Testing the Evolutionary Link Between Submillimetre Galaxies and Quasars CO Observations of

a r X i v :0806.0618v 2 [a s t r o -p h ] 8 J u l 2008

Mon.Not.R.Astron.Soc.000,000–000(0000)Printed 8July 2008

(MN L A T E X style ?le v2.2)

Testing the evolutionary link between submillimetre

galaxies and quasars:CO observations of QSOs at z ～2

K.E.K.Coppin 1, A.M.Swinbank 1,R.Neri 2,P.Cox 2, D.M.Alexander 3,Ian Smail 1,M.J.Page 4,J.A.Stevens 5,K.K.Knudsen 6,R.J.Ivison 7,8,A.Beelen 9,F.Bertoldi 6,A.Omont 10

1

Institute for Computational Cosmology,Durham University,South Road,Durham,DH13LE 2Institut de RadioAstronomie Millim′e trique (IRAM),300rue de la Piscine,Domaine Universitaire,38406Saint Martin d’H`e res,France 3Department of Physics,Durham University,South Road,Durham,DH13LE

4UCL,Mullard Space Science Laboratory,Holmbury St.Mary,Dorking,Surrey RH56NT

5Centre for Astrophysics Research,University of Hertfordshire,College Lane,Hat?eld,AL109AB 6Argelander-Institut f¨u r Astronomie,University of Bonn,Auf dem H¨u gel 71,D-53121Bonn,Germany

7Institute for Astronomy,University of Edinburgh,Royal Observatory,Blackford Hill,Edinburgh,EH93HJ 8UK Astronomy Technology Centre,Royal Observatory,Blackford Hill,Edinburgh,EH93HJ 9Institut d’Astrophysique Spatiale,Universit′e Paris-Sud,F-91405Orsay,France 10Institut d’Astrophysique de Paris,Universit′e Pierre &Marie Curie,98bis Boulevard Arago,F-75014Paris,France

8July 2008

ABSTRACT

We have used the IRAM Plateau de Bure millimeter interferometer and the UKIRT 1–

5μm Imager Spectrometer (UIST)to test the connection between the major phases of spheroid growth and nuclear accretion by mapping CO emission in nine submillimetre-detected QSOs at z =1.7–2.6with black hole (BH)masses derived from near-infrared spectroscopy.When combined with one QSO obtained from the literature,we present sensitive CO(3–2)or CO(2–1)observations of 10submillimetre-detected QSOs se-lected at the epoch of peak activity in both QSOs and submillimetre (submm)galaxies (SMGs).CO is detected in 5/6very optically luminous (M B ～?28)submm-detected QSOs with BH masses M BH ?109–1010M ⊙,con?rming the presence of large gas reser-voirs of M gas ?3.4×1010M ⊙.Our BH masses and dynamical mass constraints on the host spheroids suggest,at face value,that these optically luminous QSOs at z =2lie about an order of magnitude above the local BH-spheroid relation,M BH /M sph ,al-though this result is dependent on the size and inclination of the CO-emitting region.However,we ?nd that their BH masses are ～30times too large and their surface density is ～300times too small to be related to typical SMGs in an evolutionary sequence.Conversely,we measure weaker CO emission in four fainter (M B ～?25)submm-detected QSOs with properties,BH masses (M BH ?5×108M ⊙),and surface densities similar to SMGs.These QSOs appear to lie near the local M BH /M sph rela-tion,making them plausible ‘transition objects’in the proposed evolutionary sequence linking QSOs to the formation of massive young galaxies and BHs at high-redshift.We show that SMGs have a higher incidence of bimodal CO line pro?les than seen in our QSO sample,which we interpret as an e?ect of their relative inclinations,with the QSOs seen more face-on.Finally,we ?nd that the gas masses of the four fainter submm-detected QSOs imply that their star formation episodes could be sustained for ～10Myr,and are consistent with representing a phase in the formation of mas-sive galaxies which overlaps a preceding SMG starburst phase,before subsequently evolving into a population of present-day massive ellipticals.

Key words:galaxies:high-redshift –galaxies:evolution –galaxies:formation –galaxies:kinematics and dynamics –quasars:emission lines –submillimetre

2Coppin et al.

1INTRODUCTION

It has been established that every massive,local spheroid harbours a supermassive black hole(SMBH)in its cen-tre whose mass is proportional to that of its host(e.g. Magorrian et al.1998;Gebhardt et al.2000).This suggests that the black holes(BHs)and their surrounding galaxies were formed synchronously.This hypothesis has found sup-port from hydrodynamical simulations of galaxy formation, which use feedback from winds and out?ows from active galactic nuclei(AGN)to link the growth of the SMBH to that of its host(e.g.Di Matteo,Springel&Hernquist2005; Hopkins et al.2005;Bower et al.2006).Thus these models support a picture,?rst presented by Sanders et al.(1988), where a starburst-dominated ultra-luminous infrared galaxy (ULIRG),arising from a merger,evolves?rst into an ob-scured QSO and then into an unobscured QSO,before?-nally becoming a passive spheroid.

The high-redshift population of ULIRGs in this proposed evolutionary cycle are the submillimetre (submm)galaxies(SMGs;Smail,Ivison&Blain1997; Chapman et al.2005;Coppin et al.2006).These systems have ULIRG-like bolometric luminosities,L IR≥1012L⊙(Kov′a cs et al.2006;Coppin et al.2008),and they have many of the properties expected for gas-rich mergers (Swinbank et al.2004,2006;Tacconi et al.2006).This population evolves rapidly out to a peak at z～2.3,crudely matching the evolution of QSOs(Chapman et al.2005) and providing additional circumstantial evidence for a link between SMBH growth and spheroids.Two further results have shed light on the evolutionary link between SMGs and QSOs.Firstly,a modest fraction of optically luminous QSOs at z～2are detected in the submm/mm (～(25±10)%;Omont et al.2003)showing that the QSO-and SMG-phases do not overlap signi?cantly,given the lifetime estimates of the two populations(QSOs make up ～4%of?ux-limited samples of SMGs;Chapman et al. 2005).But when a QSO is detected in the submm/mm then it could to be in the transition phase from an SMG to an unobscured QSO,making its properties a powerful probe of the evolutionary cycle(e.g.Stevens et al.2005;Page et al. 2004).Secondly,the evolutionary state of the SMBHs within SMGs can also be judged using the2-Ms Chandra Deep Field North observations(Alexander et al.2003)to derive accurate AGN luminosities and hence lower limits on the BH masses(M BH)in those SMGs with precise redshifts in this region(Alexander et al.2005a,b;Borys et al.2005). These studies suggest that the AGN in typical SMGs are growing almost continuously–but that the SMBHs in these galaxies appear to be several times less massive than seen in comparably massive galaxies at z～0(Alexander et al. 2008).

Together these results argue for a fast transition from a star-formation-dominated SMG-phase to the AGN-dominated QSO-phase(Page et al.2004).The latter phase will result in the rapid BH growth necessary to account for the present-day relation between spheroid and SMBH masses(e.g.Magorrian et al.1998;Gebhardt et al.2000). Can we con?rm this and more generally test the proposed evolutionary link between SMGs and QSOs at the peak of their activity at z～2?verse by comparing the properties of QSOs and ULIRGs (e.g.Tacconi et al.2002).However,both of these popula-tions are1000×less abundant in the local Universe than they were at the era of their peak activity at z～2and so we have to be cautious about extrapolating from local examples to the high-redshift progenitors of the bulk of to-day’s massive spheroids(Genzel et al.2003;Swinbank et al. 2006).Thus,to properly test the validity of this cycle for typical spheroids we have to compare QSOs and ULIRGs at the era where their populations peaked:z～2.

The critical pieces of information needed to test the link between SMGs and QSOs are the relative dynamical, gas and SMBH masses of these two populations.In prin-ciple the dynamical masses can be derived from optical or near-infrared observations of emission line gas in the SMGs or QSOs(see Swinbank et al.2004,2005,2006).However, there is the problem of removing the QSO emission in these observations,as well as the e?ects of dust obscuration and out?ows.In contrast,molecular CO emission line pro?les are relatively immune to the e?ects of obscuration and out-?ows,while at the same time yielding additional constraints on the relationship between QSOs and SMGs from their gas masses.

There is currently a lack of sensitive CO observations of QSOs at z～1–3,with data published on only eight sources (e.g.Frayer et al.1998;Guilloteau et al.1999;Beelen et al. 2004;Hainline et al.2004).Instead the focus has been on CO studies of QSOs at z>～4(e.g.Omont et al.1996; Walter et al.2004;Riechers et al.2006),although these QSOs have little overlap with the redshift range where SMGs are typically detected.The paucity of CO constraints for z>1QSOs also re?ects the di?culty in determining their systemic redshifts with su?cient precision to guarantee that the CO emission falls within the bandwidth of typical mil-limetre(mm)correlators.However,sensitive near-infrared spectroscopy of the C iv,Mg ii,and[O iii]5007emission lines in QSOs can provide redshifts with required precision as well as Hαor Hβ?uxes and line widths to yield M BH estimates (Takata et al.2006;Alexander et al.2008).

We have carried out a quantitative test of the proposed link between SMGs and QSOs at z～2where both popula-tions are most common.We have obtained precise systemic redshifts from near-infrared spectroscopy of potential transi-tion QSOs(i.e.submm/mm-detected QSOs)and then used the IRAM Plateau de Bure Interferometer(PdBI)to search for CO emission.We relate their dynamical,gas and SMBH masses to SMGs from the PdBI CO survey(Greve et al. 2005).We test:a)whether the cold gas masses in these QSOs are similar to those in SMGs;b)whether the line widths and dynamical masses of these two populations are comparable; and c)how the ratio of SMBH to dynamical masses for these submm-detected QSOs relate to the estimates for SMGs and those for optically luminous QSOs(which lie on the present-day M BH–Mσrelation;McLure&Dunlop2004).Together these observations can constrain the proposed evolution-ary sequence which links QSOs to the formation of massive young galaxies and SMBHs at high redshift.

We describe the sample selection,observations and data reduction in§2.The results of the near-infrared and mm CO spectra are given in§3.The CO properties of the submm-detected QSOs are compared and contrasted with

Gas and BHs of QSOs and SMGs at z ～2

3

16

18

2022

24

R mag

R e d s h i f t

Figure 1.R -band apparent magnitude versus redshift for our sample of submm-detected QSOs.We show that these lie within the interquartile range (shaded region)of the redshift distribution of SMGs (median ?2.3;dashed line)from Chapman et al.(2005).This shows that we are probing a wide range in optical luminosities (and potentially a wide range in M BH )at the epoch where the QSO and SMG populations peak.

submm-detected QSOs in §5.Our conclusions are given in §6.We adopt cosmological parameters from the WMAP ?ts in Spergel et al.(2003):?Λ=0.73,?m =0.27,and H 0=71km s ?1Mpc ?1.All quoted magnitudes are on the Vega system.

2OBSER V ATIONS,REDUCTION AND REDSHIFTS 2.1

Sample Selection

The objective of this study is to test the link between SMGs and QSOs at the era where these populations peaked.We have therefore selected radio-quiet QSOs with a redshift distribution matched to SMGs ( z ～2.3,Chapman et al.2005)and with 850μm ?uxes >～5mJy.The S 850μm >

～5mJy ?ux limit selects comparatively bright QSOs and SMGs and is imposed so that the sources would be su?ciently bright to be detected in CO.We choose submm-detected QSOs from three main samples:we select ?ve QSOs from a submm photometry survey of optically-bright QSOs by Omont et al.(2003)–HS1002+4400,J140955.5+562827,J154359.3+535903,HS1611+4719,and J164914.9+530316,we select two QSOs from the submm photometry sur-vey of X-ray absorbed QSOs by Stevens et al.(2005)–RXJ121803.82+470854.6and RXJ124913.86–055906.2,and ?nally we include two QSOs found in blank-?eld submm surveys by Chapman et al.(2005)and Greve et al.(2004),SMM J131222.35+423814.1and MM J163655+4059,selected QSO targetted in submm photometry mode by Chapman et al.(2005).

The ?nal sample comprises 10submm-detected QSOs with similar ?uxes to the SMG sample (Greve et al.2005;>～5mJy at 850μm,assuming 850μm/1.2mm colours typi-cal of SMGs;Greve et al.2004),with a median of 10mJy.The SEDs of these QSOs demonstrate that the bulk of the mm/submm luminosity arises from dust emission and is very likely associated with star formation,rather than an AGN (Beelen et al.2006;Omont et al.2003).Our sample spans six orders of magnitude in QSO R -band brightness,enabling us to potentially trace how CO properties depend on QSO luminosity or M BH (see Fig.1).2.2

Near-Infrared Spectroscopy

The QSOs in this sample were spectroscopically identi-?ed through rest-frame UV spectroscopy.However,since the rest-frame UV emission lines can be extremely broad (FWHM ～5000km s ?1),and can be o?set in velocity with respect to the systemic redshift,precise systemic redshifts are required to ensure the CO emission falls within the PdBI receiver bandwidth.To obtain precise systemic redshifts for our targets we use the nebular emission lines of [O iii ]λ5007and H βwhich are redshifted to near-infrared wavelengths.

Seven of our submm-detected QSOs were observed with the United Kingdom InfraRed Telescope (UKIRT)1–5μm Imager Spectrometer (UIST;Howatt et al.2004).We used the HK grism which provides a spectral resolution of λ/?λ～1000and covers a wavelength range 1.4–2.4μm,thus allowing us to cover H β,[O iii ]λ4959,5007and H αsi-multaneously for most of our QSOs.The QSOs were ob-served between 2006Feb and 2006Oct.All observations

were taken in <～0.8′′

seeing and clear conditions,and were carried out using the standard ABBA con?guration where the QSO was nodded by 6–12′′along the slit to achieve sky subtraction.Individual exposures were 240s and the total integration times varied between 1.2ks and 4.8ks depend-ing on the K -band magnitude of the QSO (see Table 1).In addition,three QSOs in our sample have suitable archival spectroscopy from Takata et al.(2006)and Swinbank et al.(2004,2006).

The relevant orac-dr pipeline (Cavanagh et al.2003)was used to sky-subtract,extract,wavelength calibrate,?at-?eld and ?ux calibrate the data (and in the case of integral ?eld unit observations,forms the datacube).Spectra for the QSOs in our sample are shown in Fig.2.To derive systemic redshifts,we simultaneously ?t the H βand [O iii ]λ4959,5007emission lines with Gaussian pro?les.The emission line ?ux ratio of the [O iii ]λ4959,5007emission lines is ?xed such that I 4959/I 5007=2.9,and the [O iii ]λ4959,5007line widths are as-sumed to be the same.The ?t assumes a linear continuum plus Gaussian emission line pro?les allowing the velocity centroid,?ux and width of the H βline pro?le to be dif-ferent from [O iii ].This makes a total of seven parameters to be ?tted.We also derive emission line ?uxes and widths for the H αemission line (which are important for deriving M BH using the method of Greene &Ho 2005)in a similar manner,?tting a single Gaussian emission line pro?le superimposed on a linear continuum.The resulting redshifts and ?uxes are given in Table 2.These redshifts were subsequently used to

4Coppin et al.

