LEAPS-INNOV D2.3 - Preliminary qualification of the assembled detector head

Deliverable no. D2.3 Page 1 of 8 Project information Project full title LEAPS pilot to foster open innovation for accelerator - based light sources in Europe Project acronym LEAPS - INNOV Grant agreement no. 101004728 Instrument Research and Innovation Action (RIA) Duration 01/0 4 /202 1 – 31/0 3 /202 5 Website https://www.leaps - innov.eu/ Deliverable information Deliverable no. D2.3 Deliverable t itle Preliminary qualification of the assembled detector head Delivearable responsible ESRF, (SOLEIL and DIAMOND) Related Work - Package/Task WP2 (XAFS - DET) /Task 2.7. Type (e.g. Report; other) Report Author(s) F.J. Iguaz , E N Gimenez Dissemination level Public Document Version V1 Date 27/01/2025 Download page Document information Version no. Date Author(s) Comment V1 27/01/2025 FJ. Iguaz, EN. Gimenez Deliverable no. D2.3 Page 2 of 8 [Table of Contents] 1. Introduction 2. Detector mechanical assembly, vacuum and vibrational tests 3. Electronics assembly and tests in the detector 4. Sensor assembly in the detector 5. Preliminary qualification of the detector head 6. Conclusions 7. References 1. Introduction This document summarizes the assembly of the detector head involving the integration of the germanium sensor (task 2.2), the front-end and back-end electronics (task 2.3), the interconnection (task 2.4) and the cryostat and cooling system (task 2.5) [1]. It also describes the first thermal, vibrational and electrical tests of the detector head once the assembly was completed. The assembly and preliminary tests of the detector head was performed at ESRF with the contributions of other partners (DIAMOND, ESRF, SOLEIL) and regular fortnightly meetings took place months beforehand with all the partners of the collaboration to achieve the deliverable. 2. Detector mechanical assembly, vacuum and vibrational tests All the mechanical parts were clean first to ensure vacuum and were assembled to the cryostat in June 2024, as illustrated in Figure 1. During the assembly of all components, only a minor issue regarding a different size of screw-hole diameter of the PEEK holder (shown in image 2) was observed. This issue was corrected by increasing the holes to M4 size. Then, the cryostat was closed (image 4) with the corresponding flanges (Allectra feedthrough flange, operator valve, the nose tube without the Beryllium window) and a vacuum leak test was made. The vacuum level reached values within specifications (around 10-6 mbar) and no leaks were observed. Deliverable no. D2.3 Page 3 of 8 Figure 1: Images of the mechanical assembly: 1) the Allectra feedthrough flange with the Back-End Board (BEB), 2) the copper bar, the copper arm and the PEEK holder, 3) the cryostat closed except for the operator valve ,4) the nose tube without the Beryllium window. In September 2024, the thermal strap, which links the cryocooler copper bar to the detector copper bar, was added to the assembly. After its installation, the vibration tests were made with a sensor installed at the copper arm, as shown in Figure 2. The sensor measured values of 5, 7 and 4 mg in X, Y and Z, respectively. These values are within specifications (less than 10 mg). Figure 2: The vibrational test setup (top-left), and spectra for axis X (top-right), Y & Z (bottom). The measured value for each axis is provided by the most intense peak at a frequency higher than 50 Hz. 3. Electronics assembly and tests in the detector The electronic chain to assemble in the detector head consists of two blocks: the Front-End Board (FEB) with the TETRA ASICs, which are in vacuum, and the Back-End Boards (BEBs), which sits outside vacuum and it is attached to one of the flanges. Both parts are connected by a flexi cable (FEB flex) to transfer the data out. In October 2024, some partners (SOLEIL, DIAMOND, ESRF) meet in ESRF to perform the assembly and tests of the remanufactured electronic chain. The FEB and the FEB flex were integrated in the cryostat, as illustrated in Figure 3. First, the plastic cover of TETRA ASIC was replaced by a vacuum compatible PEEK cover (images 1). Then, the FEB was screwed to an interface copper piece and to the sensor arm (image 2). At the sensor arm there was glued a PT100 thermometer to monitor the temperature. The FEB flex and the PT100 cable were respectively