www.perkinelmer.com/opto l-series linear cmos spectroscopy sensor 25 or 50 m pitch, 2.5 mm aperture, 128 to 1024 elements datasheet lighting imaging telecom imaging product line ? differential video output for clock noise cancellation ? high saturation charge 10 pc (25 m) or 20 pc (50 m) ? antiblooming function for low crosstalk ? line reset mode for simultaneous reset of all photodiodes ? wide spectral response: 300 to 1000 nm ? polished fused silica window ? on-chip diodes (two) for temperature monitoring description the perkinelmer l-series of monolithic self-scanning linear photodiode arrays offers a high quality, low cost solution for spectroscopy and colorimetry appli- cations in the 300 C 1000 nm range. the combination of high sensitivity, low dark current, low switching noise and high saturation charge of the l-series provides excellent dynamic range and great flexibility in setting integration time. l-series sensors consist of a linear array of silicon photodiodes, each connected to a mos switch for readout controlled by an integrated shift register scanning circuit. under external clock control, the shift register sequentially enables each of the switches directing the charge on the associated photodiode to an output line. a dummy output pro- vides clock noise cancellation. the l-series devices are mounted in ceramic side-brazed 22-pin dual-inline pack- ages with ground and polished fused silica windows and are pin-compatible with earlier perkinelmer sb and tb series sensors. features ? 128, 256, 512 or 1024 photodiode elements with 25 m center-to-center spacing ? 128, 256 or 512 photodiode elements with 50 m center-to-center spacing ? 2.5 mm photodiode aperture ? extremely low dark leakage current ? low power dissipation ? clock-controlled sequential readout at rates up to 1 mhz ? single-supply operation with hcmos-compatible inputs ? single shift register design caution: while the l-series imagers have been designed to resist electrostatic discharge (esd), they can be damaged from such discharges. always observe proper esd precautions when handling and storing these sensors. dsp-106.01c - 10/2001w page 1 ? wide dynamic range
figure 2. sensor geometry and aperture response function figure 1. equivalent circuit ? 1 ? 2 start shift register end of scan dummy video reset gate active video reset drain v (substrate) dummy photodiodes active photodiodes 12 c c v v antiblooming gate antiblooming drain 12 n ss n cmos spectroscopy sensor www.perkinelmer.com/opto dsp-106.01c - 10/2001w page 2 sensor characteristics the l-series self-scanning photodiode arrays are supplied in two versions to meet broad application requirements. the models with 25 m diode spacing are applied primarily where a compact array is needed while the 50 m mod- els accommodate applications requir- ing the highest dynamic range. both types are available in 128, 256 and 512 pixel lengths and the 25 m models also include a 1024 pixel version. the simplified equivalent circuit of an l-series sensor photodiode array is shown in figure 1. each cell consists of an active photodiode and a dummy photodiode, both with an associated junction capacitance. these diodes are connected through mos multiplex switches to active and dummy video lines. the shift register is driven by complementary square wave clocks with periodic start pulses being intro- duced to initiate each scan. the pixel sampling rate is determined by the clock frequency. integration time is the interval between start pulses. the out- put signal obtained from each scan of an n-element array is a train of n charge pulses, each proportional to the light exposure on the corresponding photodiode. in addition to the signal charge, switching transients are capac- itively coupled into the active video lines by the multiplex switches. similar transients are introduced into the dummy video lines and, therefore, can be reduced by reading out the video and dummy lines differentially. in all models, the diodes are conti- guous to collect the maximum amount of incoming light. charge generated by light incident on the p-type surface between two n-regions will divide between the adjacent diodes to pro- duce the