Table1.Summary of the new near-infrared spectroscopy and mm-wave PdBI CO observations of submm-detected2

including nine new CO observations and one published object from Beelen et al.(2004).The CO t exp corresponds to the on-source

integration time with the equivalent of a6-element array.

Date(s)t exp Dates t exp Noise per channel f Beam FWHM Detected?

(ks)(hrs)(mJy beam?1)

Notes:

(a)All observed in the CO(3–2)transition,except for RXJ121803.82+470854.6observed in CO(2–1)

(b)From UIST spectroscopy(this work)

(c)From Swinbank et al.(2004)

(d)From Takata et al.(2006)and Alexander et al.(2008)

(e)From Beelen et al.(2004)

(f)With a channel width of20MHz

Due to the very weak[O iii]λ5007emission in RXJ124913.86,it was very di?cult to measure a reliable redshift for the CO observations,and thus a range of possi-ble redshifts were covered on this QSO(z～2.24–2.26).

During the course of assembling the data for this project,we obtained a CO detection of HS1611after se-curing a systemic redshift from identi?cation of[O iii]with UIST.The positional centroid of the CO was o?set by ～1.3arcsec north of the optical and near-infrared QSO po-sition,possibly indicating a companion,and HS1611was re-observed with the UIST integral?eld unit(IFU)for a total of3.6ks on2006Sep01.The UIST IFU uses an im-age slicer to take a3.3′′×6.0′′?eld and divides it into14 slices of width0.24′′.The dispersed spectra from the slices are reformatted on the detector to provide two-dimensional spectro-imaging,in our case also using the HK grism.We reduced the data using the relevant orac-dr pipeline which extracts,?at?elds,wavelength calibrates the data and forms the datacube.We discuss these data in§3.1.8.

2.3Millimetre Interferometry

We surveyed seven QSOs for CO emission using IRAM PdBI between2006June to2007June and two QSOs between 2001December and2002April in conditions with good at-mospheric phase stability and reasonable transparency.The observations were carried out using3,4,5,and6anten-nae in the D con?guration,giving a total of3,6,10,and 15baselines,respectively.For uniformity with the previous observations observations of SMGs,which will comprise our comparison sample,we have aimed to achieve the same1-σsensitivity as for the SMG survey(Greve et al.2005)for the new CO observations of our sample of submm-detected QSOs:S CO?v～0.3Jy km s?1.We also include results on a single QSO from the literature,J140955.5from Beelen et al. (2004),to yield a?nal sample of10QSOs.The PdBI obser-vations are summarised in Table1.

For data taken prior to2007January,the spectral corre-tion of2.5MHz at a bandwidth of580MHz.For data taken after January2007,the spectral correlator was adjusted to detect the line with a frequency resolution of2.5MHz in the 1GHz band of the new generation receivers.The visibilities were resampled to a frequency resolution of5MHz.

The data were calibrated,mapped and analyzed in the IRAM gildas1software package(Guilloteau&Lucas 2000).Anomalous and high phase-noise visibilities were ?agged.Passband calibrations were typically made using a nearby bright quasar.Phase and amplitude calibration within each observation were calibrated using frequent ob-servations of nearby quasars approximately every20min. The?ux calibration was done using standard calibrators,in-cluding3C273,MWC349,1637+574,1749+096,0923+392, 1243-072,and CRL618,for example.We estimate a conser-vative error on the?ux calibration of10–15%.

For each QSO the central observing frequency of the 3.1mm band receiver was tuned to either the redshifted CO(3–2)or CO(2–1)rotational transition,depending on the systemic near-infrared redshift derived from our spec-troscopy given in Table2.In the case of RXJ124913.86, the uncertainty on the redshift was large enough that we performed two separate observations with di?erent central frequency tunings of106.160and106.720GHz to ensure that the CO(3–2)line was covered.We have combined these data to cover more e?ective bandwidth and have centered the resulting spectrum at the mean central observed fre-quency,106.435GHz,corresponding to CO(3–2)redshifted to z?2.2487.

Naturally weighted datacubes were made and inspected for line emission close to the QSO position.Fig.3shows the velocity-integrated mm emission maps over the line in each ?eld.In Fig.4,we show the spectra of the CO emission in the brightest pixel of the central source in each velocity-integrated mm emission map,centered on the systemic red-shift of each system.We have included the maps and spec-

Gas and BHs of QSOs and SMGs at z ～2

5

Figure 2.Restframe UV–near-infrared spectra for our sample of submm-detected QSOs.The insets show the region around the H βand [O iii ]λ5007emission lines (all insets cover the wavelength range 4675–5100?A ),except in the case of J163655,and the redshifts derived from the [O iii ]λ5007emission are indicated.The redshift for J163655was taken from the redshifted H αdue to ambiguity in the [O iii ]λ5007emission lying near the edge of the wavelength coverage of the spectrograph;the inset therefore shows the region around H α.The region around 5100–5500?A in J1409has been masked due to strong atmospheric absorption.

tra of non-detections and in the latter cases we show the maps averaged over the central 500km s ?1of the spectra (the average line width of our sample)and spectra extracted at the phase tracking centre (or at the continuum peak for J164914.9).In all cases,we ignore the small correction for the primary beam attenuation,given that the detected sources are all close to the phase centres of the maps.

3ANALYSIS AND RESULTS 3.1

CO properties

We present new CO data for nine QSOs,including ?ve new CO detections.Four of our QSOs have CO detections with S/N >～5,and an additional QSO is marginally detected in CO with S/N ?2.5with emission coincident with the phase centre (see Fig.4).None of the QSOs appear to be resolved a previous CO detection reported by Beelen et al.(2004)in our sample of submm-detected QSOs.In total therefore this gives six detections in our sample of 10QSOs,yielding a detection rate comparable to that for SMGs by Greve et al.(2005)of ～60%.

Looking at the non-detections,it is unlikely that the cause of the failure to detect CO emission in these four QSOs is due to the systemic redshifts being wrong,as the quality of the optical/near-infrared spectral data for these is high and similar to the successful detections (see Fig.2).Similarly,it is unlikely that the velocity o?set of the CO emission and the systemic redshift are larger than the correlator coverage (the rms deviation of z NIR and z CO for our CO-detected QSOs is 0.0046),since only a few systems detected at high-redshift so far show larger o?sets (e.g.Hainline et al.2004).It is also unlikely that the CO line widths are ?1000km s ?1since our typical detected lines are 550km s ?1which is

6Coppin et al.

CO line widths could be one reason for some of the non-detections,the most likely reason appears to be that the QSOs are simply too faint in CO to be detected given the depth of our observations.However,the non-detections do still provide useful upper limits on the CO luminosity and gas masses and therefore we include them in our analysis.

Other than one QSO(J164914.9;§3.1.10),we do not detect the continuum at3.1mm in any of the QSOs.This is consistent with the measured submm/mm?uxes and grey-body dust emission with a spectral form Sν∝ν2+βin the Rayleigh-Jeans regime of the spectral energy distribution with a dust emissivityβ=1.5.

The luminosity,velocity width and spatial extent of the CO line emission can be used to place limits on the gas and dynamical mass of each system.For each CO-detected QSO a Gaussian pro?le is?t to either20-or30-MHz binned data by minimising theχ2statistic.The1σerrors on the best-?tting parameters are quoted where appropriate.To calcu-late the velocity-integrated?ux density of the CO emission, we integrate the best-?tting Gaussian pro?le(note that we do not?t the line and continuum simultaneously).The error on the velocity-integrated?ux density is calculated by boot-strapping from a Normal distribution centered on the best ?tting amplitude above,with a width given by the1σerror in the?tted amplitude,while holding the FWHM and cen-tral velocity of the best-?tting Gaussian pro?le?xed.The noise in the averaged channel maps,constructed from the av-erage emission over the full width at zero intensity(FWZI) in each QSO,is determined by summing over the weights and takes into account the phase noise visibilities.

For the non-detections we calculate3σupper lim-its S CO=3σ(δv?v FWHM)1/2,whereσis the channel-to-channel rms noise,δv is the velocity resolution and ?v FWHM is the line width(see e.g.Greve et al.2005; Seaquist,Ivison&Hall1995).We use the channel-to-channel rms noise of the spectra binned to a resolution of 100MHz and adopt a line width of500km s?1.Since our de-tections have typical line widths 550km s?1,this is a con-servative assumption.In addition,this line width assump-tion facilitates the direct comparison of our non-detections with those of SMGs in Greve et al.(2005).

We have summarised the line properties of each source (the CO position,z CO,L′CO(3?2)and FWHM)in Tables3 and4.O?sets of the CO emission from the radio position are given in Table3,assuming that the source is unresolved us-ing equation B2in Ivison et al.(2007),and are less than2σsigni?cant in most cases.We note that we discover serendip-itous>～4σsources in two of the?elds surveyed which may be associated with gas-rich companions to the QSOs (SMM J123716.01and SMM J131222.35).Overdensities of submm sources have already been found by Stevens et al. (2004)around similar QSOs on slightly bigger scales than those probed here.We now discuss the CO properties of each QSO detection and non-detection in detail,as well as any serendipitous detections.

3.1.1HS1002+4400

The integrated CO emission in this QSO is detected at ?6σ,0.8′′north of the Sloan Digital Sky Survey(SDSS; Fan et al.1999)position(see Fig.3).The CO(3–2)spec-olution of20MHz(54km s?1)and is shown in Fig.4.The CO emission line is well?tted by a Gaussian pro?le with a FWHM of640±160km s?1,a velocity-integrated?ux den-sity of S CO(3?2)=1.7±0.3Jy km s?1,and a CO redshift of z CO=2.1015±0.0007,o?set by20±65km s?1from the near-infrared systemic redshift.No signi?cant contin-uum emission is detected from the line-free region down to a1-σlimit of0.4mJy.

3.1.2RXJ121803.82+47085

4.6

Neither CO(2–1)emission at the systemic redshift nor con-tinuum emission are seen in the map near the QSO posi-tion.To calculate a3σupper limit to the velocity-integrated CO(2–1)line?ux of RXJ121803.82,we extract a spectrum at the QSO position(Table2)and rebin the spectrum to a frequency resolution of100MHz(360km s?1),resulting in

a channel rms of0.5mJy,and so derive an upper limit of

0.6Jy km s?1,assuming a line width of500km s?1,centred on the z NIR=1.7416.No signi?cant continuum emission is detected down to a1-σlimit of0.2mJy.

We do?nd two?4σserendipitous detections in the datacube,o?set in velocity from the central frequency tun-ing,and approximately10-20arcsec away from the phase centre.Neither show strong broad line detections in the spec-tra extracted at the peak locations,so we believe these are unlikely to be true sources.

3.1.3SMM J123716.01+620323.3

Neither CO(3–2)emission nor continuum emission are seen in the datacube near the QSO position.To calculate a3σupper limit to the velocity-integrated CO(3–2)line?ux, we extract a spectrum at the radio position(Table2)and rebin the spectrum to a frequency resolution of100MHz (270km s?1),resulting in a channel rms of0.2mJy,giving an upper limit of0.3Jy km s?1,assuming a line width of 500km s?1at the systemic redshift from[O iii]λ5007(Ta-ble2).No signi?cant continuum emission is detected down to a1-σlimit of0.2mJy.

We?nd a?4σdetection in the datacube,o?set in velocity from the central frequency tuning and approxi-mately18arcsec2south of the nominal pointing position at123715.36,620306.5(J2000),indicating a redshift of z CO=2.0597±0.0001,which is close to the QSO z NIR, suggesting it may be a gas-rich companion of the QSO with a separation of170kpc.See Fig. 5.This position corresponds to within0.5arcsec of an I=24.7counter-part(Smail et al.2004)with a radio?ux of≤15μJy (Biggs&Ivison2006)which is also undetected in the Chan-dra observations(Alexander et al.2003).The CO line has an integrated?ux of S CO(3?2)=0.2±0.06Jy km s?1and is very narrow(FWHM=100±30km s?1).

2Note that the map noise at this distance from the phase centre is only slightly degraded since the primary half power beam width

Gas and BHs of QSOs and SMGs at z ～2

7

Table 2.Summary of the near-infrared observed properties of the submm-detected QSOs.

HS1002+4400

100517.43434609.316.114.22.1015±0.0010

10000±1000

750±100

1500±300

RXJ121803.82+470854.6121804.54470851.019.517.0 1.7416

–

––

SMM J123716.01+620323.3123716.00620323.420.215.62.0568±0.00132100±500b 48±10b 1100±200RXJ124913.86–055906.2124913.85?055919.416.213.52.2400±0.01004820±2001000±100700±300SMM J131222.35+423814.1131222.32423813.920.218.02.5543±0.00102600±1000b 60±10b 670±300J140955.5+562827140955.50562827.017.014.92.5758±0.00504250±200220±20600±200J154359.3+535903154359.44535903.216.814.32.3692±0.00158280±300770±501000±200HS1611+4719

161239.90471157.017.115.12.4030±0.00084010±300300±201000±300MM J163655+4059163655.79405910.523.219.12.6070±0.00063000±40016±21840±400J164914.9+530316164914.90530316.016.914.22.2704±0.00093100±300580±501000±200

Source

CO position (J2000)O?set

Resolution g

z h CO

S CO ?v h FWHM h CO 850μm ?ux

1.2mm ?ux

RA Dec.(arcsec)

(kpc)

(Jy km s ?1)(km s ?1)

(mJy)(mJy)

Notes:

(a)1.2mm ?ux

from Omont et al.(2003)(b)850μm ?ux from Page et al.(2001)

(c)850μm ?ux from Chapman et al.(2005)(d)1.2mm ?ux from Greve et al.(2004)

(e)CO parameters from Beelen et al.(2004)(f)Detected in continuum only

(g)Approximate CO Resolution at the target redshifts,given the beamsizes in Table 1

(h)This quantity was obtained by ?tting a Gaussian distribution to the CO spectrum (see text)

3.1.4RXJ124913.86–055906.2

The integrated CO emission is detected at ?5σ,1.1arcsec southeast of the QSO position (Table 2).The CO(3–2)spec-trum of RXJ124913.86has been binned to a frequency res-olution of 30MHz (84km s ?1)and is shown in Fig.4.We have combined two sets of data with di?erent central fre-quency tunings (see §2.3)and have arbitrarily set the cen-tral frequency to 106.435GHz,corresponding to CO(3–2)redshifted to z =2.2487.The CO emission line is ?tted by a Gaussian pro?le with a FWHM of 1090±340km s ?1,a velocity-integrated ?ux density of S CO(3?2)=1.3±0.4Jy km s ?1,and a CO redshift of z CO =2.2470±0.0016,o?set by ?170±140km s ?1from the near-infrared redshift.This is our broadest CO line detection.No signi?cant con-tinuum emission is detected from the line-free region down to a 1-σlimit of 0.8mJy.

3.1.5SMM J131222.35+42381

4.1

The integrated CO emission of this QSO is marginally de-tected at ?2.5σ,coincident with the phase centre of the PdBI observations (Table 3).The CO(3–2)spectrum of tion of 20MHz (62km s ?1)and is shown in Fig.4.The CO emission line is ?tted by a Gaussian with a FWHM of 550±220km s ?1,a velocity-integrated ?ux density of S CO(3?2)=0.4±0.1Jy km s ?1(S/N ?3),and a CO redshift of z CO =2.5564±0.0011which is consistent with the near-infrared redshift within the 1σerrors.No signi?cant contin-uum emission is detected from the line-free region down to a 1-σlimit of 0.1mJy.

