Figure 1 - uploaded by Ian Baker
Content may be subject to copyright.
Source publication
The performance of the current high speed near infrared HgCdTe sensors
operating in fringe trackers, wavefront sensors and tip-tilt sensors is
severely limited by the noise of the silicon readout interface circuit
(ROIC), even if state-of-the- art CMOS designs are used. A major
improvement can only be achieved by the amplification of the
photoelect...
Contexts in source publication
Context 1
... solid state mechanisms for avalanche gain in HgCdTe have been well described and two recommended papers are [1] and [2], together with the reference lists. Figure 1 shows a potential energy schematic for a typical photodiode which illustrates the single-carrier, cascade-like gain mechanism, together with the history-dependent behaviour that leads to the observed low excess noise factors. In HgCdTe this is an electron gain mechanism so the absorber must be P-type, which favours an N-on-P structure. ...
Context 2
... the array common, at the anode side of the HgCdTe junction, at a fixed voltage and changing the voltage PRV on the cathode side of the junction, yields the number of digital ADU's per Volt. The same calibration can be made by keeping the video output of the detector fixed and varying the external offset voltage of the symmetric preamplifier shown on the right side of Figure 13. The ratio of the voltage change at the integrating node and the detector output is the transfer function which was measured to be 0.58 for the SWALLOW ROIC. ...
Context 3
... quantum efficiency is a function of temperature as shown in Figure 10. It rises from 28.4% to 51.8% in H-band and from 37.6% to 66.6% in K-band when the detector is cooled from T=80K to T=40K. ...
Context 4
... differences of images observed at two blackbo- dy temperatures generates a pattern having a calibrated flux which is 1.75 electrons. It can be detected as shown in Fig- ure 11. The image is an average of 16 DCS frames taken with an integration time of 5 ms. ...
Context 5
... Figure 12 the shift register based addressing of the SAPHIRA windowing scheme is shown. For the row addresses, a bit stream of 256 bits have to be loaded and for the column addresses, 10 bits (320 pixel= 10x32 pixel) have to be loaded to address the required readout window. ...
Context 6
... overhead can be reduced with faster clocking. The separate window addressing capability for readout (green regions in Figure 12) and reset windows (yellow regions in Figure 12) makes it possible to have different integration times for different windows when the multiple non- destructive readout scheme is used. Furthermore, possible edge effects can be avoided when the reset region is made larger than the readout region. ...
Context 7
... overhead can be reduced with faster clocking. The separate window addressing capability for readout (green regions in Figure 12) and reset windows (yellow regions in Figure 12) makes it possible to have different integration times for different windows when the multiple non- destructive readout scheme is used. Furthermore, possible edge effects can be avoided when the reset region is made larger than the readout region. ...
Context 8
... the SAPHIRA array with 32 channels, the video outputs are amplified directly at the focal plane, on the detector board as shown in Figure 13. The robust and well proven symmetric preamplifier design used for the SWALLOW proto- type has been ported to the SAPHIRA detector. ...
Context 9
... provides the required pixel speed of 5Mpixel/s/channel and operates reliably at cryogenic temperatures down to T=40K. The detector fan-out board with low- pass filters, antistatic protection and cryogenic preamplifiers is shown on the left side of Figure 13. N the right side is the schematics of the symmetric preamplifiers. ...
Context 10
... the plane conjugate to the detector in front of the cryostat window is at room temperature a test pattern such as a grid of holes illuminated by an extended blackbody can be used to generate a pattern of calibrated flux on the detector. The test pattern is shown in the left lower corner of Figure 14. To calculate the photon flux, the transmission curves of the filters have been measured at cryogenic temperatures. ...
Context 11
... small capacitance at the integrating node increases the conversion gain in terms of Volt/electron. The first goal of lowering the noise floor was met as demonstrated by Figure 15. The noise histograms of the voltage measured at the detector outputs of the SWALLOW and the SAPHIRA ROICs are compared. ...
Context 12
... the edges of the reset window the coupling of the pixels adjacent to the reset window is measured. The coupling is < 2% as shown by the signal profile on the left side of Figure 16. The second method employs the autocorrelation calculated from a series of flat-field differences [9]. ...
