Classifications

• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
  • H04N25/70—SSIS architectures; Circuits associated therewith
  • H04N25/701—Line sensors
  • H04N25/7013—Line sensors using abutted sensors forming a long line
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
  • H04N25/70—SSIS architectures; Circuits associated therewith
  • H04N25/71—Charge-coupled device [CCD] sensors; Charge-transfer registers specially adapted for CCD sensors
  • H04N25/711—Time delay and integration [TDI] registers; TDI shift registers
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
  • H04N25/70—SSIS architectures; Circuits associated therewith
  • H04N25/71—Charge-coupled device [CCD] sensors; Charge-transfer registers specially adapted for CCD sensors
  • H04N25/713—Transfer or readout registers; Split readout registers or multiple readout registers
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
  • H04N25/70—SSIS architectures; Circuits associated therewith
  • H04N25/76—Addressed sensors, e.g. MOS or CMOS sensors
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
  • H04N25/70—SSIS architectures; Circuits associated therewith
  • H04N25/76—Addressed sensors, e.g. MOS or CMOS sensors
  • H04N25/78—Readout circuits for addressed sensors, e.g. output amplifiers or A/D converters
• • H—ELECTRICITY
  • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
  • H10F—INORGANIC SEMICONDUCTOR DEVICES SENSITIVE TO INFRARED RADIATION, LIGHT, ELECTROMAGNETIC RADIATION OF SHORTER WAVELENGTH OR CORPUSCULAR RADIATION
  • H10F39/00—Integrated devices, or assemblies of multiple devices, comprising at least one element covered by group H10F30/00, e.g. radiation detectors comprising photodiode arrays
  • H10F39/80—Constructional details of image sensors
  • H10F39/809—Constructional details of image sensors of hybrid image sensors