We ?nd a ?4σserendipitous emission line in the datacube at 131223.63,423819.33(J2000),approximately 15arcsec northeast 3of the phase centre,corresponding to a low S/N detected line with a FWHM=450±150km s ?1,z CO =2.5408±0.0007,and S CO(3?2)=0.40±0.17Jy km s ?1(S/N ?2.5).See Fig.5.We do not ?nd an optical counter-part for this source down to K ～20and z ～24.If real,this source is o?set by 120kpc and 1300km s ?1from the QSO.

3

Note that the map noise at this distance from the phase centre is only slightly degraded since the primary half power beam width

8Coppin et al.

3.1.6J140955.5+562827

We brie?y summarise the details for J140955.5here since its detection is published in(Hainline et al.2004;Beelen et al. 2004).Beelen et al.(2004)report a CO(3–2)emission de-tection at the central observed frequency(corresponding to z CO=2.5832±0.0001)that is well?tted by a Gaus-sian with a FWHM=311±28km s?1,yielding S CO(3?2)=

2.3±0.2Jy km s?1.

3.1.7J154359.3+535903

We detect the integrated CO emission of this QSO at?5σ, and coincides with the phase centre of the PdBI observations (Table3,Fig.3).The CO(3–2)spectrum of J154349.3has been binned to a frequency resolution of20MHz(58km s?1) and is shown in Fig.4.The CO emission line is well?tted by a Gaussian with a FWHM of500±130km s?1,a velocity-integrated?ux density of S CO(3?2)=1.0±0.2Jy km s?1,and a CO redshift of z CO=2.3698±0.0006consistent with the near-infrared redshift of the QSO within the1σerrors.No signi?cant continuum emission is detected from the line-free region down to a1-σlimit of0.4mJy.

3.1.8HS1611+4719

The integrated CO emission in this QSO is detected at?6σ, 1.3arcsec north of the phase center of the PdBI observations (see Fig.3).The CO(3–2)spectrum of HS1611+4719has been binned to a frequency resolution of20MHz(59km s?1) and is shown in Fig.4.The CO emission line is well?tted by a Gaussian with a FWHM of230±40km s?1,a velocity-integrated?ux density of S CO(3?2)=1.7±0.3Jy km s?1 (S/N?6),and a CO redshift of z CO=2.3961±0.0002.No signi?cant continuum emission is detected from the line-free region down to a1-σlimit of0.5mJy.

The apparent1.3arcsec spatial and～?600km s?1ve-locity o?sets between the CO and near-infrared emission might be naturally explained if the CO is associated with a companion galaxy.To test this we search the UIST IFU dat-acube for a companion galaxy to the north,but are not able to identify any strong line emitting galaxies to a?ux limit of ～2×10?17W m?2(corresponding to an Hαstar-formation rate of～40M⊙yr?1;Kennicutt1998).We note that the near-infrared spectrum shows an extended blue wing on the [O iii]emission line which is coincident with the redshift of the CO emission(see§2.2).Higher resolution mm spec-troscopy and/or deeper near-infrared spectral imaging may be required to determine whether the spatial and velocity o?sets arise due to a companion galaxy,an out?ow from the QSO,or a merger.This is the only example of a potentially signi?cant o?set between the CO and near-infrared position and/or redshifts.

3.1.9MM J163655+4059

Neither CO(3–2)emission nor continuum emission are seen in the map near the QSO position.To calculate a3-σupper limit to the velocity-integrated CO(3–2)line?ux,we extract a spectrum at the radio position(Table2)and rebin the spectrum to a frequency resolution of100MHz(310km s?1)a channel rms of0.2mJy,and we?nd an upper limit of

0.2Jy km s?1,assuming a line width of500km s?1.No sig-ni?cant continuum emission is detected down to a1-σlimit of0.2mJy.

3.1.10J16491

4.9+530316

No CO(3–2)emission line is seen in this QSO,although con-tinuum emission is detected in the averaged channel map with a?ux of1.6±0.28mJy(S/N?6)at a position of 164914.853,530316.35(J2000)(see Fig.3).

We have determined that the continuum is dominated by synchrotron or free-free emission,with a radio spectral index,α′,of0.1,where Sν∝να′,based on1.4and5GHz ?uxes of820±20and910±80μJy,respectively(Petric et al. 2006)which predicts a3mm?ux of1.3mJy,in agreement with our detection.Based on the3mm?ux,we estimate that approximately a third of the1.2mm?ux(1.2mJy of the S1.2mm=4.6±0.8mJy)is due to synchrotron/free-free emission rather than dust emission.We calculate a3σupper limit to the continuum-subtracted velocity-integrated CO(3–2)line?ux of0.8Jy km s?1,by rebinning the spec-trum to a frequency resolution of100MHz(280km s?1),re-sulting in a channel rms of0.7mJy,and assuming a line width of500km s?1centred on z NIR=2.2704.

3.2CO Luminosities and Gas Masses

For each submm-detected QSO,we calculate the line lumi-nosity and estimate the total cold gas mass,M gas,from the integrated CO line?ux following Solomon&Vanden Bout (2005).We assume a line luminosity ratio of r32=r21= r=L′CO(3?2)/L′CO(1?0)=1(i.e.a constant bright-ness temperature).For our QSO sample,we derive a me-dian of L′CO(1?0)=(3.1±0.9)×1010K km s?1pc2for the non-detections and a median of L′CO(1?0)=(4.2±1.2)×1010K km s?1pc2for only the CO detections(see Table3).We then convert the CO luminosities into to-tal cold gas masses(including a correction factor for He-lium)using M gas=M(H2+He)=αL′CO(1?0).We use α=0.8M⊙(K km s?1pc2)?1as the CO-to-gas conversion factor appropriate for local galaxy populations exhibiting similar levels of star formation activity to submm-bright galaxies or QSOs(e.g.ULIRGs).r=1andα=0.8are also used by Greve et al.(2005)for the SMG sample and so our QSOs can be directly compared with SMGs(if these val-ues are appropriate for both populations).The gas masses for our QSOs can be found in Table4.The median gas mass of the entire sample is(2.5±0.7)×1010M⊙when the upper limits are included,and the median gas mass for the CO-detected QSOs is(3.4±0.8)×1010M⊙.These gas masses are comparable to those observed in higher-redshift(z>～4) QSOs(e.g.Riechers et al.2006).

Our observations are of J≥2CO transitions,and it is possible that signi?cant amounts of cold,possibly subther-mal molecular gas could be present,but only detectable in lower CO transitions.Moreover,if the gas is metal-poor,it is possible that the gas mass could be higher,or we could be missing clumpy dense gas,as these CO observations pri-marily trace di?use ISM.Fortunately,these possibilities are

Gas and BHs of QSOs and SMGs at z ～2

9

Figure 3.The velocity-integrated mm emission of our nine new submm-detected QSOs.The solid contours are 1,2,3...×σ(dashed lines represent equivalent negative contours),where σrefers to the noise in each velocity-integrated map (integrated over the width of the CO line)determined by summing over the weights and includes a contribution from the phase noise visibilities.These are the dirty maps (except for RXJ124913.86which has been cleaned for presentation here to remove signi?cant sidelobe structure)and the synthesized PdBI beams are shown in the insets as hatched ellipses for reference.Crosses indicate the radio or optical position of each source.Detections:The CO-detected sources shown here are HS1002,RXJ124913.86,J131222.35,J154359.3,and HS1611.Note that the CO detection of J131222.35is marginal.J164914.9shows a strong continuum detection but no CO detection (see §3.1.10).The positional uncertainties are ?0.5,0.6,0.9,0.7,0.6and 0.5,respectively,for the CO or continuum detected sources (ranked in RA)based on the S/N of the detections and the beam sizes (see text).Non-detections:The three CO and continuum non-detections are shown for comparison and show the 3mm emission over the central 500km s ?1of the spectra (the average line width of our sample).

that the CO emission in three well-studied z >4QSOs is well-described by a single centrally-concentrated molecular gas component and is highly excited.But a concern remains that if the CO-to-gas conversion factor for a faint extended component is higher (e.g.Galactic),a higher H 2mass may be hidden in such an extended component.Nevertheless,with suitable caution these observations can still be directly comparable to the SMG sample as they su?er from the same 3.3Line Widths and Dynamical Masses

The median CO line width of our sample is 550±180km s ?1(Table 3).CO line widths can be directly converted into dynamical masses,assuming a size and inclination for the gas reservoirs.In the following we assume a disk model (Solomon &Vanden Bout 2005).Following Tacconi et al.(2006),we derive v c sin (i ),where v c is the circular veloc-ity at the outer CO radius and i is the inclination of the

10Coppin et al.

LSR Velocity [km s -1]

F l u x D e n s i t y [m J y ]

Redshift

LSR Velocity [km s -1]

F l u x D e n s i t y [m J y ]

Redshift LSR Velocity [km s -1]

F l u x D e n s i t y [m J y ]

Redshift

LSR Velocity [km s -1]

F l u x D e n s i t y [m J y ]

Redshift

LSR Velocity [km s -1]

F l u x D e n s i t y [m J y ]

Redshift LSR Velocity [km s -1]

F l u x D e n s i t y [m J y ]

Redshift

LSR Velocity [km s -1]

F l u x D e n s i t y [m J y ]

Redshift

LSR Velocity [km s -1]

F l u x D e n s i t y [m J y ]

Redshift LSR Velocity [km s -1]

F l u x D e n s i t y [m J y ]

Redshift Figure https://www.doczj.com/doc/4810921296.html,limetre spectra of the nine new submm-detected QSOs presented in our sample.The LSR velocity scale is relative to the near-infrared redshift (Table 2)of each QSO,except in the RXJ124913.86,where two central frequency tunings were used,re?ecting the uncertainty of the systemic redshift of the source at the time of the CO observations were made,and so we have arbitrarily centered the spectrum on CO(3–2)redshifted to z =2.2487.Detections:The CO-detected sources HS1002,J154359.3,J131222.35and HS1611,have been binned to a frequency resolution of 20MHz,and RXJ124913.86has been binned to 30MHz.The best-?tting CO redshift for each QSO is obtained by ?tting each spectrum with a single Gaussian distribution (overplotted)and is indicated by an arrow.See Beelen et al.(2004)for the CO spectrum of J140955.5.Non-detections:The CO-undetected spectra have been extracted from the phase tracking centre of the map and have been binned to a resolution of 20MHz for comparison with the CO detections.The spectrum of J164914.9has been continuum-subtracted,and the original spectrum is given as the dashed histogram.

of the CO line by 2.4.We calculate the dynamical masses as

M dyn =R v 2

c csc 2(i )/G ,where R is the radius of the gas disk which we assume to be 2kpc,an

d list thes

e in Table 4.With a typical resolution o

f ～40kpc for our sample (see Table 3),puttin

g a 2kpc radius limit is a major assumption (as are assumptions on the system inclination),althoug

h we explain why a 2kpc disk is probably an appropriate assumption for our QSOs in §4.2.We note that our estimates are conserva-tive as the dynamical masses will be higher by a factor of about two if a merger model is adopted (Genzel et al.2003).

For the CO-detected QSO sample we have a median dynamical mass of M(<2kpc)?(2.5±1.6)×1010csc 2(i ).The main uncertainties in the dynamical mass limits are the assumed disk size and the inclination angle,i .We note that one of our detected QSOs,RXJ124913,has a CO FWHM amongst the largest ever observed for a QSO at any red-shift,although our measurement is highly uncertain (the and z >4objects,is 300km s ?1;Solomon &Vanden Bout 2005),suggesting that RXJ124913is viewed at a high in-clination angle or that the CO disk size is larger than we have assumed.However,high resolution observations of the CO distribution are needed to better constrain the gas disk sizes,inclination angles,and hence dynamics of our QSOs.3.4

Far-infrared Luminosities,SFRs and SFEs

We have derived L FIR for the submm-detected QSOs from their restframe far-infrared ?uxes and report these in Ta-ble 4.We scale a modi?ed greybody model for far-infrared emission to match the 850or 1200μm photometry assum-ing a dust temperature of T d =40K and a dust emissivity factor of νβ,with β=1.5(see e.g.Coppin et al.2008)and integrate the SED to obtain L FIR .Our assumptions for T d and βagree with the results of Beelen et al.(2006)for three

Gas and BHs of QSOs and SMGs at z ～2

11

Figure 5.The velocity-integrated mm emission (left)and spectra (right)of serendipitous emission-line candidate ’companion’detections in two of the CO datacubes.See §3.1.3and 3.1.5for details about the individual detections.Left:A ～4σsource appears in each map (solid contours are 1,2,3...×σ),indicated by an asterisk,o?set from the PdBI phase centre (indicated by a cross).See the caption of Fig.3for details.Right:20-MHz binned spectra are shown for each serendipitous emission-line candidate companion source (see the caption of Fig.4for details).

mean ?tted T d ?(35±7)K when βis ?xed to 1.6.Note that assuming T d =50K e?ectively increases L FIR by a factor of ?2.5(Wang et al.2008).We have corrected the observed 1.2mm ?ux of J164914.9by ?–1.5mJy,assuming a radio spectral index of 0.1(see §3.1.10)for synchrotron emission,before computing L FIR .

These L FIR estimates are comparable to those of SMGs (Greve et al.2005)which is not surprising given the simi-is produced predominantly by star-formation as opposed to AGN activity (see discussion in Omont et al.2003).Re-cent observations have shown that the L FIR in QSOs is not generally contaminated by AGN,but is due to star for-mation,even in the most powerful QSOs (e.g.Lutz et al.2007;Wang et al.2007),although there are clear exceptions (e.g.Weiss et al.2007).

This calculation yields a median of L FIR =(8.0±1.9)×

12Coppin et al.

L FIR into a star-formation rate(SFR)for our sample fol-lowing Kennicutt(1998);SFR=1360±320M⊙yr?1.This relation assumes a Salpeter et al.(1955)initial mass func-tion(IMF)and applies to starbursts with ages less than 100Myr.See Omont et al.(2001)for a discussion of the full range of conversion factors taking into account the uncer-tainty in the burst age,IMF,and metallicity.

The star formation e?ciency(SFE)is a measure of how e?ective a galaxy is at converting its gas into stars and can be represented by the continuum-to-line ratio,or L FIR/L′CO,as this ratio presumably traces the star forma-tion rate per total amount of gas in a galaxy.We?nd a me-dian SFE=250±100L⊙(K km s?1pc2)?1for the full sample of submm-detected QSOs.

3.5SMBH masses

We determine M BH for our QSOs using the Greene&Ho (2005)virial M BH estimator which calculates the BH mass from the Hαor Hβemission line widths and?uxes.We can measure these lines from our near-infrared spectra(Fig.2): the observed emission line properties are given in Table2 and the BH masses are given in Table4.We?nd a me-dian M BH of(1.8±1.3)×109M⊙.Note that six out of nine of the submm-detected QSOs have very large BH masses (M BH>～109M⊙)and are among the largest masses known for local BHs(e.g.Valtonen et al.2007;Macchetto et al. 1997;Humphrey et al.2008),and that two objects in the sample are at the upper envelope of known BHs at any red-shift,con?rming that these BHs are probably hosted by the progenitors of the most massive present-day ellipticals.

