be to select the cuts on the data, and then use real testbeam tracks. Multiply the charge of the clusters on the track by 30%, or whatever the spillover is measured to be for the standard Beetle settings, and check that you still have 100% efficiency to select these real tracks
Suggested datasets :	NZS data, with the Vetra translation to ZS, so as to have the tracking available. 0 degrees, all 6 sensors functional.

Evaluation of TELL1 algorithms

Analysts:	Aras Papdelis, Ann Van Lysebetten	Status:	-
Links:	Web/TWiki page	Presentation	Documentation
Description:	To check the performance of the algorthms running on the TELL1
Suggested datasets:	Runs 1576, 1620-1632, 1604-1668

Performance of the clustering

Analysts:	JC Wang, Marina Artuso	Status:	-
Links:	Web/TWiki page	Presentation	Documentation
Description:	To study the clustering: Performance vs. cluster-threshold; charge-sharing (att. re-ordering!) in clusters as a function of tracks' incidence angle. Compare offline clustering with TELL1 clustering algorithm.
Suggested datasets:	The angle runs are: 0 degree: 754-773, 815-819 5 degree: 774-793 10 degree: 794-813 16 degree: 726-751 -10 degree: 1473-1494

Evaluation of the tracking and alignment

Links:	Web/TWiki page	Presentation	Documentation
Analysts:	Tomas Lastovicka, Seb Viret,...	Status:	ANALYSIS UNDERWAY

| Description: | To quantify the performance of the tracking and the alignment algorithms on real data. Align the three modules and study track residuals and tracking efficiencies as a function of radius, strip-pitch and incidence angle.

There seems to be a major large overlap with 'Evaluation of resolution Analysis'.

Suggested datasets:	Suggested runs: first start with 0 degree runs to understand the data, continue with angle runs as in previous one and interaction setting (1704-1891)

Vertexing

Analysts:	Tomas Lastovicka, Aras Papadelis, ...	Status:	Not started yet
Links:	Web/TWiki page	Presentation	Documentation
Description:	To recontruct multiple-track events and recontruct their vertex.
Suggested datasets:	Interaction runs: 1704-1891