Definitions

• the present invention relates to high speed imaging using time delay integration (TDI) sensors and in particular to an analog front end (AFE) and an analog to digital converter (ADC) that can be used in conjunction with the TDI sensors to form a high density digitizer (HDD) .
• TDI time delay integration
• AFE analog front end
• ADC analog to digital converter
• Time delay integration is an image scanning process that produces a continuous Image of a moving two- dimensional object.
• imag photons ar
• the TDI can integrate signal intensity at a fixed position on the moving object to generate the imag .
• the total integration time can be regulated by changing the speed of the image motion and providing more/less pixels in the direction of the movement, TDI inspection systems can be used for
• [00053 ⁇ 3 ⁇ 4 module fo high speed image processing can include an image sensor and a plurality of high density digitizers ⁇ HDDs) .
• the image senso can generate a plurality of analog outputs representing an image.
• the image sensor can include a time delay integration (TDI) sensor, which can sense a range of wavelengths from deep ultraviolet through visibl radiation.
• TDI time delay integration
• the HDDs can concurrently process th plurality of analog output .
• Each HDD can b implemented as an integrated circuit, Hotably, each HDD can process in parallel a predetermined set of the analog outputs representing a portion of the image. These HDDs can achieve high average data rates while maintaining high signal-to ⁇ noise ratios.
• Each channel of the HDD can include an analog front end (AFE) and an analog- o-digital converter (ADC) ,
• the AFE can condition a signal (in one embodiment , a differential signal) representing one sensor analog output,
• the ADC can convert the resulting conditioned signal into a digital signal ,
• the AFE can include a programmable gain amplifier (PG&) with switch-out capacitors.
• the PGA can include a plurality of comparators for determining when each of the switch-out capacitors is to be disconnected from an input of the PGA,
• HDD high definition HDD
• a data rate multiplier phase locked loop PLL
• the AFE can be configured to convert a single-ended signal to a differential signal, which has high immunity to substrate noise and also increases the swing of the signal and enhance the signal-to-nois ratio ( SNR) of the system.
• the AFE can also include a correlated double sampling ⁇ CDS ⁇ circuit with offset control to optimize dynamic range ,
• the clocking of the CDS circuit can be reconfigurabie , thereby allowing one reset and multiple readings to provide averaging and increase of the system SMR.
• a data formatting block of the HDD ca be configured to provide black-level correction.
• the HDD can further include a low voltage
• each channel can include an analog driver coupled to bypass the ADC and provide the conditioned signal to an o f-chip device,
• the HDD can further include a control block for enabling/disabling a calibration mode and a test mode.
• the control block can include a ramp generator for providing a ramp signal to each ADC,
• the control block can include self-test logic configured to introduce a predetermined ramp function to each channel and monitor output pins of the HDD for deviations from the
• predetermined ramp function to introduce a DC value to each channel and monitor output pins of the HDD for noise on each channel , and/or to introduce a known signal pattern to each channel and monitor output pins of the HDD to determine when the known signal pattern starts and ends.
• HDD can also include a sensor block for selectively accessing and monitoring a digital input voltage, an inner chip voltage, a peripheral chip voltage, and a temperature sensor voltage.
• This sensor block can include a temperature sensor that can measure precisely the on-chip temperature. Because multiple HDD dies can be provided in one package, th sensor block can be used to generate a thermal map of the HDD package, which can then be used for debugging purpos .
• the HDD can further include a register control block for providing general and channel configuration bits to the HDD-
• the register control block can provide interlinking of the plurality of HDDs.
• the configuration bits of the registers of the register control block can be
• a system for high speed image processing is also described.
• This system can included a plurality of modules configured as described above.
• the plurality of HDDs can be connected for selectively providing general and channel
• the system can included a package for securing the image sensor and the plurality of HDDs , These HDD dies can be attached to the package through standard "bum s", i.e. flip-chip technology, in one embodiment, the image sensor and the plurality of HDDs can be coupled using wire bonds, wherein the package includes ill- package capacitors . These in-package capacitors form part of the overall grounding scheme of the package and can
• the module and system described herein can achieve high signal integrity, overall reliability, and lower material and assembly costs.
• FIG. 1 illustrates a top view of an exemplary TDI sensor module that includes localized driving and signal processing circuitr .
• FIG. 2 illustrates an exemplary modula array of TDI sensor module .
• FIG. 3 illustrates an exemplary inspection technique using TDI sensor modules
• FIG. 4 illustrates exemplary inputs and outputs of two high-density digitizers r • which form par of the processing circuits of a TDI sensor module.
• FIG. 5 illustrates a functional block diagram of an HDD.
• FIGS. 6A and 6B illustrate the channels and pins of exemplary HDDs .
• FIG. 7 illustrates an exemplary register control block, which can provide both general and channel configuration bits to the HDD.
• FIG . 8A and 8B illustrate exemplary configurations in which a plurality of HDDs can be connected in series or in parallel, respectively, to transfer the general and channel configuration bits.
• FIG. 9A illustrates an exemplary CDS t that be configured in a reset mode or a sampling mode.
• FIG. 9B illustrates an exemplary programmable gain amplifier 920 that can form part of an ⁇ for the HDD.