Note that we have not corrected the Hα?uxes for extinction,since extinction corrections are rarely applied to broad-line?uxes when estimating virial BH masses, even when rest-frame UV lines are used.Additionally, Alexander et al.(2008)do not?nd evidence for large amounts of extinction of the broad-line region in their sam-ple of broad-line SMGs(the mean extinction is only A V～1.2mags,assuming an intrinsic ratio of Hα/Hβ=3.1and a Calzetti et al.(2000)reddening law).Applying this mean extinction correction to our sample would increase the BH masses by～15%(0.1dex).We also investigate whether the Eddington ratios,η,for our submm-detected QSOs are sim-ilar to more typical optically selected QSOs in the same redshift range.Following McLure&Dunlop(2004),we cal-culate the mass accretion rates and Eddington ratios,η,for our submm-detected QSOs and?nd a range inηconsistent with typical QSOs in the same redshift range(η～0.1–1; McLure&Dunlop2004),with a medianη≈0.7,which is slightly higher than,but consistent with typical QSOs given the large uncertainties involved.

4COMPARISON OF THE

SUBMM-DETECTED QSOS AND SMGS

We now compare the CO properties of the submm-detected QSOs with SMGs.In order to fairly compare the SMG and QSO samples,we have discarded three strongly lensed galax-ies from the CO-observed SMG sample from Greve et al.at850μm.These would lie below our sample?ux limit ex-cept for the boost from the lensing magni?cation,and they thus probe an intrinsically fainter population.This leaves a sample of17SMGs(of which11are CO-detected)from Greve et al.(2005)with a median observed850μm?ux of 8.2mJy.

We compare median values of the gas masses,CO linewidths,SFEs,M dyn,and M BH/M sph.The error bars on these median values represent the spread of values,and have been obtained by taking the standard deviation of the me-dian values from50trials of randomly drawing out the ap-propriate number of values out of each population with re-placement.We note that the dominant uncertainties in M gas, L FIR,and M dyn are likely the assumed CO-to-gas conver-sion factor,the assumedβand T d in the SED?tting,and the assumed CO radius and inclination angle,respectively, and have not been included in each error budget,although the relative M gas,L FIR,and M dyn in our samples will be correct if the same assumptions hold for all our SMGs and submm-detected QSOs.

4.1The gas masses and star-formation e?ciencies

of submm-detected QSOs and SMGs

The median gas mass of our CO-detected QSOs is M gas= (3.4±0.8)×1010M⊙(see Fig.6).This is similar to the me-dian gas mass of the comparison sample of11CO-detected SMGs(M gas=(3.0±0.5)×1010M⊙).Including the up-per limits from the CO-undetected objects in each sam-ple yields median gas masses of(2.5±0.7)×1010and (2.4±0.5)×1010M⊙for submm-detected QSOs and SMGs, respectively.Not surprisingly,we?nd that the gas mass dis-tributions of both samples are statistically indistinguishable at the present sample sizes,as we show below.

Given the moderate fraction of non-detections in both samples,we employ survival analysis(see e.g. Feigelson&Nelson1985)to estimate the intrinsic distribu-tions of the gas masses of submm-bright SMGs and QSOs. This includes the detections and upper limits,assuming that the data are censored at random(i.e.that the chance of only a gas upper limit being available for an object is inde-pendent of the true value of the gas mass).Given the cur-rent detection limit and small dynamic range in our sample, there is no strong apparent trend with the selection criteria (i.e.with submm?ux)versus whether a particular source is detected or not,and thus the assumption of the data being randomly censored seems reasonable.We employ the widely-used product-limit Avni estimator(Avni et al.1980;see also Wall&Jenkins2004)to construct a maximum-likelihood-type reconstruction of the true distribution of gas masses in the QSO and SMG samples(see Fig.6).We now test whether or not our censored samples of SMG and QSO gas masses are likely to have been drawn from the same distribution using the non-parametric Gehan test,follow-ing Feigelson&Nelson(1985).The Gehan statistic yields L n=42±39(equivalent to?1σ),revealing that the di?er-ence between the gas mass distributions for SMGs and QSOs is about35%likely due to chance.We consider35%to be a high value,and therefore conclude that the gas masses of our QSO sample are indistinguishable from that of the SMG sample at the same epoch of z～2–3for the high mass end

Gas and BHs of QSOs and SMGs at z～213 Table4.Physical properties of our submm-detected QSO sample derived from the near-infrared and CO observations.

HS1002+44004.2±0.81.1+1.3

?0.91900+2200

?1600

3.4±0.73.3+1.9

?1.5

10.14±0.2

RXJ121803.82+470854.6≤2.40.6+0.8

?0.61100+1400

?960

≤1.9––

SMM J123716.01+620323.3≤0.60.5+0.7

?0.4930+1200

?620

≤0.5–8.1±0.5

RXJ124913.86–055906.23.6±1.00.7+0.9

?0.61200+1500

?1000

2.9±0.89.7+7.0

?5.1

9.76±0.2

SMM J131222.35+423814.11.2±0.40.3+0.4

?0.2520+670

?360

1.0±0.3

2.5+2.4

?1.6

8.2±0.5

VV96J140955.5+5628278.2±0.62.7+2.9

?2.64600+5000

?4500

6.6±0.50.8+0.2

?0.1

9.28±0.2

VV96J154359.3+5359033.1±0.71.0+1.3

?0.71700+2200

?1200

2.5±0.62.0+1.2

?0.9

10.13±0.15

HS1611+47195.1±0.81.2+1.4

?1.02100+2400

?1400

4.1±0.60.4+0.2

?0.1

9.26±0.2

MM J163655+4059≤0.80.6+0.7

?0.4950+1200

?690

≤0.6–8.4±0.5

VV96J164914.9+530316≤2.20.8+0.9f

?0.71400+1600

?1100

≤1.8–9.16±0.15

14Coppin et al.

log L FIR [L sun ]

l o g L ’C O [K k m s -1 p c -2]

Figure https://www.doczj.com/doc/4810921296.html,parison of the CO and far-infrared luminosities for QSOs,LIRGs,ULIRGs,and SMGs.The line is the best-?tting relation with a form of log L ′CO =αlog L FIR +βto the LIRGs,ULIRGs and SMGs from Greve et al.(2005).The CO-observed SMGs and QSOs appear to occupy the same part of the diagram (?lled circles/stars represent CO-detections and open circles/stars are CO-nondetections),and lie just above the relation though the o?set is not signi?cant given the large errors in L FIR .This is a particularly useful diagnostic,since it does not depend on the CO-to-gas conversion factor.

4.2

The CO line pro?les and dynamical masses of submm-detected QSOs and SMGs

A comparison between the observed CO line pro?les of SMGs and QSOs requires care as we know that there is an inherent di?erence in the types of lines seen in SMGs versus QSOs:?ve of the 11SMGs detected in CO in our comparison sample show hints of double-peaked pro?les (Greve et al.2005;Tacconi et al.2006),while none of our QSO CO emis-sion lines appear double-peaked (which is perhaps not sur-prising given our small sample size and low SNR data).These double-peaked pro?les could indicate merger activ-ity (as likely for SMGs from high-resolution observations),or a rotating gas disk/ring (e.g.Genzel et al.2003).In both cases,measuring the separation between the peaks is a more appropriate measure of the dynamics of the merger or ro-tating disk system than the FWHM of the individual peaks.

In cases where double-peaked pro?les are present we need to ?t two Gaussian pro?les simultaneously in order to derive the velocity o?sets of the peaks.We have there-fore re?t all of the SMGs and QSOs with single and double Gaussian pro?les using the χ2statistic as a measure of the goodness-of-?t.We ?nd that 4/11of the CO-detected SMGs are better ?t at the >3.5σlevel with a double Gaussian pro-?le over a single Gaussian pro?le (Greve et al.2005).None of the submm-detected QSOs are signi?cantly better ?t by double Gaussian pro?les,although we note that the random ment)out of a sample containing seven single-peaked and four double-peaked pro?les (i.e.the SMG sample above)is 7percent.The fact that the QSO pro?les appear to be single-peaked could indicate that our data are too low SNR to de-tect double-peaked pro?les (especially for a merger,where both the width and peak strength of the lines can be di?er-ent),or that we are viewing the QSOs almost face-on 4if the CO is contained in a disk,or that they are at a late stage of a merger,or that they are not in the course of a merger.

For our six submm-detected QSOs we compare the CO FWHMs and dynamical masses with the CO-detected sub-sample of 11SMGs taken from the literature (Greve et al.2005;Tacconi et al.2006).The median FWHM of the six CO-detected QSOs and seven single-Gaussian pro?le ?t SMGs are (550±180)and (600±130)km s ?1,respectively.When we include the double Gaussian pro?le ?ts in the median estimate,taking the separation between the peaks as a proxy for the dynamics of the system,the median for the SMGs slightly decreases to (530±110)km s ?1.We have plotted the histogram of the submm-detected QSO and SMG line widths for comparison in Fig.8(c.f.?g.2in Carilli &Wang 2006).

The CO FWHM of QSOs and SMGs appear to be sim-ilar.To quantify this we use a standard two-sided KS test which reveals a 95%probability that the two distributions are drawn from the same parent population (this drops to 77%if only the single-line Gaussian pro?le SMGs are used).This is in stark contrast with Carilli &Wang (2006),who ?nd only a 0.9%probability for an inhomogeneous sam-ple of 12SMGs and 15QSOs from a literature compi-lation (Solomon &Vanden Bout 2005).However,we note that they have used the SMG single-Gaussian pro?le ?t line widths even in cases where clearly double-peaked CO pro-?les are apparent,leading to arti?cially broad FWHM in these cases.Their analysis is also based on a mixed sam-ple over a wide redshift range and includes strongly lensed sources.Also,the QSOs studied here have larger linewidths than have been reported for other QSOs and they are also those with the lowest SNRs and thus the most di?cult to ?t.Therefore the di?erences in our results for z ～2QSOs and those in Carilli &Wang (2006)for higher redshift QSOs can probably mainly be attributed to the small sample sizes in-volved and the somewhat di?erent approaches (rather than a real statistically signi?cant e?ect)when all possible sources of error are taken into account.

Since the CO line widths of SMGs and QSOs are simi-

4

An intrinsically double peaked pro?le could appear single-peaked if the velocity o?set between the peaks is small enough (e.g.at high SNR a close merger could show an asymmetric line pro?le,hinting at two components).We attempt to put a limit on the inclination angles in our data by investigating a simple test case.We take two Gaussian distributions with σ=150–220km s ?1(similar to those found for double-peaked SMGs)and equal line ?uxes and shift them to a velocity separation such that the FWHM of a single-Gaussian pro?le ?t is ?550km s ?1(our QSO sample median FWHM),yielding typical velocity sep-arations of 300–150km s ?1.This translates into a conservative limit on the inclination of the system of i 40?by using the relation v obs =v circ sin (i )and taking v obs ?250km s ?1and v =400km s ?1(the latter value is what is observed typically

Gas and BHs of QSOs and SMGs at z ～215

200

400

600800100012001400FWHM CO [km s -1]

01

2

3

4

N u m b e r p e r b i n

CO detected QSOs CO detected SMGs

Figure https://www.doczj.com/doc/4810921296.html,parison of the distributions of CO FWHM for 11SMGs and 6QSO with detected CO emission.The distribu-tions have medians of (530±110)km s ?1(for a mix of single and double Gaussian pro?le ?ts),and (550±180)km s ?1for SMGs and QSOs,respectively.A KS test reveals a 95%probability that the SMG and QSO CO FWHM distributions are drawn from the same parent population (see text).The histograms have been o?-set slightly for clarity.

lar,it may follow that the dynamical masses of SMGs and QSOs are also similar.The uncertainties in this are if the gas is distributed similarly in both populations (i.e.either a disk or merger scenario),if their sizes (R )are the same,and if their average inclination angles (i )are similar.R ?2kpc is believed to be appropriate for the gas distribution in SMGs (e.g.Greve et al.2005;Tacconi et al.2006,2008).One of our QSOs,J140955.5,has been observed at PdBI by Beelen et al.(2004)in CO(7–6)with a 1.0′′×0.5′′res-olution and was unresolved,giving an upper limit to the gas reservoir of 5kpc (though this could be too high a J transition to trace the bulk of the gas reservoir).No sizes have been measured for our sample of QSOs,although the CO emission seems to be con?ned to a compact region in other samples of high-z QSO (Riechers et al.2006)as in-ferred by comparing a range of CO line ratios (see also Walter et al.2004who observe gas distributed over a 2.5kpc radius and Riechers et al.2008).This agrees with the re-cent results from Maiolino et al.(2007)who ?nd that CO emission in a z ?5QSO is unresolved with a beam size of ～1arcsec,implying that the molecular gas is contained within a compact region with a radius of <3kpc.The QSO from Maiolino et al.(2007)was observed in a high J transi-tion,CO(5–4),and one might worry that such high transi-tions might not trace the bulk of the reservoir.CO(1–0)and (2–1)observations of other z >～4QSOs (e.g.Riechers et al.2006)have con?rmed that the high J transitions do trace the bulk of the reservoir,and thus provide good size constraints ies).Spatially resolved observations exist for local QSOs,albeit the data are sparse:Downes &Solomon (1998)mea-sure a R 1kpc disk in Mrk 231;Staguhn et al.(2004)?nd a R 1kpc ring-like molecular gas distribution;and Krips et al.(2007)estimate R ?3±1kpc for the CO emis-sion in a low luminosity QSO.Therefore,assuming 2–3kpc radius regions will probably still be reasonable once these galaxies are resolved.

While the average orientation of the SMGs is likely to be close to the value expected for random orientations,i ?30?,since they are not prone to orientation selection ef-fects (see Carilli &Wang 2006),the submm-detected QSOs could be orientated preferentially with respect to the sky plane due to their selection as optically luminous QSOs in the ?rst place.We ?nd evidence for this from a comparison of the [O iii ]line width (see Table 2),a proxy for the mea-sure of the stellar velocity dispersion (supporting the idea that the narrow-line region gas is in orbital motion in the gravitational potential well of the bulge;Nelson &Whittle 1996;Nelson 2000;Bonning et al.2005),with the CO line width.We ?nd that the [O iii ]FWHM are a median fac-tor of 1.9±0.5broader than the CO line pro?les (with a large scatter;ratios range from 0.6–4.3),suggesting indeed that inclination e?ects could be important in the CO line widths of QSOs (see also Shields et al.2006).The lack of the double-peaked CO pro?les in the submm-detected QSOs could be interpreted as a later phase of a merger in the typ-ical proposed evolutionary sequence,with SMGs represent-ing an earlier phase of the merger.We also know that at least two of our QSOs must have i 20(alternatively the assumed standard CO-to-H 2conversion factor could be too high or else the assumed 2kpc radius is too small)or else the cold gas mass estimates exceed the dynamical masses (even given the large error bars;see Table 4).Conversely,some of our broadest CO line detections suggest that their incli-nation angles might be larger than 20?.Other constraints come from the obscured:unobscured AGN ratio (?5–10)for a sample of SMGs,where Alexander et al.(2008)sug-gest that QSOs could be seen close to the line of sight (i =18–25?).We thus adopt the best available estimates of the average inclination angles for SMGs and QSOs of 30?and 20?,respectively,in calculating the dynamical masses below.High-resolution CO observations could help to place tighter constraints on the inclination angles of our systems.