response function shown in figure 2. the 6 m interdiode gap is maintained in all models to provide a uniform aperture response function across the product line. the quantum efficiency of all models is shown in figure 3. the smooth curve is the basic response of the pho- todiode array showing a peak quantum efficiency approximately 60% at 650 nm. n-silicon n - 1 n n + 1 p-silicon response diode n position of incident light spot 100% 50% silicon dioxide a c 2.5 mm n - 1 n + 1 0.5 mm b a b c 25 m 50 m 19 m 6 m 44 m 6 m figure 3. quantum efficiency 70 60 50 40 30 20 10 0 300 400 500 600 700 800 900 1000 qe (%) wavelength (nm) detail smooth
cmos spectroscopy sensor www.perkinelmer.com/opto dsp-106.01c - 10/2001w page 3 interference effects in the transparent passivation layer over the diodes mod- ulate this response. since the details of this modulation can vary from sensor to sensor, the curve shown should be taken only as representative of the short- term wavelength-dependent variation in quantum efficiency. the detail curve was taken at 10 nm intervals while the smoothed curve represents an average over a bandwidth of approximately 100 nm. figure 4 shows the typical output charge as a function of exposure at 650 nm wavelength for the 25 and 50 m pitch models. exposure in nj/cm 2 is calculated by multiplying the irradiance in w/cm 2 by the integration time in msec. the response is very linear up to the saturation charge - 10 pc in 25 m models and 20 pc in the 50 m models. the dark current of an l-series sensor at 25c is typically 0.2 pa for the 25 m models or 0.4 pa for the 50 m, doubling for every 7c increase in temperature. see table 2 for detailed electro-optical characteristics. clock and voltage requirements scanning is achieved by means of an integrated shift register. the shift regis- ter is driven by complementary square wave clocks, ? 1 and ? 2 . table 1 gives rise and fall times and crossover points for these clock waveforms. the clock amplitude should be equal to v dd C v ss . with v dd = 5 v and v ss = 0 v, the clock inputs will be hcmos-compatible. since each photodiode is read out on a negative transition of ? 2 (see figure 6), the frequency of the clock signal should be set equal to the desired video data rate. the start pulse of similar amplitude to the clocks is required to load the shift register and initiate each readout period (each scan of the array). the start pulse is loaded when ? 2 is high; the start signal is pulsed high for a minimum of 10 ns during one and only one ? 2 clock high cycle. a timing diagram for the start and clock signals is shown in figure 6. caution: ? 1 and ? 2 clocks need to be continuously applied to the sensor characteristics (contd.) notes: 1. measured. table 1. electrical characteristics (25 c) (all voltages measured with respect to v sub ) signal sym min typ max units v dd v dd 4.5 5 5.5 v v dd guard v ddg -v dd -v v ss v ss -0-v v sub v sub -0-v antiblooming drain v abd v sh high v dd - 0.1 - v dd v v sl low v ss -v ss + 0.4 v clock ? 1, ? 2v h1 ,v h2 high v dd C 0.1 - v dd v v l1 ,v l2 low v ss -v ss + 0.4 v antiblooming gate v habg high v dd C 0.1 - v dd v v labg low v ss -v ss + 0.4 v v habg high v dd C 0.1 - v dd v v labg low -3.0 - -2.0 v video bias v v 2v dd /2 v dd -2 v reset drain v clock rate - 0.001 - 1 mhz start rise time t rs -1050ns start fall time t fs -1050ns start pulse width t pws 10 - - ns ? 1 rise time t r1 -1020ns ? 1 fall time t f1 -1020ns ? 2 rise time t r2 -1020ns ? 2 fall time t f2 -1020ns video delay time t vd -20-ns clock crossing x 1 0-50% x 2 0-50% capacitance ? 1, ? 2 at 5 v bias rl1201 c c -9-pf rl1202 c c -50-pf rl1205 c c -77-pf rl1210 c c pf rl1501 c c -30-pf rl1502 c c -48-pf rl1505 c c -93-pf capacitance, each video line at at 2.5 v bias 1 rl1201 c v --pf rl1202 c v --pf rl1205 c v --pf rl1210 c v - -pf rl1501 c v - - pf rl1502 c v - -pf rl1505 c v - pf rl12xx rl15xx v dd - 0.1 v dd /2 - 0.1 - - v dd v dd /2 v v 2 - - ss v rl12xx rl15xx reset gate 2 - - - ss v v rl12xx rl15xx v v v 4.5 dd v 5.5 v abd 154 - - 6.7 10.2 15.4 28.7 9.1 14 capacitance, each video line at 5.0 v bias 1 - 25 start v rd 2. clocking of reset gate may decrease dynamic range.