• FIG. 10 illustrates an exemplary senso block.
• FIG. 11 illustrates an exemplary ADC auto-test con iguratio .
• FIG. 12A illustrates exemplary signal and ground paths of an image sensor as well as the signal and DC current paths of a package without in-package capacitors ,
• FIG. 12B illustrates exemplary signal and gro3 ⁇ 4nd paths of an image sensor as well as the signal and DC current paths of a package with in-package capacitors.
• FIG. 13 compares conventional single-signal
• FIG. 14 illustrates that a CCD sensor can be timed produce different waveforms for the same image signal.
• FIG. 1 illustrates a top view of an exemplary TDI sensor modul 100 that includes localized driving and signal processing ci cuitry (also called localized circuits herein) , Specifically, TDI sensor module 100 includes a TDI sensor 102, processing circuits 103 for processing the signals from TDI sensor 102, timing and serial drive circuit 104, and pixel gate driver circuits 105.
• localized driving and signal processing ci cuitry also called localized circuits herein
• TDI sensor module 100 includes a TDI sensor 102, processing circuits 103 for processing the signals from TDI sensor 102, timing and serial drive circuit 104, and pixel gate driver circuits 105.
• processing circuit 103 can provide correlated double sampling (CDS) and other analog front end (AFE) functions (e.g. analog gain control or DC offset), analog to digital conversion (ADC) and digital post-processing such as black-level correction, per pixel gain and offset corrections, linearity corrections, look-up tables (LUTs) , and data compressio .
• CDS correlated double sampling
• AFE analog front end
• ADC analog to digital conversion
• digital post-processing such as black-level correction, per pixel gain and offset corrections, linearity corrections, look-up tables (LUTs) , and data compressio .
• the processing may be ixed or rely on additional, possibly real-time, input from the inspection system to perform functions such as sub-pixel interpolatio , analog gain control to prevent digital saturation, image position shifting, and image spatial distortion correction.
• the timing and serial drive circuits 104 can control clock timing and drive for TDI .
• Features such as reset pulse generation, raulti-phase serial-register clock generation, and ADC synchronization raay be included. This allows for very accurate timing which is needed to achieve high SNR at high clocking speeds .
• the pixel gate driver circuits 105 provide slower but higher-current TDI gate drive signals to synchronise data capture with the inspection image motion and with other TDI sensors .
• Pixel gate driver circuits 105 may typically provide three-phase or four-phase drive waveforms of square-wave and/or sinusoidal waveforms. More generally, pixel gat driver circuits 105 may use digital-to-analog conversion to provide arbitrary function generation in order to optimise the charge transfer ? thermal dissipation, and S R of th sensor.
• each of processing circuits 103, timing and serial driv circuits 104, and pixel gate drive circuits 105 can be implemented on integrated circuits (ICs) positioned around TDI sensor 102 on a PCB (printed circuit board ⁇ 101. Note that the number of ICs used to implement the driving/processing circuits can vary based on embodiment. In one embodiment ? PCB 101 can be implemented using a multi-layer, ceramic substrate .
• digital data from TDI sensor module 100 can be transmitted o f-board using a programmable, low voltage differential signaling (LVBS) , or similar
• LVBS low voltage differential signaling
• Th specific protocol can be selected from an industry standard or
• drive programmability can be added to reduce the digital noise to the LVDS for specific package traces.
• FIG. 2 illustrates an exemplary modular array 200 of TDI sensor modules 201 ⁇ hereinafter called a modular sensor array) .
• the driving/processing circuits positioned around the TDI sensor take ti a predetermined space , T us , the TDI sensors in adjacent rows can be aligned such that at least 100% image coverage is achieved when used in a continuous scanning configuration.
• each row can be offse with respect t an adjacent row such that the TDI sensor is positioned in the same vertical the driving/processing circuits of an adjacent row.
• the width of each TDI sensor is equal to or greater than the space between TDI sensors. I this configuration, as the inspected
• modular sensor array 200 can ensure at least 100 image captu e ,
• the effective data rate for modular array 200 can be significantly higher than a single,, large TDI sensor. This rate is achieved because the modular array can have an effective total size and number of output channels that is larger than can be practically manufactured in a single TDI sensor. Further note that any number o rows o TDI sensor modules can be included in a modular array, i.e. TDI sensor modules f cilitate scaling .
• FIG. 3 illustrates an exemplary inspection technique using TDI sensor modules.
• Step 301 can position an object (e.g. a reticle , mask, or integrated circuit) for inspection.
• Step 302 can begin inspection of the object, for example using a microscop .
• Step 303 can generate TDI sensor outputs , i.e. multiple analog outputs .
• Step 30 can condition these analog outputs and step 305 can adjust the programmable gain of these outputs, if necessary to optimize the conditioning.
• Step 306 can perform correlated double sampling (CDS) , which is a known process that measures electrical values (i.e. voltages or currents) in order to remove an undesired offset.
• CDS correlated double sampling
• the first measurement can then be subtracted from the second measurement, thereby providing a value that can be used to correct for offset.
• Step 307 ca perform analog-to-digital conversion (ADC) using the of set-co ec ed measurements.
• Step 308 can process the digitized data into streams of data.
• programmable logic devices such as field
• Step 309 can format the streams of calibration data and transport the resulting signals to a processing device.