Assuming a 2kpc radius and an average inclination angle of 20?for the submm-detected QSOs and 30?for the SMGs,we ?nd median dynamical masses,M dyn ∝FWHM CO 2R csc 2(i ),of M(<2kpc)?(2.1±1.4)×1011and (0.9±0.4)×1011M ⊙,respectively.This results in gas-to-dynamical mass fractions of ～15%and 30%for QSO and SMG populations,respectively.The gas mass accounts for a signi?cant fraction of the dynamical mass of both classes,indicating that the host galaxies are at an early evolutionary stage.4.3

The BH-spheroid ratios of submm-detected QSOs

We place a constraint on the spheroid mass by taking the dynamical mass estimate from §4.2,assuming i =20?and a spheroid radius of 2kpc or less (as we assumed for the

16Coppin et al.

0.0

0.5 1.0

1.5

2.0 2.5

3.0

Redshift

l o g M B H /M s p h

Figure 9.The variation in M BH /M sph with redshift,z ,for di?er-ent populations of galaxies and AGN.We show the ratio measured for BHs in local spheroids (dotted line;H¨a ring &Rix 2004)and those inferred for high-redshift radio-loud AGN (McLure et al.2006),QSOs (Peng et al.2006),and SMGs (Alexander et al.2008).We compare these to the ratio determined for our CO-and submm-detected QSOs.We ?nd good agreement between our estimate and that of Peng et al.(2006)for similar,optically lu-minous QSOs at z ～2,supporting the assumptions used in our analysis,although we show how our estimate changes for di?erent assumed inclination angles (dashed lines).Similarly,the typical QSO detected in our survey has a signi?cantly higher M BH /M sph ratio than inferred for z ～2SMGs.

(2.1±1.4)×1011M ⊙.In principle one should subtract the gas and BH masses,although we note that doing so makes a negligible di?erence in general,given the large error bars on M dyn ,and we also have an additional complication that two of our CO-detected QSOs have M gas >M dyn .Other uncertainties to our M sph estimates include the possibility that the cold gas in submm-detected QSOs is more widely distributed (e.g.due to entrainment in out?ows)than we have assumed here (>～2kpc),which would mean that all of our dynamical (and hence spheroidal)masses have been un-derestimated.Similarly,if the spheroid is signi?cantly more extended than the CO emission then the spheroid masses will be larger.For example,assuming a characteristic radius of ～10kpc for the spheroid would result in an increase in the mass of a factor of ～5times.

Based on these assumptions we estimate M BH /M sph ～(9+21

?6)×10?3for our QSO sample.This is approximately an order of magnitude larger than the local ratio of (1.4±0.4)×10?3for the BHs in local spheroids (Fig.9;Marconi &Hunt 2003;H¨a ring &Rix 2004).Thus our CO survey of z ～2QSOs appears to support the claim that M BH /M sph for z >1AGN is higher than observed locally (e.g.Peng et al.2006;McLure et al.2006).Our M BH /M sph ratio could be made to match the local relation by either decreasing the ten times larger than we have assumed.Our results and those of Peng et al.(2006)suggest that BH growth may oc-cur more rapidly than spheroid formation in QSOs at z ～2(although both studies could be prone to strong selection ef-fects;see Alexander et al.2008).Similar CO-based studies of rarer individual QSOs at higher redshifts also support this conclusion (e.g.Walter et al.2004;Maiolino et al.2007).

This result contrasts strongly with the M BH /M sph claimed for SMGs,which lie just below the local ratio (Fig.9;Alexander et al.2008).However,as mentioned in §3.5,six out of nine of our submm-detected QSOs have very mas-sive BHs (M BH >～109

M ⊙)which are far more massive than those inferred for typical SMGs (Alexander et al.2008;M BH ≈108M ⊙)and would place them among the largest known masses for local BHs.They are therefore likely to be considerably rarer than typical SMGs.To make a fairer comparison between SMGs and submm-detected QSOs we need to focus on those CO-detected submm-detected QSOs with M BH more similar to that found for SMGs.We have three submm-detected QSOs in our sample with BHs more similar to those of SMGs,unfortunately two of these are undetected in CO,so we cannot determine their dynamical masses.Nevertheless,the one example with a marginal CO detection,SMM J131222.35,has an implied spheroid mass of M sph =2.1×1011M ⊙(assuming i =20?)and hence M BH /M sph ?9×10?4.This is an order of magnitude be-low the ratio for the more optically luminous QSOs and places SMM J131222.35approximately on the local relation (Fig.9).While based on a single source marginally detected in CO,this result hints that submm-detected QSOs with M BH ?108M ⊙might be the SMG–QSO ‘transition’ob-jects we are interested in.We discuss this possibility further in §5.

5

DISCUSSION OF THE EVOLUTIONARY STATUS OF SUBMM-DETECTED QSOS

In this section we discuss the various observations which constrain the evolutionary relationship between SMGs and submm-detected QSOs.For the purposes of this discus-sion,we adapt the proposed evolutionary pathway of Sanders et al.(1988)and so test the scenario where the SMG population evolves through a submm-detected QSO phase,into a submm-undetected QSO and ?nally a passive elliptical.Now we ask –are the observed properties of the SMGs and submm-detected QSOs consistent with their ob-served gas depletion,amount of BH growth,and space den-sities required to link them in this evolutionary sequence?

To provide a characteristic timescale for our discussion we ?rst determine the lifetime of the submm-detected QSO phase.We assume that the far-infrared luminosities of the submm-detected QSOs arise from dust-obscured star forma-tion and that the molecular gas reservoirs we detect via their CO emission are the fuel for this star formation.In this way we determine that the QSOs have enough cold gas to sus-tain their current episode of star formation (assuming a star formation e?ciency of 100%)for τdepletion ～M gas /SFR ～2.5×1010M ⊙/1360M ⊙yr ?1～20Myr.τdepletion is on the order of a couple of dynamical times within our assumed CO radius of 2kpc (τdyn =R/v c <9Myr),which could be

Gas and BHs of QSOs and SMGs at z～217

as is commonly found with other populations of starburst galaxies such as ULIRGs at low-redshift and SMGs and other QSOs at high-z.Since we are most likely catch-ing the QSOs halfway through the current star formation episode,we infer a submm-detected QSO phase lifetime of 2×τdepletion=40Myr.The reader should of course note that τdepletion and our estimated lifetime are both lower limits if L FIR includes a signi?cant contribution from AGN activity, or if the starburst is discontinuous,or if the IMF is more top heavy than Salpeter’s IMF,or if the gas reservoirs contain signi?cant masses of sub-thermal gas(Greve et al.2006),or if the CO reservoir is widely distributed on scales?30kpc (not that there is any evidence that such galaxies exist).

Next we note that in our submm-detected QSO sam-ple CO is typically detected in only the optically luminous QSOs(see Fig.1),i.e.those QSOs with the most massive BHs(M BH>109M⊙;see Fig.10).As we discussed in§4.3 these BHs are much more massive than found in typical SMGs–is it feasible for the BHs to grow su?ciently be-tween these two phases?In Fig.10we plot M gas as a func-tion of M BH for our sample of submm-detected QSOs and for an‘average’SMG(M BH from Alexander et al.2008and M gas from Greve et al.2005).We indicate gas consump-tion and BH growth evolutionary tracks for the average SMG on timescales of10and20Myr,assuming a SFR of 1000M⊙yr?1and Eddington-limited BH growth.Fig.10 demonstrates that an SMG could evolve into a submm-detected QSO with M BH?108M⊙and M gas?0.6×1010M⊙(the median M gas of the CO-undetected QSOs) in a reasonable timescale,whereas an average SMG would need substantially(～10times)more gas and more time(>～SMG lifetime)to evolve into the optically luminous submm-detected QSOs in our sample with M BH?4×109M⊙and M gas?3.0×1010M⊙.

The gas consumption and BH growth timescales sug-gest that the optically luminous submm-detected QSOs in our sample cannot be the direct descendents of typical SMGs.Indeed the CO-detected optically luminous QSOs in our sample are extremely luminous AGN(M B≈?28, Omont et al.2003)with M BH>109M⊙and will most likely evolve into the most massive elliptical galaxies found in local rich clusters.These QSOs are so rare(?2deg?2 over all redshifts;Fan et al.2001)that they are unlikely to be associated with any known SMGs,since S850μm> 4.5mJy SMGs at z=1–3have much higher surface den-sities(?600deg?2;Coppin et al.2006)or space densities (?10?5Mpc?3;Chapman et al.2005).Even correcting for the relative timescales of the two phases still leaves an or-der of magnitude di?erence in the space densities.Based on the relative number or space densities,these optically lumi-nous submm-detected QSOs are unlikely to be involved in the SMG/QSO evolutionary sequence we are interested in testing.

In contrast,the less luminous submm-detected QSOs appear to be consistent with the Sanders et al.(1988)pro-posed evolutionary sequence,linking them to typical SMGs (see Fig.10).Thus we concentrate on the three submm-detected QSOs(including one marginal CO detection and two upper limits:SMM J131222.35,SMM J123716.01and MM J163655)with BHs more similar to that of SMGs of M BH?108M⊙.The median gas mass and SFR of this Hence,these systems can sustain their current star forma-tion episode forτdepletion<～7Myr or a total lifetime of the submm-detected QSO phase of<～15Myrs.This esti-mate of the lifetime of submm-detected QSOs can then be used to relate them to SMGs as follows.The incidence of blank-?eld submm sources identi?ed with QSOs is～4% (Chapman et al.2005).In our test scenario this is the frac-tion of an SMG lifetime in which it appears as an submm-detected QSO.Conversely,about15–25%of optically lu-minous QSOs are detected in the submm(Omont et al. 2003).Assuming that all bright SMGs go through a QSO phase,and that all QSOs are formed via SMGs and there is no strong mass or luminosity dependence of their life-times,then these relative fractions imply that the total QSO lifetime(i.e.submm-detected plus submm-undetected) is about20%/4%=5times shorter than the average lifetime of an SMG.Taking our estimated lifetime of less luminous submm-detected QSOs,<～15Myrs,these detection fractions imply total lifetimes for QSOs of<～70Myrs and for SMGs <～350Myrs.These estimates are consistent with indepen-dent estimates of the lifetimes for QSOs of～20–40Myr (Martini&Weinberg2001;Goncalves et al.2008)and～100–300Myr for SMGs from modelling(Swinbank et al. 2006).

There are two other observational links between these QSOs with M B≈?25(M BH～5×108M⊙)and SMGs. Firstly,the QSOs have space densities at z～2some5–10 times lower than SMGs(Boyle et al.2000;Chapman et al. 2003).This ratio of volume densities is roughly consistent with the ratio of the relative SMG and QSO lifetime,～100–300Myrs and～20–40Myrs,of～5times.Moreover,in§4.3 we found a hint that this subset of our submm-detected QSOs lies on the present-day M BH/M sph relation and simi-lar to SMGs,which is consistent with them being‘transition objects’between SMGs and QSOs in the proposed evolu-tionary sequence in which they would eventually evolve into passively evolving spheroids.

Although this conclusion is based on a small sam-ple of QSOs,if the dynamical and gas properties of these three QSOs are representative of this subset of submm-detected QSOs,then our results are consistent with～L?, M BH?108M⊙submm-detected QSOs being‘transition objects’in the proposed evolutionary sequence,represent-ing a very brief prodigious star formation phase where the BH is undergoing rapid growth synchronously with the stel-lar mass.Submm-detected QSOs do not possess su?ciently large gas reservoirs(containing less gas than SMGs on av-erage)to sustain the implied SFRs for very long,providing a possible explanation for why some of our submm-detected QSOs are undetected in CO(a single weak CO detection and two non-detections of M BH?108M⊙submm-detected QSOs).

Where do the optically luminous submm-detected QSOs?t into this evolutionary scheme?The average SMG would need≈1Gyr(more than its lifetime)in order to grow their BHs by the required factor of>～60to host a M BH?4×109M⊙BH seen in the optically lumi-nous subset of submm-detected QSOs.In addition,these SMGs would need to begin with gas reservoirs containing M gas>～4×1011M⊙in order to leave luminous QSOs with detectable reservoirs of residual gas.Alternatively,these rare

18Coppin et al.

SMGs,with SMG progenitors with M gas>～1011M⊙and a

surface density of?10deg?2,which are consuming their gas and growing their BHs synchronously,but which accu-mulate additional BH mass and gas through merging with other

SMGs in dense environments(i.e.damp‘dry merg-ing’;e.g.Bell et al.2004).However,we currently have not observed any SMGs with such large gas masses,and given their rarity,we will require large scale submm blank-?eld surveys such as those planned with SCUBA-2to?nd these potential progenitor SMGs.

Hence our current results split the submm-detected QSOs in our survey into two categories(although a bigger sample would probably give a continuous trend):

(1)Our observations of the optically luminous submm-detected QSOs indicate they have M BH>109M⊙.It is very di?cult to grow the BHs in typical SMGs to such large masses in the typical lifetimes of SMGs or submm-detected QSOs.We therefore suggest that these QSOs cannot be re-lated to the evolutionary cycle of typical SMGs.Instead,

these intrinsically rare QSOs most likely evolve from an equally rare subset of the SMG population–whose sur-face density is low enough that it has yet to be detected in current small-area submm surveys.Simply based upon the masses of their BHs,it is clear that these systems can only evolve into the most massive elliptical galaxies found in local rich clusters.

(2)We have found a subset of our submm-detected QSOs with lower optical luminosities which have gas and BH masses and space densities which are consistent with them being potential‘transition objects’in the proposed evolutionary scenario where SMGs evolve into QSOs.This subset of submm-detected QSOs have M BH?108M⊙and gas consumption timescales consistent with a phase of prodi-gious star formation where the BH has grown substantially during the preceding～200Myr SMG phase.The fast build up of the stellar population is consistent with an evolu-tionary path ending in a population of present-day massive spheroids,where homogeneous old stellar populations are seen(Swinbank et al.2006).