www.perkinelmer.com/opto cmos spectroscopy sensor clock & voltage requirements (contd.) dsp-106.01c - 10/2001w page 4 sensor. the integration period should be controlled by varying the time between start pulses. for optimum performance and minimum switching noise, it is important that the clocks are exact complements and that their rise and fall times comply with table 1. a recommended circuit for generating these clocks is shown in figure 7. end of scan an output pulse useful primarily for test purposes is provided two clock cycles after the last photodiode is sampled by the shift register scanning circuit. the timing of the eos output is shown in figure 6. the voltage levels on the eos output will be determined by the v dd and v ss voltage levels sup- plied to the photodiode array. when v dd is at +5v and v ss is operated at 0 v, the eos output will be compatible with the hcmos family of logic devices. amplifier requirements the recommended amplifier circuit for use with an l-series sensor is a simple current amplifier. a current amplifier holds the video line at a virtual ground and senses the current pulses flowing into the video line to recharge the diodes through their respective multiplex switches as they are sampled in sequence. these current pulses, which each can contain a charge up to saturation, are converted to a train of voltage pulses corresponding to the light intensity on the various diodes. in this mode of operation, the current amplifier must provide a positive bias voltage to the video line since the photodiode anode (the p-substrate) is biased to 0 v (v ). figure 8 shows a ss differential recharge amplifier suitable for use with l-series sensors. line reset / antiblooming control under certain operating conditions, it may be desirable to control integration time independent of the line scan time (time between start pulses). this can be accomplished by the use of the antiblooming gate control input. when the antiblooming gate is held high, all photodiodes are simultaneously table 3. absolute maximum ratings notes: 1. measured at specified video line bias. 2. average 600-700 nm, includes 8% window loss. 3. photo response non-uniformity (prnu) is defined as: a. +% = (v max - v avg ) / v avg x 100% b. C% = (v avg - v min ) / v avg x 100% where i. v max is the output of the pixel closest to saturation level. ii. v min is the output of the pixel closest to the dark level. iii. v avg is the average of all array pixel outputs. the first and last pixels are not counted in this measurement. 4. measured at an exposure level of e sat /2. 6. maximum dark current < 1.5 x average dark current. min max units voltage applied to any terminal with respect to v ss 0 +10 v storage or operating temperature -78 +85 c table 2. electro-optical characteristics (25c) characteristic typ max units center-to-center spacing rl12xx 25 - m rl15xx 50 - m sensitivity 1, 2 2x 10 -4 - c/j/cm 2 prnu 2, 3, 4 510% saturation exposure (e sat ) 1, 2 50 - nj/cm 2 saturation charge (q sat ) rl12xx - pc rl15xx - pc dynamic range (q sat / q noise(rms) ) rl12xx 70,000 - - rl15xx 100,000 - - average dark current 6 rl12xx 0.20 0.50 pa rl15xx 0.40 1.0 pa spectral response peak 650 - nm spectral response range 300-1000 250-1050 nm rl12xx rl15xx rl12xx rl15xx nj/cm 2 50 - - 4 x 10 -4 c/j/cm 2 10 20 reset to the bias voltage on the anti- blooming drain (typically v dd /2). conversely, when the antiblooming gate is held low, the antiblooming transistor is off and the photodiodes can then integrate photocurrent. thus, when an active high pulse is applied to vabg, the integration time for diode blooming gate to the time in which diode "n" is read out through the diode multiplex switch. under normal operating conditions, the l-series sensors do not require antiblooming control due to excellent antiblooming characteristics. however, under extremely high ex- posure conditions, antiblooming control "n" then becomes the time between the negative-going transition of the anti- can be implemented to further enhance this performance. in this mode of 5 5. minimum q for rl15xx is 15 pc; rl12xx is 8 pc. sat