Context 13
... second method employs the autocorrelation calculated from a series of flat-field differences [9]. The autocorrelation of next neighboring pixels is caused by interpixel capacit- ance and is < 0.5% as shown on the right side of Figure 16. Therefore, the error of the conversion gain due to interpixel capacitance is < 2% when the variance instead of the integrated autocorrelation is used to derive the conversion gain from the photon transfer curve. ...
Context 14
... sub-pixel sampling is enabled and 4 ADC conver- sions are made for each pixel and the average is taken for the pixel value, then the frame time is 226 s. The minimum integration time for four Fowler pairs with 4 sub-pixel samples is 904 s In Figure 17, the median of the readout noise of the described windowed readout is plotted versus the number of nonde- structive readouts for two different operating temperatures. The selected bias voltage was 10.55V corresponding to an APD gain of 20.4. ...
Context 15
... readout noise is possible with the APD gain in combination with the described advanced sampling techniques. To further illustrate the potential of eAPDs, an average of 32 frames of 256x256 pixels taken with double correlated sampling, is shown in Figure 18. The pattern was taken with the grid of holes as described in chapter 6. ...
Citations
... Therefore, HgCdTe EAPDs of conventional planar designs have also been successfully developed. Nowadays, planar EAPDs exhibit a lower dark current and technological reliability to be competitive with the cylindrical HDVIP design, having better performance in the SWIR spectral range [69]. ...
Conventional designs of an avalanche photodiode (APD) have been based on a planar p–n junction since the 1960s. APD developments have been driven by the necessity to provide a uniform electric field over the active junction area and to prevent edge breakdown by special measures. Most modern silicon photomultipliers (SiPM) are designed as an array of Geiger-mode APD cells based on planar p–n junctions. However, the planar design faces an inherent trade-off between photon detection efficiency and dynamic range due to loss of an active area at the cell edges. Non-planar designs of APDs and SiPMs have also been known since the development of spherical APDs (1968), metal-resistor-semiconductor APDs (1989), and micro-well APDs (2005). The recent development of tip avalanche photodiodes (2020) based on the spherical p–n junction eliminates the trade-off, outperforms the planar SiPMs in the photon detection efficiency, and opens new opportunities for SiPM improvements. Furthermore, the latest developments in APDs based on electric field-line crowding and charge-focusing topology with quasi-spherical p–n junctions (2019–2023) show promising functionality in linear and Geiger operating modes. This paper presents an overview of designs and performances of non-planar APDs and SiPMs.
... C-RED1 is an ultra-low noise infrared camera based on the Saphira detector and fabricated by First Light Imaging, specialized in fast imaging camera, after the successful commercialization of the OCAM2 camera [4] dedicated to extreme adaptive optics wavefront sensing. Designed and fabricated by Leonardo UK, formerly Selex, the Saphira detector is designed for high speed infrared applications and is the result of a development program alongside the European Southern Observatory on sensors for astronomical instruments [5], [6], [7]. It delivers world leading photon sensitivity of <1 photon rms with Fowler sampling and high speed non-destructive readout (>10K frame/s). ...
... C-RED1 is an ultra-low noise infrared camera based on the Saphira detector and fabricated by First Light Imaging, specialized in fast imaging camera, after the successful commercialization of the OCAM2 camera [1] dedicated to extreme adaptive optics wavefront sensing. Designed and fabricated by Leonardo UK, formerly Selex, the Saphira detector is designed for high speed infrared applications and is the result of a development program alongside the European Southern Observatory on sensors for astronomical instruments [2], [3], [4]. It delivers world leading photon sensitivity of <1 photon rms with Fowler sampling and high-speed non-destructive readout (>10K frame/s). ...
... C-RED1 is an ultra-low noise infrared camera based on the Saphira detector and fabricated by First Light Imaging, specialized in fast imaging camera, after the successful commercialization of the OCAM2 camera [4] dedicated to extreme adaptive optics wavefront sensing. Designed and fabricated by Leonardo UK, formerly Selex, the Saphira detector is designed for high speed infrared applications and is the result of a development program alongside the European Southern Observatory on sensors for astronomical instruments [5], [6], [7]. It delivers world leading photon sensitivity of <1 photon rms with Fowler sampling and high speed non-destructive readout (>10K frame/s). ...