6CONCLUSIONS AND FUTURE WORK

We have carried out a millimetre interferometry survey of nine submm-detected QSOs at z=1.7–2.6in order to test the link between SMGs and QSOs at the era where these two important populations were most numerous.We include in our analysis comparable observations of a similarly selected QSO from the literature to provide a?nal sample of ten submm-detected QSOs.To support this survey we obtained near-infrared spectroscopy of these QSOs to derive accurate systemic redshifts needed to tune the millimetre receivers to the correct frequencies.The near-infrared spectra also provide Hα?uxes and line widths needed to derive reliable BH mass estimates for the QSOs.Our main?ndings are:

(1)We detect CO emission from six of the ten submm-detected QSOs in our sample,con?rming that they con-tain a signi?cant amount of molecular gas and that a large fraction of the mm emission is from starbursts.The me-dian gas mass of our sample(including non-detections)is

67891011

log M

BH

[M

sun

]

l

o

g

M g

a

s

[

M s

u

n

]

Figure10.M gas as a function of M BH for our sample of CO-observed submm-detected QSOs.Recall that the quoted errors for M gas do not include the uncertainty on the CO-to-gas conversion factor.The arrows(o?set slightly for clarity)indicate the move-ment of an average SMG in terms of its gas mass depletion and synchronous BH growth in arbitrary scalable timecales of10and 20Myr,assuming a SFR of1000M⊙yr?1and Eddington-limited BH growth.This demonstrates that an SMG could evolve into M BH?108M⊙,M gas?0.6×1010M⊙submm-detected QSOs in a reasonable timescale,whereas an average SMG would need substantially(～10×)more gas and more time(>～SMG lifetime) to evolve into the M BH?4×109M⊙,M gas?3.0×1010M⊙submm-detected QSOs.This suggests that the submm-detected QSOs hosting BHs of M BH?108M⊙with M gas 1×1010M⊙comprise‘transition objects’that we can use to probe the inter-mediary evolutionary stage between the SMG and luminous QSO phases,while the rarer more luminous QSOs with M BH 109M⊙are not related to typical SMGs.

SMGs and to z>～4QSOs.The star formation e?cien-cies of our QSOs are also comparable to those measured for SMGs,250±100L⊙(K km s?1pc2)?1,suggesting that the gross properties of the star formation in the QSOs are like those seen in SMGs.Adopting a2kpc scale size for the gas distribution in the QSOs and a typical inclination of 20?we derive a median dynamical mass of M(<2kpc)～(2.1±1.4)×1011M⊙,similar to SMGs(assuming an incli-nation angle appropriate for random inclinations).We?nd a lower incidence of double-peaked CO line pro?les in the QSOs,compared to SMGs,which we believe results from a selection bias towards lower average inclination angles for the QSOs.

(2)Our near-infrared spectroscopy indicates a median black hole mass in our QSO sample of(1.8±1.3)×109M⊙. Combined with our dynamical estimates of the spheroid mass,these yield M BH/M sph～9×10?3.This M BH/M sph ratio for this sample of submm-detected QSOs at z=2is an order of magnitude larger than the local ratio,although

Gas and BHs of QSOs and SMGs at z～219

CO radii and inclination angles.This ratio is also signi?-cantly above that seen for SMGs at z～2.However,this comparison masks a broad range in BH masses within our QSO sample and so we split the sample into two subsets based on their BH masses.

(3)Looking at the optically luminous submm-detected QSOs in our sample we?nd that we detect CO emission in 5/6of these QSOs.However,the estimated BH masses for these QSOs,M BH?109–1010M⊙,are too large(and their number densities too small)for them to be related to typical SMGs in a simple evolutionary cycle.We propose that the progenitors of these most massive QSOs are a rare subset of SMGs with M gas>4×1011M⊙with a number density of?10deg?2which will be possible to detect with future SCUBA-2surveys.

(4)For the optically less luminous(～L?)submm-detected QSOs,we marginally detect one source in CO and obtain sensitive limits for three further QSOs.The BH masses for these systems are M BH?108M⊙,similar to the estimates for BHs in SMGs.These submm-detected QSOs are consistent with being‘transition’objects between SMGs and submm-undetected QSOs,as we show it is feasible to link their BH masses to those of SMGs by Eddington lim-ited growth for a period comparable to the gas depletion timescale of the QSOs,～10Myrs.The space density of these QSOs is also in rough agreement with that expected for the descendents of SMGs given current estimates of the relative lifetimes of QSOs and SMGs.We conclude that these～L?, M BH>108M⊙submm-detected QSOs are consistent with being in a very brief prodigious star formation phase,and that they simply do not possess su?ciently large gas reser-voirs to sustain the SFR(which is why these might be less often detected in CO),although a larger sample of CO ob-servations of submm-detected QSOs with these BH masses is required for con?rmation.

To make further progress on understanding the evo-lutionary links between SMGs and QSOs requires a larger survey of the submm and CO emission from typical QSOs (M B≈?25).In addition,measurements of other CO transi-tions for the submm-detected QSOs(e.g.from IRAM30-m, ALMA,EVLA,and SKA)are required to place better con-straints on the temperature and density of the molecular gas and thus provide a more accurate determination of the line luminosity ratios and hence total gas masses of these systems.Similarly,higher resolution CO observations are essential to put strong constraints on the reservoir sizes and inclination angles,and hence M dyn,needed to compare the two populations.Finally,better measurements of the far-infrared SEDs(with SABOCA,SCUBA-2or Herschel)will yield more accurate measures of L FIR and T dust for submm-detected QSOs to constrain the contribution from an AGN component.

7ACKNOWLEDGMENTS

This work is based on observations carried out with the IRAM PdBI.IRAM is supported by INSU/CNRS(France), MPG(Germany)and IGN(Spain).This work also makes U/06A/https://www.doczj.com/doc/4810921296.html,IRT is operated by the Joint Astronomy Cen-tre on behalf on the UK Science and Technology Facilities Council(STFC).Ned Wright’s Javascript Cosmology Calcu-lator was used in preparing this paper(Wright et al.2006).

KEKC and AMS acknowledge support from the STFC. DMA and IRS acknowledge support from the Royal So-ciety.We thank Thomas Greve for sending us the SMG CO spectra so that we could obtain the parameters of the double-Gaussian pro?le?ts.We thank Scott Chapman and Tadafumi Takata for allowing us to use their LRIS/UV and OHS/near-IR spectra of SMM J123716.01,SMM J131222.35 and MM J163655.We thank Laura Hainline,David Frayer, and an anonymous referee for useful feedback which helped improve the clarity and presentation of the paper. REFERENCES

Alexander D.M.et al.,2003,AJ,125,383

Alexander D.M.,Smail I.,Bauer F.E.,Chapman S.C.,Blain

A.W.,Brandt W.N.,Ivison R.J.,2005a,Nat.,434,738 Alexander D.M.,Bauer F.E.,Chapman S.C.,Smail I.,Blain

A.W.,Brandt W.N.,Ivison R.J.,2005b,ApJ,632,736 Alexander D.M.et al.,2008,AJ,135,1968

Avni Y.,Soltan A.,Tananbaum H.,Zamorani G.,1980,ApJ, 238,800

Beelen A.et al.,2004,A&A,423,441

Beelen A.,Cox P.,Benford D.,Dowell C.D.,Kov′a cs A., Bertoldi F.,Omont A.,Carilli C.L.,2006,ApJ,642, 694

Bell E.F.et al.,2004,ApJ,608,752

Biggs A.D.,Ivison R.J.,2006,MNRAS,371,963

Bonning E.W.,Shields G.A.,Salviander S.,McLure R.J., 2005,ApJ,626,89

Borys C.,Smail I.,Chapman S.C.,Blain A.W.,Alexander

D.M.,Ivison R.J.,2005,ApJ,635,853

Bower R.G.,Benson A.J.,Malbon R.,Helly J.C.,Frenk C.S., Baugh C.M.,Cole S.,Lacey C.G.,2006,MNRAS,370, 645

Boyle B.J.,Shanks T.,Croom S.M.,Smith R.J.,Miller L., Loaring N.,Heymans C.,2000,MNRAS,317,1014 Calzetti D.,Armus L.,Bohlin R.C.,Kinney A.L.,Koornneef J.,Storchi-Bergmann T.,2000,ApJ,533,682

Carilli C.L.,Wang R.,2006,AJ,131,2763

Cavanagh B.,Hirst P.,Jenness T.,Economou F.,Currie M.J.,Todd S.,Ryder S.D.,2003,in Payne H.E.,Jedrze-jewski R.I.,Hook R.N.,eds,ASP Conf.Ser.Vol.295, Astronomical Data Analysis Software and Systems XII, p.237

Chapman S.C.,Blain A.W.,Ivison R.J.,Smail I.R.,2003, Nat.,422,695

Chapman S.C.,Blain A.W.,Smail I.,Ivison R.J.,2005,ApJ, 622,772

Coppin K.,et al.,2006,MNRAS,372,1621

Coppin K.,et al.,2008,MNRAS,384,1597

Di Matteo T.,Springel V.,Hernquist L.,2005,Nat.,433, 604

Downes D.,Solomon P.M.,1998,ApJ,507,615

Fan X.et al.,1999,AJ,118,1

Fan X.et al.,2001,AJ,122,2833

20Coppin et al.

Frayer D.T.,Ivison R.J.,Scoville N.Z.,Yun M.,Evans A.S., Smail I.,Blain A.W.,Kneib J.-P.,1998,ApJ,506,L7 Gebhardt K.et al.,2000,ApJ,539,13

Genzel R.,Baker A.J.,Tacconi L.J.,Lutz D.,Cox P.,Guil-loteau S.,Omont A.,2003,ApJ,584,633

Goncalves T.S.,Steidel C.C.,Pettini M.,2008,ApJ,676, 816

Greene J.E.,Ho L.C.,2005,ApJ,630,122

Greve T.R.,Ivison R.J.,Bertoldi F.,Stevens J.A.,Dunlop J.S.,Lutz D.,Carilli C.L.,2004,MNRAS,354,779 Greve T.R.et al.,2005,MNRAS,359,1165

Greve T.R.,Hainline L.J.,Blain A.W.,Smail I.,Ivison R.J., Papadopoulos P.P.,2006,AJ,132,1938

Guilloteau S.,Omont A.,Cox P.,McMahon R.G.,Petitjean P.,1999,A&A,349,363

Guilloteau S.,Lucas R.,2000,in Mangum J.G.,Radford S.J.E.,eds,ASP Conf.Ser.Vol.217,Imaging at Radio through Submillimeter Wavelengths,p.299

Hainline L.J.,Scoville N.Z.,Yun M.S.,Hawkins D.W., Frayer D.T.,Isaak K.G.,2004,ApJ,609,61

H¨a ring N.,Rix H.-W.,2004,ApJ,604,L89

Hopkins P.F.,Hernquist L.,Cox T.J.,Di Matteo T.,Martini P.,Robertson B.,Springel V.,2005,ApJ,630,705 Humphrey P.J.,Buote D.A.,Brighenti F.,Gebhardt K., Mathews W.G.,2008,preprint(astro-ph/0801.3461) Ivison R.J.,et al.,2007,MNRAS,380,199

Kennicutt R.C.,1998,ARA&A,36,189

Kov′a cs A.,Chapman S.C.,Dowell C.D.,Blain A.W.,Ivison R.J.,Smail I.,Phillips T.G.,2006,ApJ,650,592 Krips M.,Eckart A.,Neri R.,Bertram T.,Straubmeier C, Fischer S.,Staguhn J.G.,Vogel S.N.,A&A,464,187 Lutz D.et al.,2007,ApJ,661,L25

Macchetto F.,Marconi A.,Axon D.J.,Capetti A.,Sparks W.,Crane P.,1997,ApJ,489,579

Magorrian J.et al.,1998,AJ,115,2285

Maiolino R.et al.,2007,A&A,472,L33

Marconi A.,Hunt L.K.,2003,ApJ,589,21

Martini P.,Weinberg D.H.,2001,ApJ,547,12

McLure R.J.,Dunlop J.S.,2004,MNRAS,352,1390 McLure R.J.,Jarvis M.J.,Targett T.A.,Dunlop J.S.,Best P.N.,2006,MNRAS,368,1395

Nelson C.H.,2000,ApJ,544,L91

Nelson C.H.,Whittle M.,1996,ApJ,465,96

Omont,A.,McMahon R.G.,Cox P.,Kreysa E.,Bergeron J., Pajot F.,Storrie-Lombardi L.J.,1996,A&A,315,1 Omont,A.,Cox P.,Bertoldi F.,McMahon R.G.,Carilli C., Isaak K.G.,2001,A&A,374,371

Omont,A.,Beelen,A.,Bertoldi F.,Cox P.,Carilli C.L., Priddey R.S.,McMahon R.G.,Isaak K.G.,2003,A&A, 398,857

Page M.J.,Stevens J.A.,Mittaz J.P.D.,Carrera F.J.,2001, Sci,294,2516

Page M.J.,Stevens J.A.,Ivison R.J.,Carrera F.J.,2004, ApJ,611,L85

Peng C.Y.,Impey C.D.,Ho L.C.,Barton E.J.,Rix H.-W., 2006,ApJ,640,114

Petric A.O.,Carilli C.L.,Bertoldi F.,Beelen A.,Cox P., Omont A.,2006,AJ,132,1307

Ramsay Howat S.K.et al.,2004,in Moorwood A.F.M., Masanori I.,eds,SPIE Conf.Proc.Vol.5492,Ground-based Instrumentation for Astronomy,p.1160Riechers D.A.,Walter F.,Brewer B.J.,Carilli C.L., Lewis G.F.,Bertoldi F.,Cox P.,2008,preprint(astro-ph/0806.4616)

Salpeter E.E.,1955,121,161

Sanders D.B.,Soifer B.T.,Elias J.H.,Neugebauer G., Matthews K.,1988,ApJ,328,L35

Sanders D.B.,Scoville N.Z.,Soifer B.T.,1991,ApJ,370,158 Seaquist E.R.,Ivison R.J.,Hall P.J.,1995,MNRAS,276, 867

Shields G.A.,Menezes K.L.,Massart C.A.,Vanden Bout P., 2006,ApJ,641,683

Smail I.,Ivison R.J.,Blain A.W.,1997,ApJL,490,L5 Smail I.,Chapman S.C.,Blain A.W.,Ivison R.J.,2004,ApJ, 616,71

Stevens J.A.,Page M.J.,Ivison R.J.,Smail I.,Carrera F.J., 2004,ApJ,604,L17

Stevens J.A.,Page M.J.,Ivison R.J.,Carrera F.J.,Mittaz J.P.D.,Smail I.,McHardy I.M.,2005,MNRAS,360, 610

Solomon P.M.,Downes D.,Radford S.J.E.,Barrett J.W., 1997,ApJ,478,144

Solomon P.M.,Vanden Bout P.A.,2005,ARA&A,43,677 Spergel D.N.et al.,2003,ApJS,148,175

Staguhn J.G.,Schinnerer E.,Eckart A.,Scharw¨a chter J., 2004,ApJ,609,85

Swinbank,A.M.,Smail,I.,Chapman,S.C.,Blain,A.W., Ivison,R.J.,Keel,W.C.,2004,ApJ,617,64 Swinbank A.M.et al.,2005,MNRAS,359,401 Swinbank,A.M.,Chapman,S.C.,Smail,I.,Lindner,C., Borys,C.,Blain,A.W.,Ivison R.J.,Lewis,G.F.,2006, MNRAS,371,465

Tacconi L.J.,Genzel R.,Lutz D.,Rigopoulou D.,Baker A.J., Iserlohe C.,Tecza M.,2002,ApJ,580,73

Tacconi L.J.et al.,2006,ApJ,640,228

Tacconi L.J.et al.,2008,ApJ,680,246

Takata T.,Sekiguchi K.,Smail I.,Chapman S.C.,Geach J.E.,Swinbank A.M.,Blain A.,Ivison R.J.,2006,ApJ, 651,713

Valtonen M.J.,et al.2007,American Astronomical Society Meeting Abstracts,211,#112.07

