Timing Performance of a Micro-Channel-Plate Photomultiplier Tube
Jonathan Bortfeldt, Florian Brunbauer, Claude David, Daniel Desforge, Georgios Fanourakis, Michele Gallinaro, Francisco Garcia, Ioannis Giomataris, Thomas Gustavsson, Claude Guyot, Francisco Jose Iguaz, Mariam Kebbiri, Kostas Kordas, Philippe Legou, Jianbei Liu, Michael Lupberger, Ioannis Manthos, Hans Müller, Vasileios Niaouris, Eraldo Oliveri, Thomas Papaevangelou, Konstantinos Paraschou, Michal Pomorski, Filippo Resnati, Leszek Ropelewski, Dimitros Sampsonidis, Thomas Schneider, Philippe Schwemling, Emmanuel Scorsone, Lukas Sohl, Miranda van Stenis, Patrik Thuiner, Yorgos Tsipolitis, Spyros Eust. Tzamarias, Rob Veenhof, Xu Wang, Sebastian White, Zhiyong Zhang, Yi Zhou
TTiming Performance of a Micro-Channel-Plate Photomultiplier Tube
J. Bortfeldt a,1 , F. Brunbauer a , C. David a , D. Desforge b , G. Fanourakis c , M. Gallinaro d , F. Garc´ıa e , I. Giomataris b , T. Gustavsson f ,C. Guyot b , F.J. Iguaz b,2 , M. Kebbiri b , K. Kordas g , P. Legou b , J. Liu h , M. Lupberger a,3 , I. Manthos g , H. M¨uller a , V. Niaouris g ,E. Oliveri a , T. Papaevangelou b , K. Paraschou g , M. Pomorski i , F. Resnati a , L. Ropelewski a , D. Sampsonidis g , T. Schneider a ,P. Schwemling b , E. Scorsone i , L. Sohl b, ∗ , M. van Stenis a , P. Thuiner a , Y. Tsipolitis j , S.E. Tzamarias g , R. Veenhof k,4 , X. Wang h ,S. White a,5 , Z. Zhang h , Y. Zhou h a European Organization for Nuclear Research (CERN), CH-1211 Geneve 23, Switzerland b IRFU, CEA, Universit´e Paris-Saclay, F-91191 Gif-sur-Yvette, France c Institute of Nuclear and Particle Physics, NCSR Demokritos, 15341 Agia Paraskevi, Attiki, Greece d Laborat´orio de Instrumenta¸c˜ao e F´ısica Experimental de Part´ıculas, Lisbon, Portugal e Helsinki Institute of Physics, University of Helsinki, 00014 Helsinki, Finland f LIDYL, CEA-Saclay, CNRS, Universit´e Paris-Saclay, F-91191 Gif-sur-Yvette, France g Department of Physics, Aristotle University of Thessaloniki, Thessaloniki, Greece h State Key Laboratory of Particle Detection and Electronics, University of Science and Technology of China, Hefei 230026, China i CEA-LIST, Diamond Sensors Laboratory, CEA-Saclay, F-91191 Gif-sur-Yvette, France j National Technical University of Athens, Athens, Greece k RD51 collaboration, European Organization for Nuclear Research (CERN), CH-1211 Geneve 23, Switzerland
Abstract
The spatial dependence of the timing performance of the R3809U-50 Micro-Channel-Plate PMT (MCP-PMT) by Hamamatsu wasstudied in high energy muon beams. Particle position information is provided by a GEM tracker telescope, while timing is measuredrelative to a second MCP-PMT, identical in construction. In the inner part of the circular active area (radius r < reference detectors has been validated. Keywords:
MCP-PMT, time resolution, Cherenkov light, beam test, t0-reference, Monte-Carlo simulation
PACS:
1. Introduction
Reliable reference detectors with high time resolution areneeded in the characterization of new detector technologiesaiming at performing ultra precise time measurements. An ex-ample is the PICOSEC-Micromegas [1, 2] concept, a newly in-troduced Micropattern Gaseous Detector for fast timing appli-cations. The time resolution of several detector prototypes hasbeen studied in great detail using minimum ionizing particle(MIP) beams at the CERN SPS secondary beam lines.There are di ff erent types of reference timing detectors pro-viding less than 10 ps time resolution. One possible option aresilicon based detectors like SiPMs that have shown a timingperformance in this range [3]. Another detector technology ∗ Corresponding author
Email address: [email protected] (L. Sohl) Now at Ludwig-Maximilians-University Munich, Germany Now at Synchrotron Soleil, BP 48, Saint-Aubin, 91192 Gif-sur-Yvette,France Now at Physikalisches Institut, Universit¨at Bonn, Germany Also at National Research Nuclear University MEPhI, Kashirskoe High-way 31, Moscow, Russia; and Department of Physics, Uluda˘g University,16059 Bursa, Turkey. Also at University of Virginia with good timing response are MCP-PMTs. Those detectorsare commonly used in various fields. One example is the usefor time-of-flight positron emission tomography (PET). Otherstudies have shown a coincidence time resolution (CTR) of30ps FWHM [4]. In this work, two MCP-PMTs of type Hama-matsu R3809U-50 Micro-Channel-Plate Photomultiplier Tubes(MCP-PMT) [5] have been studied for the beam test measure-ment of the PICOSEC-Micromegas.Particles enter the MCP-PMT, traverse a radiator and gen-erate Cherenkov light, which is then converted to charge in amultialkali photocathode between the radiator and multichan-nel plate. The radiator consists of a 3.2 mm thick synthetic sil-ica window that is integrated in the MCP-PMT. The useablephotocathode diameter is 11 mm as indicated. No further infor-mation about the photocathode and the window is given by themanufacturer. For the further simulation a generic Cherenkovangle of 45 ◦ is assumed.These MCP-PMTs provide short signals with a rise time of 160 ps. Signals of this kind are well suited as a time ref-erence for fast-timing detector studies. As discussed in a re-cent review of state-of-the-art timing detectors [6], it is ad-