There have been no significant breakthroughs in infrared imagery since the hybridization of III-V or II-VI narrow-bandgap semiconductors on complementary metal-oxide semiconductor (CMOS) read-out integrated circuits (ROICs). The development of third-generation, linear-mode avalanche photodiode arrays (LmAPDs) using mercury cadmium telluride (MCT) has resulted in a significant sensitivity improvement for short-wave infrared (SWIR) imaging. The first dedicated LmAPD device, called SAPHIRA (320x256/24 microns), was designed by Leonardo UK Ltd specifically for SWIR astronomical applications. In the past decade there has been a significant development effort to make larger LmAPD arrays for low-background astronomy. Larger LmAPD formats for ultra-low noise/flux SWIR imaging, currently under development at Leonardo include a 512 x 512 LmAPD array funded by ESO, MPE and NRC Herzberg, a 1k x 1k array funded by NASA and a 2K x 2K device funded by ESA for general scientific imaging applications. The 2048x2048 pixel ROIC has a pitch of 15 microns, 4/8/16 outputs and a maximum frame rate of 10 Hz. The ROIC characterization is scheduled in the third quarter of 2022, while the first arrays will be fabricated by end-2022. The hybridized arrays will be characterized during end-2022. At this time, First Light Imaging will start the development of an autonomous camera integrating this 2Kx2K LmAPD array, based on the unique experience from the C-RED One camera, the only commercial camera integrating the SAPHIRA SWIR LmAPD array.The detector will be embedded in a compact high vacuum cryostat cooled with low vibration pulse at 50-80K which does not require external pumping. Sub-electron readout noise is expected to be achieved with high multiplication gain. Custom cold filters and beam aperture cold baffling will be integrated in the camera.
... C-RED1 is an ultra-low noise infrared camera based on the Saphira detector and fabricated by First Light Imaging, specialized in fast imaging camera, after the successful commercialization of the OCAM2 camera [1] dedicated to extreme adaptive optics wavefront sensing. Designed and fabricated by Leonardo UK, formerly Selex, the Saphira detector is designed for high speed infrared applications and is the result of a development program alongside the European Southern Observatory on sensors for astronomical instruments [2], [3], [4]. It delivers world leading photon sensitivity of <1 photon rms with Fowler sampling and high-speed non-destructive readout (>10K frame/s). ...
We present here the latest results obtained with the C-RED One camera developed by First Light Imaging for fast ultra-low noise infrared applications. This camera uses the Leonardo Saphira e-APD 320x256 infrared sensor in an autonomous cryogenic environment with a low vibration pulse tube and with embedded readout electronics system. Some recent improvements were made to the camera. The first important one concerns the total noise of the camera. Limited to 1.75 microns wavelength cut-off with proper cold filters, looking at a blackbody at room temperature and f/4 beam aperture, we now measure total noise down to 0.6 e at gain 50 in CDS mode 1720 FPS, dividing previous noise figure by a factor 2. The total camera background of 30-400 e/s is now achieved with a factor 3 of background reduction, the camera also looking at a room temperature blackbody with an F/4 beam aperture. Image bias oscillations, due to electronics grounding scheme, were carefully analyzed and removed. Focal plane detector vibrations transmitted by the pulse tube cooling machine were also analyzed, damped and measured down to 0.3 microns RMS, reducing focal plane vibrations by a factor 3. In addition, a vacuum getter of higher capacity is now used to offer camera operation without camera pumping during months. The camera main characteristics are detailed: pulse tube cooling at 80K with limited vibrations, permanent vacuum solution, ultra-low latency Cameralink full data interface, safety management of the camera by firmware, online firmware update, ambient liquid cooling and reduced weight of 20 kg.
... We use the IOTA readout mode, as for the sky observation. However, we slightly reduce the number of Nreads (8), Nloops (2) and the size of the window (320 x 20 pixels). These numbers were selected to achieve the highest possible maximum frame rate (1916 FPS), while still achieving a theoretical sub-electron readout noise. ...
