Abstract: In a CCD type photosensitive device, the charges produced in two consecutive columns of pixels are transferred into different reading registers: the charges from the first column are loaded into the first register and the charges from the second column travel through the first register to be loaded into the second register. The two reading registers are controlled by independent potentials during the step for the loading of these registers. The device makes it possible to increase the efficiency of the transfer between the two reading registers, especially when the registers are of the type working in a two-phase mode.

Abstract: A method of detecting electromagnetic radiation imparted onto a matrix of photosensitive photomos networks in which pixels of each photomos network are simultaneously exposed to electromagnetic radiation source for a predetermined period of time. Thereafter, transfer signals are applied to each photomos pixel and accumulated charges corresponding to the strength of the radiation imparted onto the pixels are transferred in a columnwise fashion to the end of each column of the photomos networks. The charges are summed at the end of the columns and placed in a reading register. The charges in the reading register of each photomos network are summed and a signal is output corresponding to the strength of the electromagnetic radiation imparted onto the photomos network.

Abstract: A sampling circuit which obtains a level of sampled-and-held signal which is determined with respect to a well-determined reference level V0, even though the input signal is a useful signal referenced with respect to a low-stability reference level. This is the case in particular for sampling of signals derived from charge-transfer photosensitive devices for which the dark level can vary. The circuit includes a sample-and-hold device (EB1) and an input via a capacitor (C1), with a reset circuit which periodically charges the capacitor (C1) to a value which is roughly the difference between the (variable) input reference level and the (fixed) output reference level. According to the invention, it is provided that the reset circuit comprises a looped amplifier (AD1) in which the loop (B1, EB2, B2) is designed to introduce a voltage level shift equal to the shift introduced intrinsically by the sample-and-hold device (B1, EB1, B2).

Abstract: The invention relates to the encapsulation of integrated circuits, and more particularly encapsulation in a multi-layer ceramic case. In order to permit the disposition of chips of variable size on a monolithic chip (22, 24, 26) reception site (23, 25, 27) without the risk of having excessively long connection wires between the chip and the conductive regions (44) surrounding the reserved site, it is proposed according to the invention to cover the chip-reception site with a number of mutually insulated conductive contacts which can serve as soldering relays for these connection wires (80, 90, 100). If the chip is large (chip 24) it is bonded or soldered to these contacts; if it is small (chip 22) it is surrounded by relay-contacts.

Abstract: A method of detecting electromagnetic radiation imparted onto a matrix of photosensitive photomos networks in which pixels of each photomos network are simultaneously exposed to electromagnetic radiation source for a predetermined period of time. Thereafter, transfer signals are applied to each photomos pixel and accumulated charges corresponding to the strength of the radiation imparted onto the pixels are transferred in a columnwise fashion to the end of each column of the photomos networks. The charges are summed at the end of the columns and placed in a reading register. The charges in the reading register of each photomos network are summed and a signal is output corresponding to the strength of the electromagnetic radiation imparted onto the photomos network.

Abstract: A current source delivering a constant current despite sudden voltage fluctuations which may be applied on its output. This current source includes two branches in parallel, a generating branch and a reference branch. The output current is kept constant by keeping the potential difference across the terminals of the resistors in the generating branch constant, by using a differential amplifier. All the transistors are of the NPN type.

Abstract: The invention relates to lasers, and more particularly to solid state lasers such as neodymium on lithium niobate crystal lasers. The invention provides a structure with several juxtaposed optical waveguides (12) sufficiently close to each other laterally to be coupled laterally. In order to take into account the risks of inhomogeneity in the refractive indices or inaccuracies in the dimensions which would alter the distribution of the electromagnetic field of the propagation supermode in the coupled structure, it is proposed to act on the value of the index of propagation in each guide using an individual control (generally electrical). Because of the variation in index, individually controlled, phase differences in the laser wave between the different guides are produced and these phase differences react on the composition of the electromagnetic field defining the supermode. It is possible to go as far as modifying the directivity of the laser beam by this electrical control.

Abstract: In a high gain differential amplifier, in order to obtain a high gain-band product, this amplifier is provided with a controlled feedback in current. In each of the channels of the amplifier, a current generator connected to the output of the amplifier draws a current through the load resistor of the channel. Each generator is constituted by a current mirror. Application to fast analog circuits.

Abstract: The disclosure relates to current mirrors in which a high level of copying error may arise out of the collapse of the gain of the transistors. The mirror includes, in its output arm, a "Darlington" type amplifier subjected to feedback by a buffered mirror. The error gets cancelled for a gain .beta.=1. A second Darlington amplifier mounted symmetrically with the first Darlington enables the V.sub.CE values of the transistors to be balanced. The disclosed device can be applied to current mirrors when the transistors are low-gain transistors (.perspectiveto.1).

Abstract: A light signal receiver in which an input of a preamplifier receives a so-called useful electric current output by a light-sensitive device. The preamplifier comprises an input stage having an input transistor assembled with a "common base" whose emitter is connected to the input of the preamplifier and whose base is kept at a fixed potential. A low input impedance and therefore a reduction of the influence of stray capacitance exhibited by the light-sensitive device results from this arrangement Further, a second input transistor can be utilized so that the preamplifier operates as a differential type preamplifier.

Abstract: Infrared detectors whose detecting part consists of a layer of pyroelectric material deposited on a semiconductive substrate are disclosed. The object is to facilitate the polarization of the layer of pyroelectric material after it is deposited on the semiconductive substrate. The detector of the invention includes a semiconductive substrate (15) carrying a layer of pyroelectric material (11) in which are defined zones, each corresponding to a pixel (P1 to P9). Each pyroelectric zone is defined by an electrode known as lower electrode (EI 1 to EI 9), in contact with a face of pyroelectric layer (11) oriented toward substrate (15). According to a characteristic of the invention, a diode known as polarization diode (DP) is formed in substrate (15) for each pixel (P1 to P9). The polarization diode is connected to lower electrode (EI 1 to EI 9) to make possible the passage of a current between substrate (15) and the lower electrode.