10 % to 90 % amplitude
Preprint submitted to Nuclear Instruments and Methods A February 17, 2020 a r X i v : . [ phy s i c s . i n s - d e t ] F e b antageous for beam measurements to use a reference detectorwith time resolution significantly better than that expected forthe detector under study. Measurements demonstrated the suit-ability of MCP-PMTs as a (t ) timing reference for PICOSEC-Micromegas fast-timing detectors [7]. In this manuscript, wediscuss further investigations, aiming at understanding the spa-tial dependence of the time resolution over the surface of thephotocathode. The propagation of Cherenkov light includingreflection and absorption has been modeled analytically andwith a Monte-Carlo simulation. The observed radial depen-dences of the mean signal charge and time resolution have beencompared to the modeled fraction of Cherenkov light reachingthe photocathode. In the following, the di ff erent models as wellas the measurement set-up will be explained and the measureddata will be compared to the results from the simulation.
2. Modeling of Cherenkov Light Propagation in the Radia-tor
When a particle passes through the MCP-PMT at a largerdistance from the photocathode center, less light reaches thephotocathode as the Cherenkov cone is not fully projected ontothe photocathode. Figure 1 shows a sketch of the Cherenkovcones and the photocathode with the assumed dimensions ofthe detector. In our model, Cherenkov photons can be eitherconverted to charge at the photocathode, or can be reflected orabsorbed at the radiator boundaries. Part of the reflected lightcan later reach the photocathode and contribute to the detectorsignal. An analytic and a Monte-Carlo model have been devel-oped to predict the amount of light reaching the photocathode.
Figure 1: Schematic Cherenkov light cone propagation in the radiator, as as-sumed in the model. The green colored cones (diagonal hatching) are fullyprojected onto the photocathode, while the projection of the red coloured cones(vertical hatching) are fully outside of the photocathode area. The radiator hasa thickness of 3.2 mm, and the photocathode a diameter of 11 mm.
A geometric calculation of the overlapping areas of photo-cathode and Cherenkov cone for particle impact points at di ff er-ent radii has been done, with and without considering reflectionbetween the radiator and the photocathode.Figure 2 shows a sketch of the geometrical overlap of thephotocathode with the Cherenkov light. In this example, theperpendicularly incident particle is hitting the edge of the pho-tocathode area (thick black line). The blue area (horizontal hatch) shows the Cherenkov light directly reaching the photo-cathode and the red area (vertical hatch) shows the light fromthe first-order reflection reaching the photocathode. This lightwill have survived one reflection on the photocathode and oneon the opposite side of the radiator crystal before reaching thephotocathode. A part of the light can be transmitted and lostat each reflection. This loss is modelled by a weighting factor( w <
1) when estimating the amount of light reaching the redarea (vertical hatch).The amount of light reaching the photocathode decreaseswith increasing particle impact radius with respect to the photo-cathode center, due to the decreasing geometrical overlap. Withthis geometric model the relative amount of photons reachingthe photocathode depending on the radius ( P rel ( r )) is calculatedby: P rel ( r ) = A dir. ( r ) + w · A ref. ( r ) A dir. (0) + w · A ref. (0) (1)where A dir. ( r ), the area of direct light, and A ref. ( r ), the area ofreflected light, depending on the radius, and w is the weightingfactor for the loss of the reflected light. This function is scaledto the mean signal charge in the center of the MCP-PMT and w is a free parameter in the fit to the data. Figure 2: Sketch of the geometrical overlap between Cherenkov cone, in bluehorizontal hatching and violet diagonal hatching, and first-order reflected light,in red vertical hatching and yellow dotted shaded areas, with the photocathodearea, as a thick black line.