De par leur interaction avec leur environnement, les étoiles massives contribuent significativement à l'évolution de leur galaxie hôte. Cependant leur processus de formation est encore méconnu. Pour mieux contraindre les modèles de formation de ces étoiles, l'étude de leur multiplicité est essentielle. Alors que la photométrie et la spectroscopie permettent d'étudier les systèmes multiples séparés de quelques millisecondes d'angle (mas) et que l'imagerie directe permet de sonder les séparations supérieures à 50 mas, l'intervalle entre quelques mas et 50 mas de séparation ne pouvait être sondé jusqu'à récemment. Cet intervalle peut être résolu par l'interférométrie optique à longue ligne de base (OLBI), mais cette technique était limitée en sensibilité et ne pouvait observer qu'une dizaine d'étoiles massives jusqu'à il y a peu. Mener une étude statistique de la multiplicité des étoiles massives nécessite d'observer un bien plus grand nombre d'objets.L'objectif de cette thèse est l'amélioration de la sensibilité de la technique OLBI afin de réaliser un relevé de la multiplicité des étoiles massives de l'hémisphère Nord.J'ai tout d'abord participé activement à l'implémentation de deux caméras C-RED ONE dans les instruments interférométriques MIRC-X et MYSTIC pour le réseau CHARA à l’observatoire du mont Wilson, en Californie. Ces caméras, basées sur la technologie des photodiodes à avalanche (APD), ne sont utilisées en astrophysique que depuis quelques années ; notre connaissance sur leur fonctionnement est donc encore limitée. Durant ma thèse, j'ai mené une caractérisation complète de ces caméras. J'ai créé un modèle de distribution du signal des détecteurs APD afin de mieux comprendre les résultats obtenus. Ce modèle et les méthodes classiques de caractérisation ont mis en évidence des différences significatives entre les valeurs de gain et de facteur d'excès de bruit mesurées et celles fournies par le fabricant. Même si cela rend le comptage de photons individuels impossible, les caractéristiques de ces caméras restent exceptionnelles, avec un bruit total inférieur à l'électron pour des cadences de lecture allant jusqu'au kiloHertz, ce qui est fondamental pour s'affranchir au mieux des effets de la turbulence atmosphérique.Ces performances permettent un gain en sensibilité conséquent : MIRC-X atteint une magnitude limite de H = 7.5, comparé à H = 5 pour son prédécesseur MIRC. Cette magnitude limite est confirmée par le relevé de démonstration sur 44 étoiles massives. Lors de ce relevé, j'ai pu observer de manière routinière plusieurs cibles de magnitude H = 7.5, et jusqu'à une magnitude H = 8.1 avec des conditions atmosphériques très favorables. Dans ces observations, j'ai détecté 27 compagnons pour un total de 21 systèmes multiples, ayant des séparations comprises entre 0.5 et 50 mas. Ce relevé de démonstration permet de confirmer la possibilité d'utiliser MIRC-X pour rechercher des compagnons dans l'intervalle non couvert par les autres techniques d'observation, et cela sur un grand nombre d'étoiles massives (>100).Ce relevé de démonstration correspond à la première phase du grand relevé des étoiles massives de l'hémisphère nord ayant une magnitude H < 7.5 que j'ai préparé. J'ai ainsi sélectionné 120 systèmes observables avec le réseau CHARA, ce qui permettra une analyse statistique de la multiplicité des étoiles massives, complémentaire au relevé SMASH+ réalisé dans l'hémisphère sud. Ce type de relevés est essentiel pour contraindre les modèles de formation de ces étoiles.
... The Saphira Detector and C-RED One camera Designed and fabricated by Leonardo UK, formerly Selex, the Saphira detector is designed for high speed infrared applications and is the result of a development program alongside the European Southern Observatory on sensors for astronomical instruments [2], [3], [4]. It delivers world leading photon sensitivity of <1 photon rms with Fowler sampling and high speed non-destructive readout (>10K frame/s). ...
After the development of the OCAM² EMCCD fast visible camera [1] dedicated to advanced adaptive optics wavefront sensing, First Light Imaging moved to the SWIR fast cameras with the development of the C-RED One and the C-RED 2 cameras. First Light Imaging's C-RED One infrared camera is capable of capturing up to 3500 full frames per second with a subelectron readout noise and very low background. C-RED One is based on the last version of the SAPHIRA detector developed by Leonardo UK. This breakthrough has been made possible thanks to the use of an e-APD infrared focal plane array which is a real disruptive technology in imagery. C-RED One is an autonomous system with an integrated cooling system and a vacuum regeneration system. It operates its sensor with a wide variety of read out techniques and processes video on-board thanks to an FPGA. We will show its performances and expose its main features. In addition to this project, First Light Imaging developed an InGaAs 640x512 fast camera with unprecedented performances in terms of noise, dark and readout speed based on the SNAKE SWIR detector from Sofradir. The camera was called C-RED 2. The C-RED 2 characteristics and performances will be described. The C-RED One project has received funding from the European Union's Horizon 2020 research and innovation program under grant agreement N° 673944. The C-RED 2 development is supported by the "Investments for the future" program and the Provence Alpes Côte d'Azur Region, in the frame of the CPER.