• Step 310 can process and digitize th data into images and perform analysis of th images . Exemplary analysis includes defect inspection and/or feature
• Step 311 can store the images, while step 312 can display one or more images for user review.
• TDI sensor module 100 can implement steps 303-309, wherea other components in an inspection system can implement steps 301-302 and 310-312.
• FIG. 4 illustrates a portion of a TDI sensor module, specifically two high-density digitizers (HDDs) 02A and 402B, which can form part of processing circuits 103, and a TDI senso 401,
• HDDs 02& and 402B for example, HDDs 02& and 402B (for example, HDDs 02& and 402B (for example, HDDs 02& and 402B (for example, HDDs 02& and 402B (for
• HDDs 4Q2A and 402B can generate digital outputs 403A and 403B, respectively, which pertain to specific regions of the total image.
• HDDs 402A and 402B can also generate control and timing outputs 405.
• digital outputs 403A and 403B which are output in parallel, can provide a large optical imag field of view for inspection. Generating multiple digital outputs in parallel also facilitates achieving high data rates.
• one conventional configuration for reading an output of an image sensor shifts a row of digital (pixel) data to a shif register, which in turn shifts out data serially one bit (one pixel) at a time.
• each HDD can output multiple sets of digital bits ⁇ pixel values ⁇ in parallel.
• these parallel outputs allow relatively slow operation per channel f thereby maximizing the signal-to-nois ratio ⁇ SNR) , while allowing a very high total system data rate for all channels (i.e. based on a plurality of bits ⁇ pixel data) being effectively output in parallel ⁇ .
• FIG. 5 illustrates a functional block diagram of a high density digitizer (HDD) 500.
• HDD device 500 can include timing and control logic 502 that receives the previously- described timing and control inputs 40 ⁇ FIG. 4) .
• timing and control logic 502 can provide timing and control inputs 404 to one or more channels ⁇ described below) to compensate for propagation delay or other local requirements .
• Timing and control inputs 404 can also be provided to
• HDD device 500 can also include self-test logic 503 that receives some of timing and control inputs 404 and generates some of timing and control outputs 405.
• self-test logic 503 that receives some of timing and control inputs 404 and generates some of timing and control outputs 405.
• timing and control outputs 405 can be used for monitoring and/or controlling other TDI sensor modules in an efficient manner
• HDD 500 can include blocks that perform steps 304- 309.
• signal conditioning block 510 can perform step 304
• CDS block 511 can perform step 306
• ADC block 512 can perform ste 307
• calibration / data formatting block 513 can perform ste 309
• signal transport block 514 can perform step 309. Note that only components performing steps 304 , 306, 307, and 309 are shown fo simplicity. In an actual
• HDD 500 can include additional components to perform other steps and provide additional functionality.
• blocks 510-514 can perform additional functionality. For example, in one
• signal conditioning block 510 can provide current sinking, level shifting for the voltage domain, offset level adjustment, buffering, single-ended to differential conversion, and robust electrostatic discharge damage (BSD) control.
• BSD electrostatic discharge damage
• Correlated double sampling ⁇ CDS block 511 can remov several sources of noise found in sensor processing, e.g. low-frequency noise and/or reset noise, i3 ⁇ 4DC block 512 can advantageously digitize signals from multiple channels ⁇ e.g. the 8 exemplary channels show for purposes of illustration) in parallel, thereby providing uniform performance across HDD 500. This uniform performance is particularly desirable in an inspection or metrology system.
• Calibration data formatting block 513 can perform real-time processing, such as black-level correction, drift-compensation, and/or other calibration processes known in the art of high-performance imaging. Calibration data
• formatting block 513 can also perform digital signal processing calculations that access previous measurements and predict future data.
• the results of this real-time processing ca be fed back into the analog front-end ⁇ i3 ⁇ 4FE) ⁇ e.g. signal conditioning block 510 and CDS block 511) to control conditioning, gain control, and sampling as needed.
• such real-time processing can be performed b one or more PGAs , CPUs , or dedicated processing devices ⁇ i.e. external devices),
• Signal transport block 514 can receive the formatted data from data formatting block 513 and generate digital outputs 504.
• sets of digital outputs 504 can be created (e.g. by merging channel data) with minimal data accuracy loss.
• merged digital output can affect the quality of the analog input by coupling through the silicon substrate as well as I/O ring and ESD devices .
• FIG. 5 shows sets 506A and 506B that could be merged to rm digital outputs 504.
• This merged digital data c n be sent at high data rates ⁇ e.g. using high speed interconnect)
• Merged digital data also advantageously uses fewer pins of HDD 500 (e.g. in the case of either set 506 ⁇ or 506B, one fourth of the original pin count ⁇ .
• using localized ADC and output multiplexing in HDD device 500 can significantly improve system performance and resource management.
• FIG. 6A illustrates the channels and pins of an exemplary HDD 600.
• HDD 600 includes 16 channels, i.e. channels 601-616, other HDDs may include fewer channels ⁇ e.g. 8 channels) or more channels ⁇ e.g. 32 or 64 channels) .
• Exemplary components of the channels are shown in channel 601.
• each channel can include an analog front end 621
• AFE 621 can include one or more analog filters for smoothing waveforms and/or removing DC levels , CDS circuitr , single-ended to diff rential
• An ADC 622 can convert the processed analog signals from AFE 621 into digital signals.
• the digital outputs from ADC 622 can be stored in a shi register 623.
• a low voltage differential signaling ⁇ LVDS ⁇ block 624 can receive the outputs of shift register 623 and then generate two different voltages GUTP and GUTN, which can be compared off-chip.