The former analytical geometrical calculation describes theradial behavior of the signal amplitude well for small and largeradii but is lacking precision for medium radii. Therefore, asimulation of the light propagation and conversion to charge inthe fused silica radiator has been carried out to model the radialamplitude behavior. For this simulation the photons are createdas two dimensional points on the window surfaces as objects ina C ++ program. A random generator with a probability thresh-old decides, for each point, if the photon is reflected on the sur-face or not. If it is reflected, the new position on the opposite2ide of the window with respect to the Cherenkov angle is cal-culated. In the initial step, 25 · points have been randomlydistributed over the radiator surface that carries the photocath-ode. Photons in the photocathode region can be either reflectedor, if not reflected, they can be converted to charge or absorbedand thus lost. Outside the photocathode region, photons are lostafter not being reflected. The same holds for photons on the op-posite side of the radiator, i.e. at the air-radiator interface. Allthese e ff ects have been taken into account for each point in thesimulation.The results of this simulation give the x and y coordinatesof the point where each photon is generated as well as where itleaves the crystal, either by transmission or absorption on thecrystal surface. Another indicator is given for each photon if ithas generated a charge in the photocathode, which means thatthe photon has not been reflected or absorbed, and ends on thearea of the photocathode.The simulation is controlled by the three probabilities for re-flection at the photocathode, absorption at the photocathode ifnot reflected, and reflection at the air-radiator interface. Theseparameters have been determined from a χ minimization be-tween simulation results and data. The simulation of light propagation has been used as one pos-sible model to quantitatively describe the radial dependence ofthe time resolution σ t on the produced charge. It is based onthe relation: σ ∝ σ N P.E. (2)where σ SPTR is the single photoelectron time resolution of thedetector and N P.E. is the number of photoelectrons generated atthe photocathode. This relation has been shown to be valid forMCP-PMTs, albeit without considering the spatial dependence[8]. This relation will be extended by including, spatially re-solved, absorption and reflection of the Cherenkov photons atthe air-radiator interface and at the photocathode.As the exact quantum e ffi ciency of the photocathode is un-known, we do not extract the number of produced photoelec-trons from the simulation. Instead, the ratio between the gen-erated photons and photons reaching the photocathode ( N ) andthe ratio of generated photons and those reaching the photocath-ode after x reflections ( N x ) has been used. The number of pho-toelectrons is linearly correlated to the amount of light reachingthe photocathode. Therefore, it is valid to substitute this pa-rameter and rescale Eq. (2) to show the correlation between thetime resolution and the relative amount of light reaching thephotocathode.A distinction between photons reaching the photocathode di-rectly ( N ) and those reaching the photocathode after exactlyone reflection ( N ) is made. The variance of the timing of oneMCP-PMT is calculated by σ = (cid:18) N N (cid:19) AN + (cid:18) N N (cid:19) BN (3) where N is the ratio of all simulated photons to the onesreaching the photocathode, N and N are the ratio of photonsreaching the photocathode after 0 or 1 reflections, respectively; A is a scaling factor correlated to σ , A = ( σ / N ) ( r = inthe center of the MCP-PMT, and B is the corresponding scalingfactor for the reflected photons. The reflected photons will in-crease the signal arrival time jitter and may thus slightly worsenthe timing resolution. B is defined as: B = (cid:16) √ A + ∆ σ (cid:17) (4)with the additional parameter ∆ σ , introduced to this model todescribe the impact of the reflected photons on the rising edgeof the signal and therefore the signal arrival time (SAT).