... The SAPHIRA, like any NIR detector array, must be hybridized to a matching ROIC that provides the necessary electrical interface (Finger et al. 2012). The SAPHIRA ROIC provides four modes of read operation, output channels selectable to 4, 8, 16, or 32, and a serial programmable interface with other user-configurable options. ...
... ROIC. Five Mark 3 arrays were initially installed in the GRAVITY instrument at VLT (Finger et al. 2012(Finger et al. , 2014(Finger et al. , 2016Kendrew et al. 2012 ...
We present the dark current performance of the SAPHIRA series of HgCdTe APD arrays, characterized as afunction of bias voltage and temperature. We measure a gain-normalized dark current in multiple SAPHIRA arraysof 0.025 e⁻ s⁻¹ pix⁻¹ from unity gain (Vbias= 2.5 V) up to an avalanche gain of ∼ 5 (Vbias = 8 V). Under arestricted subarray and long exposures, we set an implied upper limit on intrinsic dark current in the SAPHIRA of0.0015 e⁻ s⁻¹ pix⁻¹. These values are still dominated by glow, NIR illumination generated by the readoutintegrated circuit.
... Selex ES has over 14 years' experience in the MOVPE growth of MCT by the interdiffused multilayer process (IMP) on GaAs substrates [1]. MOVPE allows control of the alloy composition (x) across the entire compositional range between CdTe and HgTe without any changes to the reactor growth conditions, allowing heterostructure and bandgap engineering approaches to achieve innovative device designs, such as the separation of the gain and absorption regions in APDs used in adaptive optics applications, achieving high breakdown voltage while maintaining high gain [2]. Extrinsic doping control over the range ~10 15 cm -3 -mid×10 17 cm -3 is achieved using arsenic and iodine as the acceptor and donor dopant respectively. ...
... No significant high-noise tail is evident and the median value is 0.8 electrons. Measurements by ESO have shown that the readout noise can be reduced to close to just 0.2 electrons rms with 15 Fowler pairs [2]. The breakthrough results achieved with MOVPE grown HgCdTe eAPD arrays hybridised to the SAPHIRA ROIC have shown that single photon detection can be achieved without the need for deep cooling beyond 80 K and with very low defect levels. ...
... Designed and fabricated by Leonardo UK, formerly Selex, the Saphira detector is designed for high speed infrared applications and is the result of a development program alongside the European Southern Observatory on sensors for astronomical instruments [2], [3], [4]. It delivers world leading photon sensitivity of <1 photon rms with Fowler sampling and high speed non-destructive readout (>10K frame/s). ...
After the development of the OCAM2 EMCCD fast visible camera [1] dedicated to advanced adaptive optics wavefront sensing, First Light Imaging moved to the SWIR fast cameras with the development of the C-RED One and the C-RED 2 cameras. First Light Imaging's C-RED One infrared camera is capable of capturing up to 3500 full frames per second with a subelectron readout noise and very low background. C-RED One is based on the last version of the SAPHIRA detector developed by Leonardo UK. This breakthrough has been made possible thanks to the use of an e-APD infrared focal plane array which is a real disruptive technology in imagery. C-RED One is an autonomous system with an integrated cooling system and a vacuum regeneration system. It operates its sensor with a wide variety of read out techniques and processes video on-board thanks to an FPGA. We will show its performances and expose its main features. In addition to this project, First Light Imaging developed an InGaAs 640x512 fast camera with unprecedented performances in terms of noise, dark and readout speed based on the SNAKE SWIR detector from Sofradir. The camera was called C-RED 2. The C-RED 2 characteristics and performances will be described. The project leading to this application has received funding from the European Union's Horizon 2020 research and innovation program under grant agreement N° 673944. C-RED2 development is supported by the "Investments for the future" program and the Provence Alpes Côte d'Azur Region, in the frame of the CPER.