cmos spectroscopy sensors www.perkinelmer.com/opto line reset / antiblooming control (contd.) dsp-106.01c - 10/2001w page 5 operation, a bias voltage (the same voltage as the video line bias, typically v dd/ 2) is applied to the antiblooming drain. the antiblooming drain is then biased to 1 - 3 v above . by adjusting the bias level on the antiblooming gate, excess charge present on the video line is shunted to the antiblooming drain. figure 6. timing diagram t pws 12 3 4 n-2 n-1 v sl v sh start x crossing ? 1 clock crossings v v h1 l1 200 ns min h2 l2 v v ? 1 ? 2 video eos n t vd 50 ns min 50 ns min t t t t r1 f1 f2 r2 100% 0% 1 ? 2 x crossing 2 figure 5. temperature diode read mode v -10v 100 kohm vsub temp diode output figure 4. typical signal charge vs. exposure at 600 nm 0 20 15 25 saturation charge signal charge (pc) exposure (nj/cm saturation exposure 60 2 ) 25 m pitch 50 m pitch 50 40 20 10 30 0 10 5 v labg
www.perkinelmer.com/opto cmos spectroscopy sensors a b c d ld ck clr enp ent rco q q q q a b c d rco dip switch dip switch dip switch d ck q q d ck q q clr pr clr pr 74hc74 74hc74 74hc161 74hc161 74hc161 13 8 12 11 10 1 6 5 2 3 15 15 15 +5 74hc04 +5 clock in start 4 rco 1k 1k 9 6 74hc02 phase 1 phase 2 9 4 1 5 2 74hc02 74hc04 74hc04 8 4 11 5 10 6 3 3 1 2 3 4 5 6 9 2 1 7 10 a b c d ld ck clr enp ent q q q q a b c d q q q q a b c d a b c d ld ck clr enp ent 3 4 5 6 9 2 1 7 10 3 4 5 6 9 2 1 7 10 figure 7. two-phase drive circuit dsp-106.01 c - 10 /2001w page 6 dark signal and noise there are two components of the dark signal from the l-series sensors. these are due to: (1) spatial variations in the switching transients coupled into the video line through the clocks and internal multiplex switches, and (2) the integrated dark current. a por- tion of the switching transient effect will be spatially random and a portion will have the periodicity of the clocks. the latter portion can be min- imized by matching the clock ampli- tudes and rise and fall times and by good circuit layout to minimize capacitance between clocks and video lines. the peak-to-peak fixed pattern due to all switching transient effects should be less than 1% of the saturated signal. the dark signal due to dark current is the dark current multiplied by the integration time. this can be arbitrarily reduced by lowering the temperature or by reduc- ing the integration time. there are three identifiable sources of readout noise: (1) reset noise, (2) shot noise, and (3) amplifier noise. reset noise is associated with resetting the diode capacitance to a fixed voltage. its root mean square value is given by (ktc)1/2/q where k is bolzmann?s constant, t is the absolute tempera- ture, q is the electronic charge and c is the total capacitance of the photo- diode, the associated video line and the capacitance of the external cir- cuitry. at room temperature, ktc noise is approximately 1500 electrons rms. it can be reduced somewhat by cooling. the rms value of the dark shot noise is the square root of the number of electrons in the dark sig- nal charge. for example, with a room temperature dark current of 0.25 pa and 10 ms integration time, the rms dark current shot noise is approxi- mately 125 electrons. because of the exponential temperature dependence of dark current, shot noise can be reduced dramatically with a moderate amount of cooling. amplifier noise depends on the amplifier circuit used. in general, the low video output capacitance of the l-series sensors makes it easy to achieve low amplifier noise, and values below 2000 electrons are possible. temperature diodes the l-series sensor arrays have two on-chip diodes for sensing array temp- erature. the standard method of use is to force a fixed forward current (normally 100 a) through the diodes and measure the forward diode voltage drop. the diode test circuitry is shown in figure 5. the forward diode voltage drop at 25c is 592.8 mv. the equation for computing junction temp- erature from the measured diode forward voltage (vf) is the familiar straight-line equation: t j = k * v f + t 0 where t j is the junction temperature, k = -0. 407 (c/mv) is the slope or derivative of the diode junction temp- erature versus forward voltage (c/mv), v f is the diode forward voltage and t 0 = 266.26 c is the ordinate-intercept or offset temperature.