3. Description of the Beam Measurement Setup
The MCP-PMTs were operated at a nominal gain of 8 · along with the PICOSEC-Micromegas detectors in muon beamsat the CERN SPS secondary beam lines [9]. The energy of thesemuons during our measurements is up to 180 GeV and the parti-cle rate reached up to 2 · s − . The MCP-PMTs feature an un-segmented anode and thus cannot provide position informationfor the incident particle. Consequently, a beam telescope withthree triple-GEM detectors [10] with a two-dimensional stripreadout structure has been used to track the incoming muonsand determine the impact point of the particles on the MCP-PMTs. Details of track reconstruction can be found in [11].The Scaleable Readout System (SRS) [12], interfacing APV25front-end boards [13], has been used to record GEM detectorhits, while the MCP-PMT signals were acquired with oscillo-scopes with integrated sampling digitizer with 2.5 GHz band-width at 20 GS / s sampling rate (”Teledyne LeCroy - WaveRun-ner 8254”). A common trigger signal, generated by coincidenthits in three scintillation detectors, was fed to the SRS and thesampling oscilloscopes. A reliable synchronization of the indi-vidual streams was possible, by recording an internally gener-ated SRS trigger number as analog bit stream on an additionaloscilloscope channel. The time resolution of a detector is defined by the uncertaintyof its signal arrival time (SAT) to a time reference. The SATinformation is derived by an o ffl ine 40 % constant fraction dis-crimination (CFD) analysis for each acquired waveform. Therising edge of each signal has been fitted with a generalized lo-gistic function and the time value of 40 % of the amplitude hasbeen calculated for the CFD analysis. Signals of two MCP-PMTs are recorded on two separate channels and their time res-olution is evaluated from the di ff erence of their SATs. Eventhough both MCP-PMTs are operated at the same amplificationvoltage, their response is slightly di ff erent. For charged parti-cles hitting the innermost region (radius r < . ± .
03) pC and (1 . ± .
03) pC, for MCP-PMT 1 and MCP-PMT 2, respectively. Because of di ff erentresponses, the MCP-PMTs can have di ff erent time resolution.3herefore, throughout this study, only the combined time reso-lution of the two MCP-PMTs is estimated, see Sec. 4.2.The contribution of readout electronics and CFD algorithm tothe observed time resolution is estimated by recording an iden-tical signal on two oscilloscope channels and determining thevariance in the signal arrival time. An 18 GHz power dividerwas used to split the signal of one MCP-PMT into two iden-tical signals and no impact on the signal quality could be ob-served. The time di ff erence of the SAT of both signals followsa distribution with a width of σ = (3.09 ± σ DAQ = (3 . ± .
04) ps, which yields a time res-olution per channel of σ DAQ / √ = (2 . ± .
03) ps. The mea-sured instrument response function agrees with the trigger andinterpolation jitter of ≤ .
4. Results
Combining signal charge and track impact point measure-ments, the spatial dependence of the mean signal charge hasbeen evaluated. The radial dependence of the mean signalcharges relative to the center of each MCP-PMT are shown inFigure 3 together with the results from the analytical modelwith and without reflections and from the Monte-Carlo simu-lation. The signal charge decreases as expected with increas-ing distance from the MCP-PMT center due to a decrease ofthe Cherenkov light reaching the photocathode. Maximum av-erage signal charge is observed for tracks with the full conecontained in the acceptance of the photocathode. This can beobserved in the inner 2 . r > . r =
0; it shows an agreement with thedata for small radii, but it underestimates the signal charge forevents at larger radii. For the blue dashed line, first-order reflec-tions are added to the geometrical calculation, with fitting theweighting factor to w = .
08; the curve shows an agreementwith the data in the outer part of the window ( r > χ minimization the reflection probability on thephotocathode is determined to be (0 . ± . . ± .