Wall J.V.,Jenkins C.R.,2003,Practical Statistics for As-tronomers.Cambridge Univ.Press,Cambridge Walter F.,Carilli C.,Bertoldi F.,Menten K.,Cox P.,Lo K.Y.,Fan X.,Strauss M.A.,2004,ApJ,615,L17 Wang R.et al.,2007,AJ,134,617

Wang R.et al.,2008,AJ,135,1201

Weiss A.,Downes D.,Neri R.,Walter F.,Henkel C.,Wilder

D.J.,Wagg J.,Wiklind T.,2007,A&A,467,955 Wright

E.L.,2006,PASP,118,1711

Yao L.,Seaquist E.R.,Kuno N.,Dunne L.,2003,ApJ,588, 771

(完整word版)各种接口针脚定义大全,推荐文档

3.5mm插头 最常见的立体声耳机分三层，标准分布为“左右地红白”（从端部到根部依次是左声道、右声道、地线，其中左声道常用红色线皮，右声道常用白色的）。 最常见的是银白色的和铜黄色的，银色的是铜镀银，铜黄色的就是铜。由于银的稳定性和电子工程性优于铜，所以铜镀上银后可以升级使用该插头设备的用户体验。 USB接口 USB是一种常用的pc接口，他只有4根线，两根电源两根信号，故信号是串行传输的，usb接口也称为串行口，usb2.0的速度可以达到480Mbps。可以满足各种工业和民用需要.USB接口的输出电压和电流是： +5V 500mA 实际上有误差，最大不能超过+/-0.2V 也就是4.8-5.2V 。usb接口的4根线一般是下面这样分配的，需要注意的是千万不要把正负极弄反了，否则会烧掉usb设备或者电脑的南桥芯片：黑线：gnd 红线：vcc 绿线：data+ 白线：data-

USB接口定义图 USB接口定义颜色 一般的排列方式是：红白绿黑从左到右 定义： 红色－USB电源：标有－VCC、Power、5V、5VSB字样 白色－USB数据线：（负）－DATA-、USBD-、PD-、USBDT- 绿色－USB数据线：（正）－DATA+、USBD+、PD+、USBDT+ 黑色－地线： GND、Ground USB接口的连接线有两种形式，通常我们将其与电脑接口连接的一端称为“A”连接头，而将连接外设的接头称为“B”连接头(通常的外设都是内建USB数据线而仅仅包含与电脑相连的“A”连接头)。 USB接口是一种越来越流行的接口方式了，因为USB接口的特点很突出：速度快、兼容性好、不占中断、可以串接、支持热插拨等等，

CameraLink接口数字相机图像显示装置解读

Camera Link接口数字相机图像显示装置（技术）摘要:由于目前基于CameraLink接口的各种相机都不能直接显示,因此本文基于Xilinx公司的Spartan3系列FPGA XC3S1000-6FG456I设计了一套实时显示系统,该系统可以在不通过系统机的情况下,完成对相机CameraLink信号的接收、缓存、读取并显示。系统采用两片SDRAM作为帧缓存,将输入的CameraLink信号转换成帧频为75Hz,分辨率为1,024×768的XGA格式信号,并采用ADV7123JST芯片实现数模转换,将芯片输出的信号送到VGA接口,通过VGA显示器显示出来。设计的系统可以应用于各种基于CameraLink接口的相机输出信号的实时显示。 关键词:CameraLink; FPGA; SDRAM控制器;实时显示 Research on the Real-time Display Technology Based on CameraLink Abstract: All cameras based on the CameraLink interface cannot be displayed directly at present. Therefore, we designed a real-time display system based on the Xilinx Spartan3 FPGA XC3S1000-6FG456I.Our system could receive, store, read and display the CameraLink signal without the system computer. Two SDRAMs were used as frame storage. Input CameraLink signal was converted to XGA signal with 1024×768 pixles/frame at 75 frame/s. The ADV7123JST was used as D/A convertor. Its output signal was transmitted to the VGA interface and displayed on the screen of the VGA monitor. Our system could display the output signal of all cameras based on the CameraLink interface. Keywords: CameraLink; FPGA; SDRAM controller; real-time display

完整word版各种接口针脚定义大全

3.5mm 插头 最常见的立体声耳机分三层，标准分布为“左右地红白”（从端部到根部依次是左声道、右声道、地线，其中左声道常用红色线皮，右声道常用白色的）。 最常见的是银白色的和铜黄色的，银色的是铜镀银，铜黄色的就是铜。由于银的稳定性和电子工程性优于铜，所以铜镀上银后可以升级使用该插头设备的用户体验。 USB接口 USB是一种常用的pc接口，他只有4根线，两根电源两根信号，故信号是串行传输的，usb接口也称为串行口， usb2.0的速度可以达到480Mbps。可以满足各种工业和民用需要.USB接口的输出电压和电流是： +5V 500mA 实际上有误差，最大不能超过+/-0.2V 也就是4.8-5.2V 。usb接口的4根线一般是下面这样分配的，需要注意的是千万不要把正负极弄反了，

否则会烧掉usb设备或者电脑的南桥芯片：黑线：gnd 红线：vcc 绿线：data+ 白线：data- 1 USB接口定义图 USB接口定义颜色 一般的排列方式是：红白绿黑从左到右 定义： 红色－USB电源：标有－VCC、Power、5V、5VSB字样 白色－USB数据线：（负）－DATA-、USBD-、PD-、USBDT- 绿色－USB数据线：（正）－DATA+、USBD+、PD+、USBDT+ 黑色－地线： GND、Ground USB接口的连接线有两种形式，通常我们将其与电脑接口连接的一端称为“A”连接头，而将连接外设的接头称为“B”连接头(通常的外设都是内建USB数据线而仅仅包含与电脑相连的“A”连

接头)。 USB接口是一种越来越流行的接口方式了，因为USB接口的特点很突出：速度快、兼容性好、不占中断、可以串接、支持热插拨等等， 2 所以如今有许多打印机、扫描仪、数字摄像头、数码相机、MP3播放器、MODEM等都开始使用USB做为接口模式，USB接口定义也很简单： 1 ＋5V 2 DATA－数据－ 3 DATA＋数据＋ 4 GND 地 串口 主板一般都集成两个串口，可Windows却最多可提供8个串口资源供硬件设置使用(编号COM1到COM8)，虽然其I/O地址不相同，但是总共只占据两个IRQ(1、3、5、7共享IRQ4，2、4、6、8共享IRQ3)，平常我们常用的是COM1～COM4这四个端口。我们经常在使用中遇到这个问题——如果在COM1上安装了串口鼠标或其他外设，就不能在COM3上安装如Modem之类的其它硬件，这就是因为IRQ设置冲突而无法工作。这时玩家们可以将另外的外设安装在COM2或4。 标准的串口能够达到最高115Kbps的数据传输速度，而一些增强型串口如ESP(Enhanced Serial Port，增强型串口) 、Super

JTAG各类接口针脚定义及含义

JTAG各类接口针脚定义及含义 JTAG有10pin的、14pin的和20pin的，尽管引脚数和引脚的排列顺序不同，但是其中有一些引脚是一样的，各个引脚的定义如下。 一、引脚定义 Test Clock Input (TCK) -----强制要求1 TCK在IEEE1149.1标准里是强制要求的。TCK为TAP的操作提供了一个独立的、基本的时钟信号，TAP的所有操作都是通过这个时钟信号来驱动的。 Test Mode Selection Input (TMS) -----强制要求2 TMS信号在TCK的上升沿有效。TMS在IEEE1149.1标准里是强制要求的。TMS信号用来控制TAP状态机的转换。通过TMS信号，可以控制TAP在不同的状态间相互转换。 Test Data Input (TDI) -----强制要求3 TDI在IEEE1149.1标准里是强制要求的。TDI是数据输入的接口。所有要输入到特定寄存器的数据都是通过TDI接口一位一位串行输入的（由TCK驱动）。 Test Data Output (TDO) -----强制要求4 TDO在IEEE1149.1标准里是强制要求的。TDO是数据输出的接口。所有要从特定的寄存器中输出的数据都是通过TDO接口一位一位串行输出的（由TCK驱动）。 Test Reset Input (TRST) ----可选项1 这个信号接口在IEEE 1149.1标准里是可选的，并不是强制要求的。TRST可以用来对TAPController进行复位（初始化）。因为通过TMS也可以对TAP Controll进行复位（初始化）。所以有四线JTAG与五线JTAG之分。 (VTREF) -----强制要求5 接口信号电平参考电压一般直接连接Vsupply。这个可以用来确定ARM的JTAG接口使用的逻辑电平（比如3.3V还是5.0V?） Return Test Clock ( RTCK) ----可选项2 可选项，由目标端反馈给仿真器的时钟信号,用来同步TCK信号的产生,不使用时直接接地。System Reset ( nSRST)----可选项3 可选项,与目标板上的系统复位信号相连,可以直接对目标系统复位。同时可以检测目标系统的复位情况，为了防止误触发应在目标端加上适当的上拉电阻。 USER IN 用户自定义输入。可以接到一个IO上，用来接受上位机的控制。 USER OUT 用户自定义输出。可以接到一个IO上，用来向上位机的反馈一个状态 由于JTAG经常使用排线连接，为了增强抗干扰能力，在每条信号线间加上地线就出现了这种20针的接口。但事实上，RTCK、USER IN、USER OUT一般都不使用，于是还有一种14针的接口。对于实际开发应用来说，由于实验室电源稳定，电磁环境较好，干扰不大。

CameraLink 图像采集接口电路1 (2)

CameraLink 图像采集接口电路 1．Camera Link标准概述 Camera Link 技术标准是基于 National Semiconductor 公司的 Channel Link 标准发展而来的，而 Channel Link 标准是一种多路并行 LVDS 传输接口标准。 低压差分信号（ LVDS ）是一种低摆幅的差分信号技术，电压摆幅在 350mV 左右，具有扰动小，跳变速率快的特点，在无失传输介质里的理论最大传输速率在 1.923Gbps 。 90 年代美国国家半导体公司（ National Semiconductor ）为了找到平板显示技术的解决方案，开发了基于 LVDS 物理层平台的 Channel Link 技术。此技术一诞生就被进行了扩展，用来作为新的通用视频数据传输技术使用。 如图1 所示， Channel Link 由一个并转串信号发送驱动器和一个串转并信号接收器组成，其最高数据传输速率可达 2.38G 。数据发送器含有 28 位的单端并行信号和 1 个单端时钟信号，将 28 位 CMOS/TTL 信号串行化处理后分成 4 路 LVDS 数据流，其 4 路串行数据流和 1 路发送 LVDS 时钟流在 5 路 LVDS 差分对中传输。接收器接收从 4 路 LVDS 数据流和 1 路 LVDS 时钟流中把传来的数据和时钟信号恢复成 28 位的 CMOS/TTL 并行数据和与其相对应的同步时钟信号。 图1 camera link接口电路 2．Channel Link标准的端口和端口分配 2.1 ．端口定义 一个端口定义为一个 8 位的字，在这个 8 位的字中，最低的 1 位（ LSB ）是 bit0 ，最高的 1 位（ MSB ）是 bit7 。 Camera Link 标准使用 8 个端口，即端口 A 至端口 H 。

最新各种接口针脚定义大全

各种接口针脚定义大 全

3.5mm插头 最常见的立体声耳机分三层，标准分布为“左右地红白”（从端部到根部依次是左声道、右声道、地线，其中左声道常用红色线皮，右声道常用白色的）。 最常见的是银白色的和铜黄色的，银色的是铜镀银，铜黄色的就是铜。由于银的稳定性和电子工程性优于铜，所以铜镀上银后可以升级使用该插头设备的用户体验。 USB接口 USB是一种常用的pc接口，他只有4根线，两根电源两根信号，故信号是串行传输的，usb接口也称为串行口，usb2.0的速度可以达到480Mbps。可以满足各种工业和民用需要.USB接口的输出电压和电流是： +5V 500mA 实际上有误差，最大不能超过+/-0.2V 也就是4.8-5.2V 。usb接口的4根线一般是下面这样分配的，需要注意的是千万不要把正负极弄反了，否则会烧掉usb设备或者电脑的南桥芯片：黑线：gnd 红线：vcc 绿线：data+ 白线：data-

USB接口定义图 USB接口定义颜色 一般的排列方式是：红白绿黑从左到右 定义： 红色－USB电源：标有－VCC、Power、5V、5VSB字样 白色－USB数据线：（负）－DATA-、USBD-、PD-、USBDT- 绿色－USB数据线：（正）－DATA+、USBD+、PD+、USBDT+ 黑色－地线： GND、Ground USB接口的连接线有两种形式，通常我们将其与电脑接口连接的一端称为“A”连接头，而将连接外设的接头称为“B”连接头(通常的外设都是内建USB数据线而仅仅包含与电脑相连的“A”连接头)。

USB接口是一种越来越流行的接口方式了，因为USB接口的特点很突出：速度快、兼容性好、不占中断、可以串接、支持热插拨等等，所以如今有许多打印机、扫描仪、数字摄像头、数码相机、MP3播放器、MODEM等都开始使用USB做为接口模式，USB接口定义也很简单： 1 ＋5V 2 DATA－数据－ 3 DATA＋数据＋ 4 GND 地 串口 主板一般都集成两个串口，可Windows却最多可提供8个串口资源供硬件设置使用(编号COM1到COM8)，虽然其I/O地址不相同，但是总共只占据两个IRQ(1、3、5、7共享IRQ4，2、4、6、8共享IRQ3)，平常我们常用的是COM1～COM4这四个端口。我们经常在使用中遇到这个问题——如果在COM1上安装了串口鼠标或其他外设，就不能在COM3上安装如Modem之类的其它硬件，这就是因为IRQ设置冲突而无法工作。这时玩家们可以将另外的外设安装在COM2或4。 标准的串口能够达到最高115Kbps的数据传输速度，而一些增强型串口如ESP(Enhanced Serial Port，增强型串口) 、Super

CameraLink接口 时序控制

CameraLink接口 1.CameraLink接口简介 1.1CameraLink标准概述 Camera Link 技术标准是基于 National Semiconductor 公司的 Channel Link 标准发展而来的，而 Channel Link 标准是一种多路并行 LVDS 传输接口标准。 低压差分信号（ LVDS ）是一种低摆幅的差分信号技术，电压摆幅在 350mV 左右，具有扰动小，跳变速率快的特点，在无失传输介质里的理论最大传输速率在 1.923Gbps 。 90 年代美国国家半导体公司（ National Semiconductor ）为了找 到平板显示技术的解决方案，开发了基于 LVDS 物理层平台的 Channel Link 技术。此技术一诞生就被进行了扩展，用来作为新的通用视频数据传输技术使用。 如图1.1所示， Channel Link 由一个并转串信号发送驱动器和一个串转并信号接收器组成，其最高数据传输速率可达 2.38G 。数据发送器含有 28 位的单端并行信号和 1 个单端时钟信号，将 28 位 CMOS/TTL 信号串行化处理后分成 4 路 LVDS 数据流，其 4 路串行数据流和 1 路发送 LVDS 时钟流在 5 路 LVDS 差分对中传输。接收器接收从 4 路 LVDS 数据流和 1 路 LVDS 时钟流中把传来的数据和时钟信号 恢复成 28 位的 CMOS/TTL 并行数据和与其相对应的同步时钟信号。 图1.1 camera link接口电路