www.perkinelmer.com/opto l-series photodiode array active video dummy video 1.5pf 2k 2k 1.5pf video out +5v + - figure 8. l-series readout amplifier cmos spectroscopy sensors dsp-106.01c - 10/2001w page 7 1 122 22 320 419 518 617 716 815 914 10 13 11 12 antiblooming gate start temp diode 1 v dd active video end-of-scan antiblooming drain reset gate reset drain temp diode 2 v sub dummy video nc nc nc nc dil, top view v ss v sub v ddg ? 1 ? 2 nc figure 9. pinout configurations 1.000 (2.54) 0.100 (2.54) 0.046 (1.168) fused silica window silicon chip 0.020 (0.508) pin 1 sensing area 0.2055 (5.220) 0.1845 (4.686) 2.5mm 0.390 (9.906) 0.410 (10.414) b (see table 4) a (see table 4) (all dimensions are typical and in inches (mm)) figure 10. package dimensions table 4. package dimensions and tolerances part number a b 1.080 (27.43) 0.126 (3.20) 1.080 (27.43) 0.252 (6.40) 1.080 (27.43) 0.504 (12.80) 1.600 (40.64) 1.008 (25.60) 1.080 (27.43) 0.252 (6.40) 1.080 (27.43) 0.504 (12.80) 1.600 (40.64) 1.008 (25.60) note: all dimensions in inches (mm) rl1201 rl1202 rl1205 rl1210 rl1501 rl1502 rl1505 1.080 (27.43) 200 o 200 o 1m o 1m o notes: 1. video out amplitude = signal charge / feedback capacitance video amp == = q c f 10pc 1.5pf 6.7v t=c f r f =(1.5pf)(1mohm)=1.5 s 2.5v offset *t = time constant 2. circuit shown is for a rl12xx device. 1
www.perkinelmer.com/opto for more information e-mail us at opto@perkinelmer.com or visit our web site at www.perkinelmer.com/opto . all values are nominal; specifications subject to change without notice. ? 2001 perkinelmer inc. all rights reserved. perkinelmer, the perkinelmer logo and the stylized ?p? are trademarks of perkinelm er, inc. dsp-106.01 c ? 10 /2001w reticon is a registered trademark of perkinelmer, inc. cmos spectroscopy sensor page 8 table 6. sales offices north america united states perkinelmer optoelectronics 2175 mission college blvd. santa clara, ca 95054 toll free: 800-775-opto (6786) phone: +1-408-565-0830 fax: +1-408-565-0703 europe germany perkinelmer optoelectronics gmbh wenzel-jaksch-str. 31 d-65199 wiesbaden, germany phone: +49-611-492-570 fax: +49-611-492-165 asia japan perkinelmer optoelectronics neopt. 18f, parale mitsui building 8 higashida-cho, kawasaki-ku kawasaki-shi, kanagawa-ken 210-0005 japan phone: +81-44-200-9170 fax: +81-44-200-9160 www.neopt.co.jp singapore 47 ayer rajah crescent #06-12 singapore 139947 phone: +65-770-4925 fax: +65-777-1008 pixels pixel pitch 25m 50m 128 rl1201lgq-711 rl1501lfq-711 256 rl1202lgq-711 rl1502lfq-711 512 rl1205lgq-711 rl1505lfq-711 1024 rl1210lgq-711 - table 5. stock part numbers ordering information while the information provided in this data sheet is intended to describe the form, fit and function for this product, perkinelmer reserves the right to make changes without notice.
|