05) and the reflectionprobability at the air-radiator interface is (0 . ± . M ean C ha r ge ( p C ) MeasurementGeometric modelingwithout reflexionGeometric modelingwith 1st order reflexionMonte-Carlo simulation (a) MCP-PMT 1 M ean C ha r ge ( p C ) MeasurementGeometric modelingwithout reflexionGeometric modelingwith 1st order reflexionMonte-Carlo simulation (b) MCP-PMT 2Figure 3: Radial distribution of the average signal charge for the two MCP-PMTs. Both MCP-PMTs are operated at 2800 V. The green dotted curve showsthe expected distribution by geometrical calculation without including reflec-tions. The blue dashed line shows the geometrical calculation after includingfirst order reflections. The red line shows the distribution expected from a ded-icated simulation that includes the detector characteristics (see text). ff erent mean signal charges in the center of the photocath-ode. The determined parameters need to be taken with a grainof salt. More refined models with precise knowledge on theMCP-PMT materials and including signal formation processesmight yield di ff erent results. The time di ff erence of the SAT of both MCP-PMTs results ina Gaussian-like distribution (Figure 4). The standard deviation σ tot of this distribution is determined from fitting a Gaussianfunction and σ tot = (cid:113) σ + σ + σ , (5)where σ MCP1 and σ MCP2 are the time resolutions of MCP-PMT1 and MCP-PMT 2, respectively, and σ DAQ denotes the setup-specific contribution of DAQ system and analysis method.An extraction of the time resolution of a single MCP-PMTfrom these beam measurements is only possible when assumingthat σ MCP1 ≡ σ MCP2 at the same track proximity to the center.As the respective alignment of the two MCP-PMTs is measuredto be (0 . ± .
01) mm, as the 2D-distance between the MCP-PMT centers, and they are operated at equal high voltage, thisassumption is plausible. However, the observed pulse height inMCP-PMT 2 is 30 % smaller than that in MCP-PMT 1. Thiscould be due to a di ff erent photo-conversion e ffi ciency or am-plification factor. It has to be assumed that the dependence ofthe resolution on the signal amplitude is the same despite theobserved gain di ff erence, to extract the time resolution of a sin-gle MCP-PMT. Since the available dataset does not allow fora determination of these e ff ects on the time resolution, we willonly discuss in the following the combined MCP-PMT time res-olution: σ MCP (cid:66) (cid:113) σ + σ = (cid:113) σ − σ . (6)If the two MCP-PMTs behaved similarly, a time resolution of σ MCP / √ In the following, we will discuss the time resolution of theMCP-PMTs as a function of the impact point of the incidentparticle. Events are grouped according to their impact radiussuch that the statistics of events hitting each ring-like area isequal and the time resolution is then independently calculatedfor each group. Figure 5 shows the combined time resolution σ MCP as a function of the distance with respect to the center ofMCP-PMT 2.The combined time resolution is better than 10 ps in the in-ner radius of r < − − − − − − − − − (ns) MCP2 - SAT
MCP1
SAT020406080100120140160180 N u m be r o f e v en t s ± = 7.2 tot σ ± = -5.036 µ / ndf = 129.7 / 84 χ Figure 4: Di ff erence of the signal arrival times at the two MCP-PMTs. Dataare selected by requiring a particle track passing through the inner 11 mm ofthe first MCP. A combined time resolution of (7.2 ± C o m b i ned T i m e R e s o l u t i on ( p s ) Refraction:Air: 0.8PC: 0.2Abs: 0.4Chi2: 103.068
Figure 5: Combined time resolution as a function of the track impact pointdistance from the photocathode center. The red line shows a possible modellingusing results from the Monte-Carlo simulation, as explained in Sec. 2. particles are further away from the center of the MCP-PMT, themean signal charge is reduced and the time resolution degrades.The red curve in Figure 5 shows the results of the Monte-Carlo simulation (see Eq. (3)) described in Sec. 2. Optimalagreement between simulation results and observed data isreached with A = , ∆ σ = . = . Theadditional spread for the reflected photons of both MCP-PMTs ∆ σ must not be confused with the time delay of the reflectedphotons, which is of the order of ∼
40 to 45 ps. ∆ σ dependson the probabilities for reflection and absorption used in theMonte-Carlo model of the charge distribution.
5. Conclusions
A study of the MCP-PMT timing resolution was performedusing muons with energies up to 180 GeV in a test beam atCERN. In particular, the study aimed at characterizing the timeresolution as a function of the position of the impinging muons.5he MCP-PMT provides a precise and reliable reference timewell below 10 ps for charged particles, traversing the MCP-PMT inside the photocathode area. However, when the particletraverses the detector further outside the photocathode area, thetiming resolution begins to degrade due to a decrease of the de-tected fraction of Cherenkov light.The model, based on simple geometrical arguments, hasshown good agreement with the observed behavior by assumingan additional time jitter of 7.5 ps due to the reflected light. Weconclude that the MCP-PMT is appropriate for use as a time ref-erence to determine the time resolution of detector prototypeslike the PICOSEC-Micromegas, that have shown a time resolu-tion of 25 to 30 ps [1], as long as the investigated region in thedetector under test is well aligned to the MCP-PMT and has adiameter smaller than 13 mm.