• channel 601 can output OUTIP and OUT1H
• channel 616 can output OUT16P and O0 16N.
• LVDS 624 can effectively use this difference to encode in ormation. Mot that LVDS is a standard output format for high speed networks/busses and therefore is not described further herein.
• FIG. 6B illustrates the channels and pins of another exemplary HDD 600' .
• each channel further includes an analog driver 625, which has its inputs connected to the outputs of AFE 621 and its outputs connected to the output pins (i.e. the pins providing OUTP and OUTN) .
• an analog driver 625 which has its inputs connected to the outputs of AFE 621 and its outputs connected to the output pins (i.e. the pins providing OUTP and OUTN) .
• ADC 622, shift register 623, and LVDS 624 can b bypassed, thereby allowing analog signals to b output by the channel.
• analog or digital data can be sent to the next
• HDD 600 (or HDD 600' ⁇ can further includes a register control block 632 that can receive digital input signals DLDI (digital load data input) , DI ⁇ digital read data input) , DIN ⁇ data in) , and DCKI ⁇ digital clock input) . These digital input signals can traverse HDD 600 (in various manners, as described below i FIGS. 7, 8A, and 8B) and eventually are output as digital output signals DIDO (digital load data output) , DRDO (digital read data output) , DOD (data out) , and DC O (digital clock output) .
• Additional circuits integrated on HDD device 600 can include clock circuitry 631 and an associated data rat
• multiplier phase locked loop ( PLL) 632 can provide delays for clock signals when necessary, e.g. to adjust the delay to when the actual CDS occurs ⁇ as can be indicated by a received trigger signal) .
• PLL phase locked loop
• data rate multiplier PLL 632 can ensure that all channel outputs are phase locked to that square wave (or a multiplie of that square wave) .
• Biasing circuitry 641 can generate voltage bias VB (see FIG. 9&) as well as allow the differential amplifier to handle a unipolar signal range (i.e. 0 to max) instead o a bipolar signal range ⁇ i.e. —max/2 to +max/2 ) .
• each channel may have a separate bias control capability.
• Digital control 642 can enable/disable analog driver 625 (FIG. 6B) , the digital driver associated with LVDS 62 , the output pins (e.g. to save power), calibration, and/or test modes or the channels .
• a exemplary calibration could include introducing a predetermined ramp function on the input pins of HDD device 600 and monitoring the output pins to ensure that the same ramp function is output.
• calibration could include introducing a DC value on the input pins of HDD device 600 and monitoring the output pins for noise on each channel.
• Exemplary test modes include providing a known signal pattern to the channels and analyzing the digital output, thereby facilitating identification of when that pattern starts and ends. Once the start/end of a known pattern can be ascertained, then the channels can be programmed
• digital control block 642 can include a bit clock LVDS ⁇ associated with the on- chi bit line) and a word clock LVDS (associated with the on- chip word line) , which can be used in well known techniques to synchronize on-chip data (e.g. during test mode and/or during calibration of specific channels) ,
• FIG, 7 illustrates an exemplary configuration for register control block 633, which can be used to provide both general configuration bits and channel configuration bits to the HDD,
• FIG. 7 f two sets o registers are provided: a first set 701 for the general conf guration bits and a second set 702 for the channel configruration bits .
• a multiplexer 730 can be used to write bits DIN to either general configuration registers 711 (which may control the overall timing of the chip, the enabling/disabling of the output drivers , etc.) or channel configuration register 721 (whic may control the gain for each channel or other channel- control signals) . Circuits in the HDD can access the bits of configuration registers 711 and 721 via standard techniques. Bits DIN for general configuration registers 711 can be loaded into shift registers 712 using the clock signal DC I on a clock line 71 . Once all the general con guration bits are clocked into shi registers 712 f the load signal DLDI on line 714 (i.e.
• DLDI can trigger general con iguration registers 711 to receive the values in shift registers 712 in parallel.
• bits DIM for channel configuration registers 721 can foe loaded into shift registers 722 using the clock signal DCKI on a clock line 723, Once all the channel configuration bits are clocked into shift registers 722 , the load signal DLDI on line 724 can trigger general configuration registers 711 to receive the values in shift registers 712 in parallel .
• the fir t and second sets of regxsters 701 and 702 may have the sam number of registers in some embodiments, in other embodiments the first and second sets of regxsters 701 and 702 can have different numbers of registers depending on circuits implemented on the HDD,
• multiplexer 731 can read the bits DOUT from the first set of registers 701, i.e. the general conf guration bits, or from the second set of registers 702, i.e. the channel configuration b ts. Specifically, the read signal DRDI can trigger shift registers 712 to load th values from general configuration registers 711, wherein such bits can then be clocked out of shifter register 712 using the clock signal DCKI and appropriate output selection by multiplexer 731.
• the read signal DRDI can trigger shift registers 722 to load the values from channel configuration registers 721 , wherein such bits can then foe clocked out of shifter registers 722 using the clock signal DCKI and appropriate output
• FIG, 8A illustrates an exemplary configuration in which a plurality of HDDs 801-809 can foe serially connected to receive inputs 810 for the general control circuitry or the channel control circuitry (e.g. using multiplexers 730 and 731 , FIG. 7) .
• the outputs from HDDs 801-808 form the inputs to HDDs 802-809.
• 9 chips are indicated in FIG. 8A, other embodiments may hav more or fewer HDDs in a daisy chain configuration.