connected to the DSUB25 and DSUB9 connectors at Deliverable no. D2.3 Page 4 of 8 the back flange of the cryostat (image 4). Once the cryostat was closed, its leak-tightness was verified again (image 5) and the cryocooler was switched on. Figure 3: Images of the electronics assembly in the cryostat: 1) the FEB and its PEEK cover assembled to the copper interface piece, 2) this assembly installed at the sensor arm, 3) the FEB flex along the detector nose, 4) the DSUB25 and DSUB9 connectors at the back flange, 5) the leakage test before closing, 6) the electronics box assembled, 7) the final setup during the noise measurement. An electronics box was designed by DESY to contain all the BEBs of the electronics. The electronic box was assembled and screwed to the cryostat’s rear flange. The two original bars could not be installed due to an incompatibility with the flange screw’s location, but they were temporary replaced by two 3D-printed bars (image 6). Metallic tape between the box and the cryostat flange (image 7) was added for electrical continuity. Once the full assembly was completed and the system reached a stable temperature, the first tests with the electronic chain were performed. The noise level and the fall time of test-induced signals were measured at a FEB temperature of 163 K. The temperature value was limited by the cold finger configuration (as a first test no indium foil was placed at all copper links and the FEB bridge was not installed as required the sensor). The results for noise level and fall time are shown in Figure 4. The noise level was lower at cold temperature than at room temperature, and the fall time values were shorter, as expected. Deliverable no. D2.3 Page 5 of 8 Figure 4: Dependence of the electronic noise with the peaking time (left) and the mean fall time of testinduced signals (right) at room (313 K) and cold temperature (163 K). 4. Sensor assembly in the detector In December 2024, the Large Pixel Pad (LPP) germanium sensor, 10 pogo pins and a PEEK spacer were connected to the FEB in the clean room of the ESRF detector group, as shown in Figure 5 (images 1, 2). An aluminium cover for the three TETRA ASICs with a thermal bridge, a PEEK cover for the FEB flex and the HV cable for the HV PCB were also integrated to the detector head (image 3). The IR window provided by MIRION (image 4) was used while the final IR is being manufactured, and a molybdenum collimator (image 6) was placed in front of the IR window. All these elements composed the sensor head assembly, which was screwed to the sensor arm (image 5). Two PT100 thermometers (one at the sensor arm, and a second one at the cryocooler bar) were installed to monitor the temperature at different relevant places inside the detector head. Figure 5: Images of the sensor assembly in the detector: 1) the pixels of the LPP sensor with Indium contacts, 2) the 10 pogo pins installed in the PEEK spacer, before the installation of the FEB; 3) the final Deliverable no. D2.3 Page 6 of 8 assembly, including the aluminium cover, a PEEK protection for the flex cable and the HV cable; 4) a view of the IR window, provided by MIRION; 5) the sensor head assembly installed at the sensor arm; 6) the molybdenum collimator; 7) the PT100 thermometer at the sensor arm. To complete the detector head assembly, a beryllium window of 30 mm diameter was glued to its emplacement at the nose tube and its leak tightness was verified. The FEB flex, the HV cable and the two PT100 cables were respectively connected to the DSUB25 connector, the HV feedthrough pin and the DSUB9 of the back flange. Before moving the detector to the test lab, the cryostat was reopened to replace to components: one PT100 thermometer, which failed, and the operator valve, which was damaged due to a misalignment of the operation tool. 5. Preliminary qualification of the detector head In December 2024, the first tests with the assembled detector head comprising the germanium sensors, electronic chain, cryocooler and mechanical components, took place at ESRF with partners of the collaboration (SOLEIL, DIAMOND, ESRF) and industry (XGLab) meeting in person. Once the detector head was in vacuum, the process of cooling it lasted ~3.5 hours, in agreement with the thermal simulations [2]. The FEB PT100 thermometer measured 128 K, and the PT100 thermometer at the sensor arm indicated 139 K, reaching the temperature specifications for the FEB (below 300 K [3]). After that, the germanium sensor was polarized to +100 V, a value in the operation range recommended by the manufacturer XGLab (70 to 300 V [4]). Leakage ramps were observed for the 10 channels in absence of X-rays and for the three gain settings. This fact validated the interconnection process. The reset period of all ramps was measured for high gain, with approx. 