First look at Slow Control DB LEM numbering for 3x1x1 Yuriy - - PowerPoint PPT Presentation

First look at Slow Control DB

LEM numbering for 3x1x1 Yuriy Onishchuk, Vladislav Kaluzhnyy

Kiev Taras Shevchenko National University

25-Aug-2016

Yuriy Onishchuk, Vladislav Kaluzhnyy (Kiev Taras Shevchenko National University) First look at Slow Control DB 25-Aug-2016 1 / 19

General remarks

General remarks

Slow Control DB:

– ≈1500 sensor values – Installation and arrangement in progress

Qscan needs in inpup parameters for each run The most important are the high voltages (drift, extraction, amplification and induction). But also geometrical parameter such as LEM thickness, etc. Permanent access needs to the DB during run time to retrieve those parameters Easy access to the run conditions using dedicated classes in the Qscan code (including getters like as LAr->GetDrift(), LEM->GetHVUp(), GAr->GetPressure(), GAr->GetTemperature(), etc.)

Topics

LEM indication & numbering First DB test with VMONxxxx sensors Thickness & Gain variations for installed LEMs

Yuriy Onishchuk, Vladislav Kaluzhnyy (Kiev Taras Shevchenko National University) First look at Slow Control DB 25-Aug-2016 2 / 19

LEM indication & numbering

LEM & Anode indication Anode-PCB numbers (TE0xx) LEM-PCB numbers LEM numbering Each LEM has unique number (PCB number) marked on manufacturing stage 15 LEMs have been previously tested within Argon atmosphere: 1-5, 8, 10-12, 14-17, 22, 23. LEMs with PCB numbers 3, 14, 23 were removed

Yuriy Onishchuk, Vladislav Kaluzhnyy (Kiev Taras Shevchenko National University) First look at Slow Control DB 25-Aug-2016 3 / 19

LEM disposition numbering

LEM-CRP (position numbers) indicated: 1–12. LEM-CRP numbers can be supposed as basic VMONxxxx sensors (HV applied to TOP and BOTTOM electrodes) indicated (R, G, B colors) Coordinate system: axes Y along 3 m, X – 1 m.

Yuriy Onishchuk, Vladislav Kaluzhnyy (Kiev Taras Shevchenko National University) First look at Slow Control DB 25-Aug-2016 4 / 19

LEM cabling & power supply

VMONxxxx sensors (TOP and BOT) connected to boards: 00, 02, 04. Board 08 is empty This is in contradiction with info on previous slide: used boards 00, 04 and 08 TOP, BOT sensor numbering based on 0, 4, 8 boards usage. VMON02xx numbers are absent in the DB field list at current time Discussion with Sebastien clarified this problem: the right scheme is 00, 02, 04. Field names in DB will be corrected during next 7-10 days Another proposition is to replace VMON on LEM mapping based indication: L_xx_up and L_xx_down with xx corresponding to LEM position numbers: 01, 02, . . . , 12

Yuriy Onishchuk, Vladislav Kaluzhnyy (Kiev Taras Shevchenko National University) First look at Slow Control DB 25-Aug-2016 5 / 19

Top, Bot LEM sensors

Summarization of LEM numbering Indicated numbers:

LEM-CRP layout or simply LEM numbers (LEM1, LEM2, . . ., LEM12) LEM-PCB numbers TOP , BOT sensor numbering according to DB

Qscan coordinate system (X = −1.5 . . . 1.5 m, Y = −0.5 . . . 0.5 m) used

Yuriy Onishchuk, Vladislav Kaluzhnyy (Kiev Taras Shevchenko National University) First look at Slow Control DB 25-Aug-2016 6 / 19

DB test

DB named “wa105_sc” located on host “wa105cpu0001.cern.ch” Possible access to DB using mysql commands. Session started with command mysql -u wa105 -h wa105cpu0001.cern.ch -p wa105_sc Usage of C API functions is more proper to design a programming tool for connecting to Qscan (see Thierry’s examples and Yuriy’s C and Python codes in testdb directory) Python was used to get more quick and carefull DB analysis. Package _mysl consist of identical to C API functions. Matplotlib was used to visualize results

Yuriy Onishchuk, Vladislav Kaluzhnyy (Kiev Taras Shevchenko National University) First look at Slow Control DB 25-Aug-2016 7 / 19

DB test

Time evolution of LEM sensor values from 2016-08-04 to 2016-08-15 12:00:00 shown VMONxxxx sensors were connected at the begin of August and than switched off Design of GUI-ROOT based display is in progress now

Yuriy Onishchuk, Vladislav Kaluzhnyy (Kiev Taras Shevchenko National University) First look at Slow Control DB 25-Aug-2016 8 / 19

LEM thickness measurement

LEM thickness measurements have been done in May 2016 Files with results available on https://edms.cern.ch/document/1682958/1 121 points (11 × 11) automatically processed in the x − y plane with step 50 mm LEM x−y plane corresponds to the YZ plane on the figure Double LEM side measurement procedure was based on ZEISS PRISMO ULTRA mashine usage

Yuriy Onishchuk, Vladislav Kaluzhnyy (Kiev Taras Shevchenko National University) First look at Slow Control DB 25-Aug-2016 9 / 19

LEM thickness measurement

Info about 13 LEM-PCBs (1-6, 8-14) is available Thickness measuremens have been done for 8 of 12 LEM installed only Files in directories, plotsROOTformat and fwdselectionlems, contained only 1D-historgams and 1D-, 2D-figures Solution:

– Usage of files with raw results to reconstruct XY-matrix – Expample: file LEM_1_fet.txt containes raw (fet) results of LEM-PCB 1 (LEM-CRP 3) processing. Totally 238 rows (double side measurement for 119 points): Yfet ↔ X, Zfet ↔ Y, Xfet ↔ Z – X, Y values for both surfaces averaged – Difference of Z values gives thickness – Result: 119 scattered 3D point obtained for LEM PCB 1 sample

Yuriy Onishchuk, Vladislav Kaluzhnyy (Kiev Taras Shevchenko National University) First look at Slow Control DB 25-Aug-2016 10 / 19

LEM-PCB 1 (LEM-CRP 3). Results comparison

Comparison with 2D-histogram from the file LEM1_2D.txt Lack of 4 points in 11 × 11 matrix Probably 2 points loss is a result of the bug in ROOT-visualisation of 2D graphs and histograms (min and max values points not shown)

x [mm]

100 200 300 400 500

y [mm]

• 500
• 400
• 300
• 200
• 100

1.04 1.06 1.08 1.1 1.12 1.14 1.16

LEM-PCB 1 data analysis (bottom plots):

Lack of 2 points only in 11 × 11 matrix (usage of special trick to show all points) Figure LEM_1 (TH2 & palette) use the same palette and visually is very similar to upper plot Variant of visualisation used TGraph2D and surf interpolation shows a possibilitiy to get values between XY-knot positions

X [mm]

100 200 300 400 500

Y [mm]

500 − 400 − 300 − 200 − 100 − 1.04 1.06 1.08 1.1 1.12 1.14 1.16

LEM_1 (TH2 & palette)

X [mm]

50 100 150 200 250 300 350 400 450

Y [ m m ]

450 − 400 − 350 − 300 − 250 − 200 − 150 − 100 − 50 −

Thickness [mm]

1.11 1.12 1.13 1.14 1.15 1.16

LEM_1 (TGraph2D & surf1) Yuriy Onishchuk, Vladislav Kaluzhnyy (Kiev Taras Shevchenko National University) First look at Slow Control DB 25-Aug-2016 11 / 19

Effective Gain

Effective electron gain, Geff , can be expressed as: Geff (E0, ρ) = T eα(ρ,E0)x where E0 = V/d – the electric field, depends on applyed voltage V to a LEM of the insulator (FR4) thickness d ρ – the gas density T – the electrical transparency of the chamber, T = 1 x ≈ 0.7d – the effective amplifiction length α(ρ, E0) – the 1st ionisation Townsend coefficient: α(ρ, E0) = Aρ e−Bρ/κE0 can be parametrized by Aρ = 7339 cm−1, Bρ = 183 kV/cm for pure argon at 87 K and 0.98 bar within the range between 20 and 100 kV/cm The coefficient κ = 0.95 determines the averaged effective field within the LEM gap d value can be estimated as difference of measured total thickness, dtot and cooper double side layer, dCu: d = dtot − dCu Averaged etstimation gives, dCu = 140µm, according to FR4 and cooper ELTOS (manufacturer) and CERN FR4 measurements

Yuriy Onishchuk, Vladislav Kaluzhnyy (Kiev Taras Shevchenko National University) First look at Slow Control DB 25-Aug-2016 12 / 19

Thickness & Gain variations for LEM 1

LEM-CRP 1 (LEM-PCB 2) 2D distributions for measured LEM total thickness and the electron gain Relative thickness variations, d/d, is about 1.5-2% Relative gain variations, G/G, is about 10-15% See upload slides for 8 of 12 installed LEMs

Yuriy Onishchuk, Vladislav Kaluzhnyy (Kiev Taras Shevchenko National University) First look at Slow Control DB 25-Aug-2016 13 / 19

Conclusion

Conclusions

LEM PCB and CRM numbering have been cross-checked VMON sensor boards: 0,2,4 or 0,4,8 will be corrected 1st test of WA105 DB fields have been done using mysql and C API functions Time evolution of LEM sensors (VMONxxxx) shown Design of GUI-ROOT based display is in progress Thickness & Gain variations for installed LEMs was analysed Many thanks to Sebastien and Thierry for fruitful discussions

Yuriy Onishchuk, Vladislav Kaluzhnyy (Kiev Taras Shevchenko National University) First look at Slow Control DB 25-Aug-2016 14 / 19

Upload slides

Yuriy Onishchuk, Vladislav Kaluzhnyy (Kiev Taras Shevchenko National University) First look at Slow Control DB 25-Aug-2016 15 / 19

Thickness & Gain variations for installed LEMs (1)

LEM-CRP 1 (LEM-PCB 2) LEM-CRP 2 (LEM-PCB 11)

Yuriy Onishchuk, Vladislav Kaluzhnyy (Kiev Taras Shevchenko National University) First look at Slow Control DB 25-Aug-2016 16 / 19

Thickness & Gain variations for installed LEMs (2)

LEM-CRP 3 (LEM-PCB 1) LEM-CRP 4 (LEM-PCB 4)

Yuriy Onishchuk, Vladislav Kaluzhnyy (Kiev Taras Shevchenko National University) First look at Slow Control DB 25-Aug-2016 17 / 19

Thickness & Gain variations for installed LEMs (3)

LEM-CRP 5 (LEM-PCB 8) LEM-CRP 7 (LEM-PCB 13)

Yuriy Onishchuk, Vladislav Kaluzhnyy (Kiev Taras Shevchenko National University) First look at Slow Control DB 25-Aug-2016 18 / 19

Thickness & Gain variations for installed LEMs (4)

LEM-CRP 8 (LEM-PCB 12) LEM-CRP 10 (LEM-PCB 10)

Yuriy Onishchuk, Vladislav Kaluzhnyy (Kiev Taras Shevchenko National University) First look at Slow Control DB 25-Aug-2016 19 / 19