• Each HDD may also be given a u ique fixed input pattern on CIDO - CID4 , which can be used to uniquely identify and address each device. The pattern may generated by
• SPIJMODE input may then be used to select parallel addressing mod for writing and reading register .
• each HDD in the TDI sensor package can be individually addressable, e.g. bits for the general control circuitry and the channel control circuitry can be sent over a bus with a set of lines provided for each HDD.
• This configuration can facilitat th re ogramming of individual HDDs (chips) .
• FIG. 9A illustrates an exemplary CDS circuit 900 that can be configured in a reset mod or a sampling mode.
• Circuit 900 includes a progranaiiabi gain amplifier (PGA) 902, which receives inputs via positive and negative input terminals and generates outputs pgan and V pgap for an ADC.
• PGA progranaiiabi gain amplifier
• transistor 901 which forms part of TDI sensor 401 ⁇ FIG. 4 ⁇ , is connected between a high voltage source (e.g. VDD) and a node 903, Th gate of transistor 901 receives a voltage proportional to the detected signal on the sensor; therefore, its output (provided to node 903) is also proportional to the signal on the sensor.
• VDD high voltage source
• Th gate of transistor 901 receives a voltage proportional to the detected signal on the sensor; therefore, its output (provided to node 903) is also proportional to the signal on the sensor.
• a capacitor 904 and a resistor 905 are each connected between node 903 and ground.
• a capacitor 906 is connected between node 903 and the negative input terminal of PGA 902.
• a capacitor 907 is connected between a switch 908 and the positive input terminal of PGA 902.
• a feedback capacitor 909 and a switch S 1 are connected in parallel to the negative input and output terminals of PGA 902.
• a f edback capacitor 910 and a switch SW2 are connected in parallel the positive input and output terminals of PGA 902 >
• feedback capacitors 909 and 910 are effectively programmable (e.g. via switches SW1 and SW2 ) and their values can change the voltage gain of PGA 902 (in an inversely proportional manner ⁇ .
• bias voltag 3 ⁇ 4 can be programmed to be clos to one extreme ⁇ minus full value ⁇ . Whe half a maximum signal is detected, bias voltage VB can be programmed to be close to zero. When a full signal is detected, bias voltage V B can be programmed to be close to the other extreme (plus full value) , thereby effectively doubling the signal swing.
• FIG. 9B illustrates a simplified exemplary controlled non-linea response programmable gain amplifier 920 that can form part of an &FE for the HDD.
• Vin the input to PGA 920, has an associated AC signal having a "swing" that can be measured.
• Vin must also supply some current load to the sensor. In this embodiment, that current can be provided by a switching
• a plurality o input capacitors 921-924 are connectable in parallel between an input voltage Vin and a negative input terminal of amplifier 900 (see FIG. 9ft for details) , The positive input terminal of amplifier 900 is connected to ground.
• Capacitors 922-924 can be
• Switches 925-927 are controlled by
• Comparators 928-930 receive threshold voltages VT0-VT3 , respectively, on their first input terminals and a voltage from a node 933 on their second input terminals .
• a capacitor 932 is connected between node 933 and ground, whereas a capacitor 931 is connected between node 933 and the input voltage Vin.
• amplifier 900 After a reset of PGA 920 and initially during the sampling mode, switches 925-92? are closed. This configuration generates a maximum gain for amplifier 900, which is computed as the ratio of the total parallel capacitance of capacitors 921-924 (i.e. their summed capacitances) to the capacitanc of its programmable capacitor (i.e. capacito 909, FIG, 9A) . In other words , amplifier 900 has a gain defined by the ratio of its input capacitance and its feedback capacitance. An input signal Vin is initially low, but transitions higher over time until the next reset ⁇ wherein a reset is triggered for the sampling of each pixel) . During that time, amplifier 900 is effectively integrating that signal and generating a voltage Vo for an ADC .
• comparators 928-930 are comparing that rising signal to their threshold voltages VTQ-VT2, wherein VT0 ⁇ VT1 ⁇ W2. Once the signal at node 933 reaches each of the threshold voltages, comparators 928-930 will trigger switches 925-927 to
• PGA 920 can reduce the voltage swing of input voltage Vin using the negative feedback provided by the feedback capacitor 909 and the switchable capacitors 922-927.
• the gain for PGA 920 can be defined by
• the gain of amplifier 900 can be changed without changing the offset. This feature is possible because a voltage chang in Vin with capacitors 921- 924 connected to the negative input terminal of amplifier 900 causes current to flow through those capacitors in one
• PGA 920 is particularly beneficial f r a digitizer (i.e. the ADC) .
• CDS 900 and PGA 920 can be configured to output differential signals.
• Providing a differential signal instead of a single signal can provide advantages in a sensor system.
• a single signal is typically understood to be relative to ground.
• ground may actually vary slightly across a chip. Therefore a sensor receiving signal from different areas of the chip may require additional
• a differential signal has two signals, each of which is relative to the other.
• the HDD can be more noise immune (i.e. compared to using a single signal) when using multiple channels on one chip,
• FIG. 10 illustrates an exemplary sensor block 643 (FIG. €A) that can provide high accuracy, low speed (on the order of 10-100 signals per second), and low power results.
• PGA.1003 e.g. of conventional design, can recexve xnputs from a multiplexer 1002 and provides outputs to a sigma-delta ADC 1004.
• sigma-delta ADC 1004 can generate an output, which is accessible by an off-chip device.
• multiplexer 1002 can receive a plurality of inputs, at least one of which can be a temperature sensor 1001 ⁇ depending on the size of the chip) .
