38 CHAPTER 3．∫N8TR UMEN工4TION
DEτ ：1肌m ・l DET念｛3・cm ・】 PE州a mln】
Fig．3．12：Focal plane images formed by stray light（Serlemitsos et al．2007）． The left and middle panels show simulated images of a monochromatic point−like source of 4．51 keV locatillg at（DETX、 DETY）＝（−20〜0 ）in the cases of without and with the pre−collimator，
respectively The radial dark lanes are the shades of the alignment bars． The right panels is the in−fiight stray ilnage of the Crab Nebula in the 2．5−5．5 keV band located at the same o任axis angle． The unit of the color scale of this panel is counts per 16 pixels over the entire exposure time of 8428．8s， The counting rate f拓m the whole image is O．78土0．01 cs−l includillg background． Note that the illtensity of the Crab Nebula measured with XIS3 at the XIS−default position is 458±3 c s−1 in the same 2．5−5．5 keV band． All the
images are binned with 4×4pixels followed by being smoothed with a Gaussian with a
sigma of 2 pixels， where the pixel size is 24μm．
1．．．．』：．．・．．弓 ．．1．1・．、・． 1．．、・．、「．．、ト．．一 ㌔・．．・・・・… ．1．「・「」．．・ぺ．1．． ．・．．．・ 、：．．k覗
3．1．THE 5σZ4κひSATELL∫TE 39
Fig．3．14：The four XIS detectors before installation onto 5脇α抗（Koyalna et aL 2007）．
XRT−12， and XRT−13． Each CCD camera has a single CCD chip with an array of 1024×
1024picture elelnents（ pixels「ハ）， and covers an 17．8 ×17．8 region on the sky． Each pixel is 24μm square， and the size of the CGD is 25 mm×25 mm、011e of the XISs， XIS］，uses aback−side illuminated（BI）CCDs， while the other three use仕ont−side illuminated（FI）
CCDs． The XIS has been partially developed at MIT（CGD sensors， analog electronics，
thermo−electric coolers， and temperature control electronics）， while the digital electroniぐs and a part of the sensor housing were developed in Japan， jointly by Kyoto University．
Osaka University， Rikkyo University， Ehime University， and ISAS．
ACCD has a gate structure on one surface to transfCr the charge packets to the readout gate． The surface of the chip with the gate structure is called the 仕ont side 「． A
仕ont−side illuminated GCD〔FI CCD）detects X−ray photons that pass through its gate
structures， i，e．加m the f士ont side． Because of the additional photo−electric absorption at the gate st「ucture， the low−energy quantum detection ef五ciency（QDE）of the FI CCD is rather limited． Conversely， a back−side illuminated CCD（BI CCD）receives photons hom＾back， or the side without the gate structures． For this purpose， the undepleted layer of the CCD is completely removed in the BI CCD， and a thin layer to enhance the electron collection efhdency is added in the back surface． A BI GCD retains a high QDE e▽en ill sub−keV energy band because of the absence of gate structure on the photon−detection side． However， a BI CCD tends to have a slightly thinner depletion layer， and the QDE is therefore slightly lower in the high energy band． The decision to use only one BI CCD and three FI CCDs was made because of both the slight additional risk involved in the
40 CHAPTER 3． IN8TRσMEN皿4質ON
new technology BI CCDs and the need七〇balance the overall e伍ciency for both low and high energy photons．
Tb minimize七he thermal noise， the sensors need to be kept at〜−90°C during ob−
servations． This is accomplished by thermo−electric coolers（TECs）， con七rolled by TEC Con七rol Electronics， or TCE． The Analog Electronics（AE）drives the CCD clocks， reads and amplifies the data from the CCDs， performs the analog−to−digital conversion， and routes the signals to the Digi七al Electronics（DE）． The AE and TCE are located in the same housing， and together， they are called the AE／TCE．5酩αえ％has two AE／TCEs；
AE／TCEOl is used for XIS−SO and S1， and AE／TCE23 is used for XIS−S2 and S3， The digital elec七ronics system for the XISs consists of two Pixel Processing Units（PPU）and
one Main Processing Unit（MPU）；PPUOI is associated with AE／TCEO1， and PPU23 is associated with AE／TCE23． The PPUs receive the raw data from AE， carry ou七event
detection， and send even七da七a to the MPU． The MPU edits and packets the event da七a，and sends them to the satellite，s main digital processor．
To reduce contamination of the X−ray signal by optical and UV light， each XIS has
。an Optical Blocking Filter（OBF）10cated in front of it． The OBF is made of polyimide wi七h a thickness of 1000 A， coated wi七h a七〇七al of 1200 A of aluminum（400 A on one side and 800 A on the o七her side）． Tb facili七a七e the in一且igh七calibration of the XISs， each CCD sensor has two 55】比calibration sources． One is installed on the door尤o illuminate the whole chip， while七he other is located on the side wall of the housillg and is collimated in order to illuminate two corners of the CCD． The door−moun七ed source will be used for initial calibration only；once the door is opened， i七will not illuminate the CCD． The collimated source can easily be seen in two corners of each CCD． A small number of七hese X−rays scatter onto the entire CCD． In addition七〇the emission lines created by these sources， we can utilize a new feature of the XIS CCDs， charge injection capabili七y， 七〇 assist with calibration． This allows an arbi七rary amoun七〇f charge to be input to the pixels at the top row of the imaging region（exposure area）， i．e． the far side from the