1 sec, which is equivalent to leakage current of 0.05 pA. This value is well below the specifications (less than 1 pA) and indicates that the germanium sensor is operated in the correct temperature range (below 100 K). The detector was irradiated with photons from an iron (55Fe) source of 5.9 and 6.4 keV and gammas from an americium (241Am) source of 26.3 and 59.5 keV to evaluate its spectroscopic performance and the electronic noise present in this configuration. As expected, only the 7 inner pixels showed X-ray induced signals and an energy spectrum, as the collimator was designed [5] to block the 3 outer ones which are used to discriminate events and increase the energy resolution. The energy resolution observed in this preliminary test is around 400 eV (FWHM) at 5.9 keV and between 450 and 500 eV at 59. 5 keV, as shown in Figure 6. The resolution is still far from the specifications (below 180 eV at 5.9 keV) and is limited by the electronic noise. According to preliminary tests made with XGLab staff, who also participated in person in the preliminary qualification of the detector head, the noise is mainly induced by the DC converters Deliverable no. D2.3 Page 7 of 8 of the back-end board into the HV cable and the electronic box. This hypothesis will be checked in further tests with the system in February of 2025 by rerouting the HV line to a side flange. In addition, a redesign of the electronic box may be considered after rerouting the HV line, to reduce the noise introduced to the system. Figure 6: Energy spectra from an iron source for the central pixel (left), and from an americium source (right) for the 7 inner pixels. 6. Conclusions The detector head was physically assembled between June and December 2024 by the collaboration. Regular fortnightly meetings have been held throughout the project to ensure that all the parts match together and mitigate risks. Indeed, only a couple of minor issues with an easy fix emerged during the detector assembly. Once the detector head was assembled, the first thermal, vibrational and electrical tests were performed and showed that the system was within specifications. The remanufactured electronics and the interconnection have been validated after setting-up the system. A spectroscopic signal was obtained from all the seven active sensor pixels the first time that all the main parts of the detector were put together. While the energy resolution of the system is not yet within specifications, a first source of noise was identified: an induced noise from the BEBs into the HV line. There are clear plans to study and fix this issue in the next tests during 2025 and investigate other ways to optimize the system. The system is not fully completed as there are other parts that need to be added or replaced mainly due to long delivery lead times: a tungsten collimator (to replace the molybdenum collimator), the final IR window and the zeolite container. These parts will be included in Q2 2025 and after their integration into the detector, a new validation and tests of the detector will take place. 7. References [1] E. Gimenez-Navarro et al., Development of a new generation multi-element monolithic HPGe for XAFS applications, proceedings of SRI 2024 conference, to be published in J. Phys. Conf. Ser [2] M. Quispe, FEA Germanium Detector: Transient Studies, 09/01/2025 Deliverable no. D2.3 Page 8 of 8 [3] M. Quispe et al., Thermal and vibrational studies of a new germanium detector for X-ray spectroscopy applications at synchrotron facilities, JACoW IPAC2024 (2024) TUPR75. [4] MIRION Technologies, Project LEAPS Specifications report for “big” pattern, 18/04/2023, Internal reference: M-DET-F-257-B. [5] N. Goyal et al., Progress in the Development of Multi-Element Monolithic Germanium Detectors in LEAPS-INNOV Project: Insights from Detector Performance Simulation, proceedings of SRI 2024 conference, to be published in J. Phys. Conf. Ser