• Temperature sensor 1001 can include a transistor, a current source, or some other set of components for sensing temperature that generates a voltage output.
• sensor block 643 can also monitor critical supply voltages, such as a digital supply voltage supplied to the chip, an analog input voltage supplied to the chip (which could be read out in analog format from a test pin, or converted by the on-chip sigma-delta ADC 1004 and then read out digitally ⁇ , a voltage generated in an inne area of the chip, and/or a voltage generated in a periphery area of the chip.
• sensor block 643 can advantageously provide a diagnostic function for the chip including the HDD.
• a TDI sensor module including multiple HDDs e.g.
• sensor block 643 can use differential output of multiplexer 1002 and PGA 1003.
• FIG. 11 illustrates an exemplary ADC auto-test configuration in which a ramp generator 1106, when enabled using an internally-generated signal ENABLE, provides a digital ramp (i.e. saw tooth) signal to adders 1107-1110, Adders 1107- 1110 also receive independent inputs IN1-IN4, respectively.
• inputs IN1-IN4 can foe the channel inputs described above.
• the sums of adders 107-1110 are provided to ADCs 1101-1104, respectively, which in turn provide bit outputs to a mult lexe 1105.
• the ramp signal of rajra generator 1106 is a 10-bit signal and each independent input IH1-IH4 is a 2 ⁇ bit signal, thereby resulting in both adders 1107-1110 and ADCs 1101-1104 outputting 12-bit results.
• a control signal provided by internal control logic can select which ADC result to output as the OUT signal ⁇ logically, a 12- bit signal) . Motafoly, even in the absence of tual channel input, the conf guration of FIG. 11 allows ADCs 1101-1104 to be tested.
• FIG. 12A illustrates exemplary signal and ground paths of an image sensor 1201 ⁇ including components fo
• a set of sensor components to generate a channel can include transistors 1-M3, connected as shown in FIG, 12A) as well as the signal and DC current paths of a package 1202 without in ⁇ package capacitors.
• FIG. 12A these paths are indicated using arrows.
• image sensor 1201 and package 1202 can be connected using wire bonds in an air medium. In other embodime t, wire bonds in epoxy can be used. In either embodiment, the bonding can also include flip chi bonding (i.e. solder bumps connecting to pads), which is well known .
• image sensor 1201 can be supported by package 1202 (e.g. the edges only of image sensor 1201 can be supported by a "frame" package) , but are shown separately in FIG. 12 ⁇ for simplicity.
• FIG. 12B illustrates exemplary signal and ground paths of image sensor 1201 as well as the signal and DC current paths of a package 1203 with in-package capacitors .
• the signal path ⁇ i.e. current extends outside package 1202 to the VDD power supply and then back to sensor 1201 to provide power to sensor 1201.
• This signal path is the same for both DC and high frequency current.
• having high frequency signals traveling that far ⁇ e.g. on the order of 10-30 cm$ is quite problematic and can result in significant performance
• This degradation can include ground voltages ⁇ AVSS and VSS planes, for example) that fluctuate.
• the current path is localized within package 1203 ⁇ e.g. on the order of 1-2 cm ⁇ .
• This shortened path, compared to that of package 1202, can advantageously ensure good performance, e.g. ensuring consistent, stable reference voltages,
• the ADCs of the device can be calibrated using an internally generated reference, or can use a reference from an external connection.
• multiple devices can be connected together with this connection and calibrated to the same to reference for high precision measurement .
• On-chip logic is designed support calibrations of either type in the preferred embodiment
• multipl analog signals can be encoded or mixed before digital conversion, and composite digital data can then foe decoded for improved signal- to-noise ratio.
• multiple digital samples can be taken per pixel (nrultisampling) , thereby
• digital data can fo locally stored and processed for calibration, compression, and pre-processing.
• th results from adjacent channels and a "history" of channel values can be used to provide real time control of the analog and digital
• a signal level can be monitored, wherein when the signal level exceeds a
• the HDD can further include a precision timing delay control to adjust the reset and sampling clock of the &DG. This delay is controlled by digital block.
• the timing control circuit can include a reliable voltage supply generated internally through a
• Th HDD can furthe includ analog multiplexer to read exact DC voltage o the image sensor outputs for optimum setting of the resistor and bias current.
• som local processing to perform defect detection can be performed. For example, in designs with highly repetitive features, adjacent features can be compared, wherein adjacent defect-free features should be identical.
• defect detection can include subtracting the image o one feature from th imag of the adjacent feature, and indicating a defect is detected only when the difference of the images exceeds a predetermined threshold. Output data generation can foe entirely disabled or highly compressed when the difference data is below a predetermined threshold .
• FIG. 1.3 compares conventional single-signal and multi-signal CDS modes.
• REF 1301 arid SIG 1302 waveforms define the reference and signal clocks with one reference and one sample per digital readout.
• REF 1303 is the same as REF 1301, which SIG 1304 clocking produces multiple sample measurements and digital readouts , each readout relative to the reference value collected at t__ref .
• all timing signals may be implemented as differential pairs as shown in REF 1301,
• FIG. 14 illustrates that a CCD sensor can be timed to produce different waveforms for the same image signal.
• CCD waveform 1401 can foe used to collect three signal measurements , all relative to the re erence sampl (REF 1303) .
• CCD waveform 1402 can foe used to collect two reference levels and two signal samples, where the measurement at t_jsigl is relative to the first reference measurement, and would foe 0 counts for an system with no noise sources .