fra血e−store region． The charge injection capability may be used to measure the CTI
（charge transfer ine伍ciency）of each column， or even to reduce the CTI， How the charge injection capability will be used is still in progress as of this writing．Pulse Height Determination， Residual Dark−current Distribution， and Hot Pixels
When a CCD pixel absorbs an X−ray photon，七he X−ray is converted to an electric charge，
which in turn produces a voltage at the analog output of the CCD． This voltage（ pulse−
height ）is proportional to the energy of the incident X−ray In order to determine the true pulse−height corresponding to the input X−ray energy， it is necessary to subtract the Dark Levels and correct possible oμical Light Leaks．
Dark Levels are non−zero pixel pulse−heights caused by leakage currents in the CdD．
In addition， optical and UV light may enter the sensor due七〇imperfect shielding（ 1ight leak ）， producing pulse heights that are not related to X−rays． In the case of the ASCA
雛THE Sσ蹴σ8AT肌∫TE 41
SIS，七hese were handled via a single mechanism：Dark Levels of 16×16 pixels were sampled
and their（truncated）average was calculated for every exposure・Then the same average
Dark L，evel was used to de七ermine the pulse−height of each pixel in the sample・After 七he launch of ASCA， it was found that the Dark Levels of different pixels were actually，；≠塩 eren七， and their distribution around七he average did not necessarily follow a Gaussian・
．The non−Gaussian distribution evolved with time（referred to as Residual Dark−current Distributlon or RDD）， and resulted ln a degrada七10n of七he energy resolutlon due to ihcorrec七Dark Levels．
For七he S批αえμXIS， Dark Levels and Light Leaks are calcula七ed separately in norma1 mode． Dark Levels are defined for each pixe1；those are expected to be constant for a given observa七ion・The PPU calculates the Dark Levels in七he Dark Ini七ial mode（one of lhe special diagnostic modes of the XIS）；those are stored in the Dark Level RAM・The ふerage Dark Level is determined for each pixe1， and if the dark level is higher than七he hot−pixel七hreshold， this pixel is labeled as a hot pjxe1・Dark Levels can be updated by the Dark Update mode， and sent to the teleme七ry by the Dark Frame mode． Unlike the case of ASCA， Dark Levels are not determined for every exposure， but the same Dark Levels are used for many exposures unless they are initialized or updated． Analysis of
the ASCA data showed that Dark Levels tend to change mostly during the SAA passage
of the satellite． Dark Update mode may be employed several times a day after the SAA passage・Hot pixels are pixels which always output over threshold pulse−heights even without input signals． Hot pixels are not usable for observa七ion， and their output has七〇be dis−
regarded during scientific analysis． The ASCA SIS did not identify hot pixels on−board，
and all the hot pixel data were telemetered and removed during七he data analysis proce−
dure．・The number of hot pixels increased wi七h time， and eventually Occupied signi丘cant parts of the telemetry In七he case of XIS， ho七pixels are detected on−board by the Dark Initial／Update mode， and their positions and pulse−heights are stored in the Hot−pixel RAM and sent to the telemetry Thus， hot pixels，can be recognized on−board， and they are excluded from the event detec七ion processes． R is also possible to specify the hot pix−
els manuallyl There are， however， some pixels which output over threshold pulse−heights intermit七ent1γ Such pixels are called flickering pixels．1七is difncult to identify and re一 血ove the flickering Pixels on board；they are inevitably output to the telemetry and need to be removed during the ground processing． Flickering Pixels sometimes cluster around specific columns， which makes it relatively easy七〇iden七ify．
The Light Leaks are calculated on board with the pulse height data after the subtrac−
tion of the Dark Levels． A七runcated average is calculated for 64×64 pixels（this size may be changed in the future）in every exposure and its running average produces the Light Leak． Thus， the Light Leak is basically the same as the Dark Level in AscA sls．
The Dark Levels and the L，ight Leaks are merged in the parallel−sum（P−Sum）mode， so Dark Update mode is not available in P−Sum mode． The Dark Levels， which are defined fbr each pixel as the case of the normal mode， are updated every exposure． It may be
42 CHAPTER 3．1NSTRσMEN工4質ON
considered that the Ught Leak is de丘ned for each pixel in P−Sum mode．
On−board Event Analysis
The main purpose of the on−board processing of the CCD data is to reduce the total amount transmitted to ground． For七his purpose，七he PPU searches for a characteristic pattern of charge distribution（called an event）in the pre−processed（post−Dark Levels and