Abstract: The invention relates to photosensitive semiconductor devices and, more particularly, to linear arrays having several parallel rows of photoconductive points and operating in the integration and charge carry mode. In particular, the object of the invention is to reduce a smearing effect.The device of the invention comprises a photosensitive surface (SP) divided into photosensitive surface elements (SI1 to SMn) placed in rows (L1 to Ln) and in columns (C1 to CM). Each column forms a shift register that ends in a storage space (CS1 to CSM) of a readout register (RL) formed by a shift register of the charge transfer type: readout register (RL) being on same semiconductor substrate (10) as photosensitive surface (SP) the device includes.According to a feature of the invention, the device an intermediate zone (ZI) protected from light used to make a separation distance (DS) between photosensitive surface (SP) and readout register (RL).

Abstract: The invention relates to an amplifying circuit with controlled gain, the gain of which is an exponential function of a gain control voltage. The circuit includes two identical amplifiers A1 and A2, the first constituting the basic amplifying cell of the circuit (input Ie, output Is), and the second receiving a reference signal (Iref). The output from the second amplifier, which is the product of the input Iref times the gain G, serves as input to a feedback control loop for the gain G of the amplifiers. This loop receives a set point voltage Vc with a view to establishing a gain G which is an exponential function of Vc. A large-gain differential amplifier AD constitutes the core of this loop: the differential voltage between its inputs tends to be held at zero by the feedback control loop. According to the invention, an element (FTL) with logarithmic transfer function is inserted into the loop between the output of the second amplifier A2 and the input of the large-gain differential amplifier AD.

Abstract: A photosensitive sensor including multilinear arrays of n lines of cells which operate by charge integration and time delay by successive observations of the same image line by the n lines in the array, and by accumulation of the corresponding n observations in the cells of the sensor.To avoid the risk of saturation inherent in the fact that the electrical charges resulting from n observations accumulate in the same cell, the number of lines between which the charge integration and delays are preformed is programmed to be between 1 and n. This programming is performed as a function of the average illumination of the image analysed. To do so, the electrodes (ES1, ES2) of certain lines of cells are used, and potentials are applied to them; either the normal charge transfer potentials or a blocking potential which isolates the lines beyond this blocking electrode from a reading stage (RL, CL).

Abstract: Disclosed is a reading register of the type formed, firstly, by a shift register of the charge-coupled device type working according to a mode other than a two-phase mode and, secondly, an output circuit. It is notably an object of the invention improve the dynamic range, in amplitude, of the output signal delivered by the output circuit for supply voltages. The reading register has a substrate, bearing a shift register and an output circuit, separated from each other by an output gate. The shift register includes a sequence of transfer stages, each having at least three electrodes receiving transfer pulses of different phases, this sequence of stages being separated from the output gate by a last transfer stage or output transfer stage having at least three successive electrodes. According to one characteristic of the invention, the last two electrodes of the transfer stage are connected to each other and receive one and the same voltage pulse.

Abstract: Disclosed is a charge-transfer or charge-coupled (CCD) type device, aimed at increasing the charge storage capacity. The disclosed photosensitive device includes a plurality of photosensitive cells. Each photosensitive cell includes an elementary photosensitive zone beneath which there is formed a potential well. According to one characteristic, the potential well has a surface area greater than the surface area of the elementary photosensitive zone. The disclosure can be applied to charge-couped devices that include a non-photosensitive zone between the pixels.

Abstract: Disclosed is an amplifier circuit with gain controlled by a current or a voltage. It is sought to obtain a linear gain as a function of the set value current or voltage, whereas amplifiers with controlled gain usually have a non-linear variation of the gain as a function of gain control voltage. According to the invention, there is proposed an amplifier circuit with the particular feature of having two identical amplifiers with controlled gain: one receives the signal to be amplified and gives an amplified output signal; the other receives a reference voltage or current and gives an output which is compared with a signal with a set value of variation in gain. The result of the comparison is used to automatically control the gain of the second amplifier to adjust its output to a desired value. The first amplifier receives the same automatic control signal as the second one.

Abstract: Disclosed is an analog/digital converter designed to convert an analog voltage into a digital voltage with n bits. The converter comprises:p=2.sup.n comparators capable of comparing the analog voltage with a reference analog voltage obtained by a voltage generator, each comparator comprising an inverter amplifier, a capacitor having a first terminal connected to the input of the amplifier and wherein, before the application of the analog voltage, each amplifier is connected in short-circuit feedback loop while the reference voltage is applied to the other terminal of the capacitor, this operation being performed no longer in parallel for all the amplifiers but successively, one after another in order to reduce the consumption of current during this operation.

Abstract: According to the invention, a first layer of conductive material (11) is submitted to an incomplete etching operation in the presence of a mask (13). After elimination of the mask, a second layer of conductive material is deposited, and the thus-obtained result is submitted to an etching operation without a mask, so allowing the inter-electrode gaps to be reduced.The process provides a very tight electrode configuration, and is particularly suited to charge-coupled devices.

Abstract: A furnace for soldering integrated circuit chips on a ceramic substrate has a heating plate supporting the substrate, the plate comprising perforations at the places where the integrated circuit chips will be soldered, behind the plate, a heat radiation source and a device to direct the radiation as desired towards only one of the apertures of the plate at a time.