Landscapes

• Engineering & Computer Science (AREA)
• Multimedia (AREA)
• Signal Processing (AREA)
• Transforming Light Signals Into Electric Signals (AREA)
• Studio Devices (AREA)

Priority Applications (4)

Applications Claiming Priority (2)

Publications (1)

Family

ID=48869892

Family Applications (1)

Country Status (7)

Cited By (1)

Families Citing this family (54)

Citations (5)

Family Cites Families (15)

• 2012
  • 2012-02-01 US US13/364,308 patent/US8754972B2/en active Active
• 2013
  • 2013-01-30 TW TW102103563A patent/TWI562094B/zh active
  • 2013-02-01 CN CN201380014322.5A patent/CN104205160B/zh active Active
  • 2013-02-01 JP JP2014555729A patent/JP6231019B2/ja active Active
  • 2013-02-01 EP EP13743326.4A patent/EP2810244A1/en not_active Withdrawn
  • 2013-02-01 WO PCT/US2013/024240 patent/WO2013116579A1/en not_active Ceased
• 2014
  • 2014-05-07 US US14/272,454 patent/US9462206B2/en active Active
  • 2014-07-30 IL IL233867A patent/IL233867A/en active IP Right Grant

Patent Citations (5)

Also Published As

Similar Documents

Legal Events