Light Leaks subtraction）frame data． When an X−ray photon is absorbed in a pixel， the photoionized electrons can spread into at most four adj acent pixels． An event is recognized when a valid pulse−height（one between the Event Lower and Upper Thresholds）is found that exceeds the pulse−heights in the eight adjacent pixels（e．9． it is the peak value in七he 3×3pixel grid）． The coordinates of七he central pixel are considered the location of the
event， Pulse−height data for the adjacent 5×5square pixels are sent to the Event RAM as well as the pixel location．
The MPU reads the Event RAM and edits the data to the telemetry format． The amount of information sent七〇telemetry depends on the editing mode of the XIS． All the editing modes（in normal mode）are designed七〇send七he pulse heights of at least
4central pixels of an event七〇the telemetry， because the charge cloud produced by an X−ray photon can spread into at most 4 pixels． Informa七ion of the surro皿ding pixelsmay or may not output to the七elemetry depending on the editing mode． The 5×5mode
outputs the most detailed information七〇the telemetryJ．e． al125 pulse−heights from七he 5×5pixels containing七he event． The size of the telemetry data per event is reduced by afactor of 2 in 3×3mode．Photon pile−up
The XIS is essentially a position−sensitive integrating instrument， with the nominal inter−
val between readouts of 8 s． If during the integration time one or more photons strike the same CCD pixel， or one of its immediate neighbors， these cannot be correctly detected as independen七photons：this is the phenomenon of photon pile−up． Here， the modes七angular resolu七ion of the SμzαえμXRT is an advantage：the central 3 x 3 pixel area receives 2％of the total counts of a poin七source， and〜10％of the counts fall within〜0．15 arcmin of the image center． We calculated the count rate at which 50％of the events within the centra1 3×3pixels are piled−up（the pile−up fraction goes down as we move out of七he image
center；this丘action is＜5％for the O．15 arcmin radius）−although we offer no formal
justification for this particular limit， this is compatible with our ASCA SIS experience（i．e．， at this leve1， the pile−up effects do not dominate the systematic uncertain七ies）．
XIS Background Rate
All four XISs have low backgrounds， due to a combination of the Sμzαえμorbit and the instrumental design． Below l keV， the high sensitivity and energy resolution of the XIS−
Sl combined with this low background means that S批αえμis the superior instrument for
3．1．THE SσZ4κσ8ATELL∫TE 43
Table 3．4：Major XIS Background Emission Lines
Line Energy XIS−SO XIS−Sl XIS−S2 XIS−S3
keV 10−9 c七／s／pix 10−9 ct／s／pix 10−9 ct／s／pix 10−9 ct／s／pix
OK O．5249 18．5±0．5 69．3±1：ζ 14．3±｝：1 14．1±｝：；
AIK 1．846 1．98±o．23 3．01±o．51 1．50±8：》； 1．57±8：1§
SiK 2．307 0．299±1：》892 2．21±0．45 0．0644（＜0．282） 0．543±8：1｝I AuM 2．1229 0．581±0．2341．13±8：；1！ 0．359±1：1｛； 6．69±；：；；
Mn Kα 5．898 8．35±1：12 0．648±o．289 0．299±8：ll16 0．394±1：｝§1 Mn Kβ 6．490 1．03±8：ll6 0．294（＜0．649） 0．00（＜0．111） 0．428±1：Ill NiKα 7．470 7．20土o．31 6．24±o．53 3．78±8：；1 7．13±8：；亨 NiKβ 8．265 0．583±0．183 1．15±8：量89 0．622±0．206 0．983±8：錨 AuLα 9．671 3．52±8：菟 3．28±；：；1 1．88±8：；； 3．54±8：§l AuLβ 11．514 2．25±8：ζ； 2．91土1．29 0．752±8：鋸2 2．67±8：91
Note：Typical accumulation time are 110−160 ks
observing soft sources wi七h low surface brightness． At the same time， the large effective
area at Fe K（comparable to七he XMM pn）combined with this low background make
飢zαえμapowerful tool for investigating hot and／or high energy sources as well．
In the XIS，七he background originates from the cosmic X−ray background（CXB）com−
bined with charged particles（the non−X−ray background， or NXB）． Currently， flickering
pixels are a negligible component of the background． When observing the dark earth
（乞．e．七he NXB）， the background rate between 1−12 keV in is O．11 cts／s in the FI CCDs and O．40 cts／s in the BI CCD；see Figure 3．15：1eft． Note七hat these are the fluxes af七er the grade selection is applied with only grade O，2，3，4and 6 selected． There are also fluorescence features arising ffom the calibra七ion source as well as ma七erial in the XIS and X㎜s． The Mn lines are due七〇the scattered X−rays丘om the calibra七ion sources．
As shown in Table 3．4 the Mn lines are almost negligible excep七for XIS−SO． The O lines are mostly contamination from the day earth（3．1．3）． The other lines are fluorescent lines from the material used for the sensor． Table 3．4 shows the current best estimates for the strength of these emission features， along with七heir 90％upper and lower limits．
The background rate on the FI chips（including all the grades）is normally less than 400counts／frame（50 cts／s）when no class discriminator is apPlied． On the BI chip， the rate is normally less than 150 counts／ffame（18．75 cts／s）． The backgro皿d rate on the FI chips is expected to reduce significantly when the class discriminator is apPlied． But little change is anticipated for the BI chip， Since 5×5，3×3， and 2×2modes require on average 40，20， and 10 bytes per event， the minimum telemetry required for any source is