Artuso summarizing the work of R Badman A Borgia PCollins X Llopart R Plackett J Wang Marina Artuso VELO Upgrade Meeting The goal Determine an absolute gain calibration average over the ASIC and per pixel in order to ID: 812823
Download The PPT/PDF document "Test pulse calibration Marina" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Test pulse calibration
Marina Artuso summarizing the work of R. Badman, A. Borgia, P.Collins, X. Llopart, R. Plackett. J. Wang
Marina Artuso VELO Upgrade Meeting
Slide2The goal
Determine an absolute gain calibration (average over the ASIC and per pixel) in order to:Determine TOT to charge calibration and produce Landaus which give an absolute measure of the collected charge for parametric studies (sensitivity to HV, thresholds…)Determine the minimum detectable charge at any given threshold (input to simulation)Verify influence of proper charge weighting on resolution and more realistic comparison with simulation
2
Marina Artuso VELO Upgrade Meeting
Slide3The problems & caveats
We had several of them:Timing of the test pulse with respect to the shutterIdentify the proper pattern to avoid cross talk (spacing 16 seems to be ok)Absolute value of the test pulseReliability of the threshold scan Best data so far taken by Xavi, will be the basis of the discussion today, much more on the failed attempts on a future presentation
Note: these results are a subset of the studies undertaken at CERN and SU, more systematic studies to come
3
Marina Artuso VELO Upgrade Meeting
Slide4The test pulse offset
Measured from difference between anin_n02 and anin_no4 & compare with
tpulse
value from
daq
program
Obtained 35.3 mV
Test_in
has Ct=7.5fF (qt=46.875
e
-
/mV)
4
Marina Artuso VELO Upgrade Meeting
Slide5The non-linear response
Edge of response to negative
tp
for
thl
= 420
Separation depends upon offset and 2*
thl
5
Marina Artuso VELO Upgrade Meeting
Slide6Fitting for the surrogate function
Use (see http://aladdin.utef.cvut.cz/ofat/Methods/TimePixCalibration/index.htm)
6
Marina Artuso VELO Upgrade Meeting
Slide7Gain Curves for THL = 400
Marina Artuso VELO Upgrade Meeting
7
Data taken by
Xavi
, 10
testpulses
per setting, readout sum of 10
ToTs
. Thus efficiency curve can not be produced.
Each point is an average over all pixels. Error bar represents the spread.
D04-W0015
Data from
Xavi
Slide8Fit Parameter Distributions for THL=400
Marina Artuso VELO Upgrade Meeting8
Gain (count/mV)
ToT
0
(count)
C Parameter (
count
mV
)
T Parameter (mV)
D04-W0015
Data from
Xavi
Slide9Threshold Determined from Calibration
Threshold determined from gain curve (intersect with TOT=0). The value is slightly lower than the true value due to noise.The mean threshold is 33.74x46.875
= 1582
e
Assuming
THLeq
=460 from noise center method, THL=400 corresponds to ~1500 e, consistent with the calibration.
Threshold spread is ~ 36 e.
9
THL = 400
Threshold ( x46.875 e )
Number of Pixels
D04-W0015
Data from
Xavi
Marina Artuso VELO Upgrade Meeting
Slide10Summary of the fit results at 3 different thresholds
THLgain
tot0
c
t
Thr
(
e
-)
420
0.21±0.008
21
±2.2
123.4±43.4
41.8±1.7
980
400
0.21
±0.007
30.6±1.8
134.8±41.9
30.2±1.5
1580
380
0.19±0.007
47.3±1.8
187.1±44.7
16.6±1.0
2090
10
Marina Artuso VELO Upgrade Meeting
Slide11Alternative threshold determination
Threshold can be determined from center of efficiency curve (n-of-nonzero TOT wrt tp).
Threshold can also be calculated from
ToT
vs
testpulse
charge
curve
.
(first
non-zero TOT)
. From
this method the threshold is measured to be (33.74x46.875=1581 e), which is 32 e lower than what is obtained from efficiency curve mode.
Marina Artuso VELO Upgrade Meeting
11
Threshold(efficiency) [mV]
Threshold(gain) [mV]
Difference (gain - efficiency) [mV]
D04-W0015
Data from Alessandra
Slide12Mean Threshold determined from efficiency curve
thl=400 (Xavi)
12
Marina Artuso VELO Upgrade Meeting
Slide13Conclusions
We have inverse surrogate functions for all the pixels of the DUT studied in test beams 2009 & 2010We have a reasonable determination of the minimum detectable charge at various thresholds, note that a threshold of about 750 e- tb2009 data are dominated by noiseThis information is now available for relevant analyses
Marina Artuso VELO Upgrade Meeting
13