Acknowledgments
We acknowledge the financial support of the Cross-Disciplinary Program on Instrumentation and Detection ofCEA, the French Alternative Energies and Atomic EnergyCommission; the RD51 collaboration, in the framework ofRD51 common projects; and the Fundamental Research Fundsfor the Central Universities of China. L. Sohl acknowledgesthe support of the PHENIICS Doctoral School Program ofUniversit´e Paris-Saclay. J. Bortfeldt acknowledges the sup-port from the COFUND-FP-CERN-2014 program (grant num-ber 665779). M. Gallinaro acknowledges the support from theFundac¸ ˜ao para a Ciˆencia e a Tecnologia (FCT), Portugal (grantsIF / / / FIS-PAR / / References [1] J. Bortfeldt, et al., PICOSEC: Charged particle timing at sub-25 picosec-ond precision with a Micromegas based detector, Nucl. Instrum. Meth.A903 (2018) 317–325. doi:10.1016/j.nima.2018.04.033 .[2] T. Papaevangelou, et al., Fast Timing for High-Rate Environments withMicromegas, EPJ Web Conf. 174 (2018) 02002. arXiv:1601.00123 , doi:10.1051/epjconf/201817402002 .[3] A. Benaglia, S. Gundacker, P. Lecoq, M. Lucchini, A. Para, K. Pauwels,E. Au ff ray, Detection of high energy muons with sub-20ps timingresolution using l(y)so crystals and sipm readout, Nuclear Instrumentsand Methods in Physics Research Section A: Accelerators, Spec-trometers, Detectors and Associated Equipment 830 (2016) 30 – 35. doi:https://doi.org/10.1016/j.nima.2016.05.030 .URL [4] R. Ota, K. Nakajima, I. Ogawa, Y. Tamagawa, H. Shimoi, M. Suyama,T. Hasegawa, Coincidence time resolution of 30 ps FWHM using a pairof cherenkov-radiator-integrated MCP-PMTs, Physics in Medicine & Bi-ology 64 (7) (2019) 07LT01. doi:10.1088/1361-6560/ab0fce .URL https://doi.org/10.1088%2F1361-6560%2Fab0fce [5] Hamamatsu PHOTONICS K. K., R3809U-50 SERIES (2015).URL [6] J. Vavra, PID techniques: Alternatives to RICH methods, Nucl. Instrum.Meth. A876 (2017) 185–193. doi:10.1016/j.nima.2017.02.075 . [7] L. Sohl, Spatial time resolution of MCP-PMTs as a t -reference, Nucl.Instrum. Meth. A936. doi:10.1016/j.nima.2018.11.138 .[8] K. Inami, N. Kishimoto, Y. Enari, M. Nagamine, T. Ohshima, A 5-psTOF-counter with an MCP-PMT, Nucl. Instrum. Meth. A560 (2006) 303–308. doi:10.1016/j.nima.2006.01.027 .[9] N. Doble, L. Gatignon, G. von Holtey, F. Novoskoltsev, The upgradedmuon beam at the SPS, Nucl. Instrum. Meth. A343 (1994) 351–362. doi:10.1016/0168-9002(94)90212-7 .[10] F. Sauli, GEM: A new concept for electron amplification in gas de-tectors, Nucl. Instrum. Meth. A386 (1997) 531–534. doi:10.1016/S0168-9002(96)01172-2 .[11] J. Bortfeldt, The Floating Strip Micromegas Detector, Springer, 2015. doi:10.1007/978-3-319-18893-5 .[12] S. Martoiu, et al., Development of the scalable readout system for micro-pattern gas detectors and other applications, JINST 8 (2013) C03015. doi:10.1088/1748-0221/8/03/C03015 .[13] M. French, et al., Design and results from the apv25, a deep sub-microncmos front-end chip for the cms tracker, Nucl. Instrum. Meth. A466 (2)(2001) 359 – 365. doi:10.1016/S0168-9002(01)00589-7 .[14] Teledyne LeCroy, WaveRunner 8000 Series Datasheet (2018).URL http://cdn.teledynelecroy.com/files/pdf/waverunner8000-datasheet.pdfhttp://cdn.teledynelecroy.com/files/pdf/waverunner8000-datasheet.pdf