1.2CameraLink端口和端口分配 1.2.1端口分配 在基本配置模式中，端口 A 、 B 和 C 被分配到唯一的 Camera Link 驱动器 / 接收器对上；在中级配置模式中，端口 D 、 E 和 F 被分配到第二个驱动器 / 接收器对上；在完整配置模式中，端口 A 、 B 和 C 被分配到第一个驱动器 / 接收器对上，端口 D 、 E 和 F 被分配到第二个驱动器 / 接收器对上，端口 G 和 H 被分配到第三个驱动器 / 接收器对上。表1.1给出了三种配置的端口分配， Camera Link 芯片及连接器的使用数量情况。 表1.1 3种配置模式的端口分配 图1.2 各种配置下的端口连接关系

电脑主板各种接口及引脚定义

电脑主板各种接口及引脚定义，下图为常见的主板外设接口 首先是ATX 20-Pin电源接口电源接口，根据下图你可方便判断和分辨。 现在为提高CPU的供电，从P4主板开始，都有个4P接口，单独为CPU供电，在此也已经标出。

主板上CPU等网风扇接口。 主板上音频线接口。

主板SATA串口硬盘接口。 PS/2接口 鼠标和键盘绝大多数采用PS/2接口，鼠标和键盘的PS/2接口的物理外观完全相同，初学者往往容易插错，以至于业界不得不在PC'99规范中用两种不同的颜色来将其区别开，而事实上它们在工作原理上是完全相同的，从下面的PS/2接口针脚定义我们就可以看出来。

上图的分别为AT键盘（既常说的大口键盘），和PS2键盘（即小口键盘），如今市场上PS2键盘的数量越来越多了，而AT键盘已经要沦为昨日黄花了。因为键盘的定义相似，所以两者有共同的地方，各针脚定义如下： 1、DATA 数据信号 2、空 3、GND 地端 4、＋5V 5、CLOCK 时钟 6 空（仅限PS2键盘） USB接口 USB(Universal Serial Bus，通用串行总线)接口是由Compaq、IBM、Microsoft 等多家公司于1994年底联合提出的接口标准，其目的是用于取代逐渐不适应外设需求的传统串、并口。1996年业界正式通过了USB1.0标准，但由于未获当时主流的Win95支持(直到Win95 OSR2才通过外挂模块提供对USB1.0的支持)而未得到普及，直到1998年USB1.1标准确立和Win98内核正式提供对USB接口的直接支持之后，USB才真正开始普及，到今天已经发展到USB2.0标准。 USB接口的连接线有两种形式，通常我们将其与电脑接口连接的一端称为“A”连接头，而将连接外设的接头称为“B”连接头(通常的外设都是内建USB数据线而仅仅包含与电脑相连的“A”连接头)。Negative data ,positive data

HDMI接口引脚定义及对照表

Edit by Simon on 2013-03-02 HDMI 接口引脚定义对照表 Pin A Type B Type C Type D Type 总共有19pin,规格为4.45mm ×13.9mm ，为最常见的HDMI 接头规格,相对等于DVI Single-Link 传输。在HDMI 1.2a 之前,最大能传输165MHz 的TMDS,所以最大传输规格1600x1200。 总共有29pin,可传输HDMI A type 两倍的TMDS 资料量，相对等于DVI Dual-Link 传输，用于传输高分辨率(2560x1600以上)。 总共有19pin,可以说是缩小版的HDMI A type,但脚位定义有所改变。主要是用在便携式装置上，例如DV 、数码相机、便携式多媒体播放机等。(又称mini-HDMI ，但实际上HDMI 官方并没此名称。) 俗称Micro HDMI 是定义为HDMI 1.4版本的，保持hdmi 标准的19pin .但是尺寸与微型USB 的接口差不多，尺寸为2.8mm ×6.4mm ，主要应用在一些小型的移动设备上，如手机，MP4等等。 1TMDS Data2+TMDS Data2+TMDS Data2Shield Hot Plug Detect 2TMDS Data2Shield TMDS Data2Shield TMDS Data2+Utility 3TMDS Data2–TMDS Data2–TMDS Data2–TMDS Data2+4TMDS Data1+TMDS Data1+TMDS Data1Shield TMDS Data2Shield 5TMDS Data1Shield TMDS Data1Shield TMDS Data1+TMDS Data2-6TMDS Data1–TMDS Data1–TMDS Data1–TMDS Data1+7TMDS Data0+TMDS Data0+TMDS Data0Shield TMDS Data1Shield 8TMDS Data0Shield TMDS Data0Shield TMDS Data0+TMDS Data1-9TMDS Data0–TMDS Data0–TMDS Data0–TMDS Data0+10TMDS Clock+TMDS Clock+TMDS Clock Shield TMDS Data0Shield 11TMDS Clock Shield TMDS Clock Shield TMDS Clock+TMDS Data0-12TMDS Clock–TMDS Clock–TMDS Clock–TMDS Clock+13CEC TMDS Data5+DDC/CEC Ground TMDS Clock Shield 14Reserved (N.C.)TMDS Data5Shield CEC TMDS Clock-15SCL TMDS Data5-SCL CEC 16SDA TMDS Data4+SDA DDC/CEC Ground 17DDC/CEC Ground TMDS Data4Shield Reserved (N.C.)SCL 18+5V Power TMDS Data4-+5V Power SDA 19Hot Plug Detect TMDS Data3+Hot Plug Detect +5V Power 20TMDS Data3Shield 21TMDS Data3-22CEC 23Reserved (N.C.)24Reserved (N.C.)25SCL 26SDA 27DDC/CEC Ground 28+5V Power 29 Hot Plug Detect

并行接口引脚定义

PC 并行接口引脚定义 2009-08-11 16:27:53 PC 并行接口外观是 25 针母插座： 引脚定义 Pin Name Dir Description 1/STROBE Strobe 2D0Data Bit 0 3D1Data Bit 1 4D2Data Bit 2 5D3Data Bit 3 6D4Data Bit 4 7D5Data Bit 5 8D6Data Bit 6 9D7Data Bit 7 10/ACK Acknowledge

11BUSY Busy 12PE Paper End 13SEL Select 14/AUTOFD Autofeed 15/ERROR Error 16/INIT Initialize 17/SELIN Select In 18GND Signal Ground 19GND Signal Ground 20GND Signal Ground 21GND Signal Ground 22GND Signal Ground 23GND Signal Ground 24GND Signal Ground 25GND Signal Ground ECP 并行口定义 ECP 是 Extended Capabilities Port 的缩写，外观同并行口，是 25 针母插座：

引脚定义 Pin Name Dir Description 1nStrobe Strobe 2data0Address, Data or RLE Data Bit 0 3data1Address, Data or RLE Data Bit 1 4data2Address, Data or RLE Data Bit 2 5data3Address, Data or RLE Data Bit 3 6data4Address, Data or RLE Data Bit 4 7data5Address, Data or RLE Data Bit 5 8data6Address, Data or RLE Data Bit 6 9data7Address, Data or RLE Data Bit 7 10/nAck Acknowledge 11Busy Busy 12PError Paper End 13Select Select 14/nAutoFd Autofeed 15/nFault Error 16/nInit Initialize

各种接口针脚定义-RJ45接RJ48接口串口并口接口(精)

各种接口针脚定义－RJ45接、RJ48接口、串口、并口接口 RJ45接口信号定义，以及网线连接头信号安排 以太网10/100Base-T 接口: Pin Name Description 1 TX+ Tranceive Data+ (发信号+) 2 TX- Tranceive Data- (发信号-) 3 RX+ Receive Data+ (收信号+) 4 n/c Not connected (空脚) 5 n/c Not connected (空脚) 6 RX- Receive Data- (收信号-) 7 n/c Not connected (空脚) 8 n/c Not connected (空脚) 以太网100Base-T4 接口: Pin Name Description 1 TX_D1+ Tranceive Data+ 2 TX_D1- Tranceive Data- 3 RX_D2+ Receive Data+ 4 BI_D3+ Bi-directional Data+ 5 BI_D3- Bi-directional Data- 6 RX_D2- Receive Data- 7 BI_D4+ Bi-directional Data+ 8 BI_D4- Bi-directional Data- 1 white/orange 2 orange/white 3 white/green 4 blue/white 5 white/blue 6 green/white 7 white/brown 8 brown/white

注：RJ45接口采用差分传输方式，tx+、tx-是一对双绞线，拧在一起可以减少干扰。 RJ48接口信号定义 RJ48是用于T1/E1等串行线路的接口。和以太网的RJ45是一样的。 对于接不同的传输，信号定义不一样。 RJ48口呈“凸” 这个形状，线序从左往右为87654321. 例如CT1/PRI-CSU (RJ-48C)信号定义如下 RJ-48C Pin Description 1 Receive Ring 2 Receive Tip 4 Ring 5 Tip 对于T1/E1 Trunk and Digital Voice Port (RJ-4 Pin1 Signal 1 RX + (input) 2 RX - (input) 3 — 4 TX + (output) 5 TX - (output) 6 — 7 — 8 — 串口、并口接口定义 并行口与串行口的区别是交换信息的方式不同，并行口能同时通过8条数据线传输信息，一次传输一个字节；而串行口只能用1条线传输一位数据，每次传输一个字节的一位。并行口由于同时传输更多的信息，速度明显高于串行口，但串行口可以用于比并行口更远距离的数据传输。 1、25针并行口插口的针脚功能： 针脚功能针脚功能 1 选通(STROBE低电平) 10 确认(ACKNLG低电平) 2 数据位0 (DATAO) 11 忙(BUSY) 3 数据位1 (DATA1) 12 却纸(PE) 4 数据位2 (DATA2) 13 选择(SLCT)

图像采集系统的Camera Link标准接口设计

图像采集系统的Camera Link标准接口设计 2009/9/24/10:20 来源：单片机与嵌入式系统作者：李天文赵磊 引言 高速数据采集系统可对相机采集得到的实时图像进行传输、实时处理，同时实现视频采集卡和计算机之间的通信。系统连接相机的接口用的是Camera Link接口，通过Camera Link接口把实时图像高速传输到FPGA图像采集卡中进行数据实时处理，并通过PCI接口实现采集卡和计算机之间的通信。本文主要研究数据采集系统 Cam-era Link接口技术。 Camera Link是专门为数字摄像机的数据传输提出的接口标准，是2000年10月由一些摄像头供应商和图像采集公司联合推出的。Camera Link标准简化了计算机和摄像头之间的连接。本设计选用Dalsa公司的DS-21-02M30相机，该相机支持Camera Link接口。相机数据通过Camera Link接口传输到一块Altera公司的FPGAStratixII中进行处理。在FPGA中进行数据的高速缓存，可以在FPGA 中设计各种图像处理程序对图像进行实时处理。 1 DS-21-02M30相机简介 DS-21-02M30相机可提供高灵敏度的8／10位图像。为了同时获得卓越的分辨率和灰度级，DS-21-02M30相机图像分辨率为1 600×1 200，像素尺寸为7．4 μm×7．4 μm，像素数据输出时钟为40 MHz，最高帧频可达60帧／s。通过设定像素数据格式命令，可以设定像素数据为8位、10位。功耗低于15 W，供电电源电压为12～25 V。 通过异步串口向DS-21-02M30相机发送ASCII码控制命令和诊断命令，可以控制相机输出图像的增益、补偿、帧频、曝光时间、曝光模式和测试图像的输出，还可以对相机进行诊断。串口协议：1位开始位，8位数据位，无奇偶校验位，1位停止位；通信波特率为9 600 bps(相机默认)，通过设定波特率命令可将其设定为19 200 bps、57 600 bps和115 200 bps。 DS-21-02M30相机共有4种曝光模式，可以通过设定曝光模式命令来为相机选择合适的曝光模式。 模式2：内部触发方式(相机的默认曝光模式)。帧频和曝光时间可用相应的命令控制。 模式3：最大曝光时间的外部触发方式。 模式4：外部触发方式。帧频和曝光时间都由外部触发信号控制，即外部触发信号的高电平阶段为曝光时间，外部触发信号的频率为帧频。 模式6：外部触发方式控制帧频，曝光时间可用相应内部命令控制。 DS-21-02M30相机的命令以ASCII码的形式发 送。向相机发送命令时，以回车符作为结束。相机上电后，相机背后的指示灯闪烁，同时通过串口发送“CameraInitial ization in process，Please Wait…OK>”字符串。当收到“OK>”字符串时，表明相机要开始传送图像数据，相机背后的指示灯不再闪烁。当相机收到有效的命令时，会返回“OK>”字符串作为应答；否则，返回“Error x：Error Massage>”字符串作为应答。其中，x

cameralink引脚定义

Camera Link连接器与电缆引脚定义 发布时间：2010-2-10 来源：admin 阅读次数：3234 Camera Link连接器与电缆引脚定义 Channel Link 的高速速率传输使选择连接器和电缆这一环节变得非常重要。必须严格依照 Camera Link 标准中关于对连接器与电缆的引脚定义去设计相机和采集卡的相关连接信号。 1. 连接器 连接器规定的制造商是 3M 公司，其规格化的 3M 26-pin MDR （ Mini D Ribbon ）产品是 Channel Link 的标准连接器（如图3 所示），故而 Camera Link 标准的连接器也选择此型号。 图3 26-pin MDR 连接器 当将这些连接器安装到一个相机或者图像采集卡上时要用到插槽（如图4 所示）。插槽上的连接器固定螺母要与标准的 Camera Link 电缆连接器上的固定螺丝匹配。 图4 26-pin MDR 连接器插槽示意图 2 ．电缆 3M 按照 Camera Link 标准设计了一种专门用于相机和图像采集卡之间的集成电缆。这种双绞屏蔽电缆能够满足高速差分信号应用中的所有严格要求。 3M 电缆产品的通用型号为 14X23 —SZLB —XXX —OLC 。

它的有效长度在 1m 至 10m 之间。另外，它有 2 种外壳可供选择。关于电缆的选型参数说明如图5 所示。本设计中采用的是 14B23 — SZLB — 200 — OLC ，即带固定螺丝的 2m 长电缆。 图5 3M 电缆产品选型说明图 3 ．连接器的引脚分布 表4 给出了安装于相机或者图像采集卡上的 26-pin MDR 连接器的引脚定义。 中级、完整配置模式基本配置模式（含控制与串行通信） 相机端图像采集 卡端 Channel Link 信 号 电缆 相机 端 图像采集卡 端连接器 Channel Link 信 号 1 1 Inner shield Inner shield 1 1 Inner shield 14 14 Inner shield Inner shield 14 14 Inner shield 2 25 Y0- PAIR1- 2 25 X0- 15 12 Y0+ PAIR1+ 15 12 X0+ 3 2 4 Y1- PAIR2- 3 24 X1- 16 11 Y1+ PAIR2+ 16 11 X1+ 4 23 Y2- PAIR3- 4 23 X2- 17 10 Y2+ PAIR3+ 17 10 X2+ 5 22 Yclk- PAIR4- 5 22 Xclk- 18 9 Yclk+ PAIR4+ 18 9 Xclk+ 6 21 Y3- PAIR5- 6 21 X3- 19 8 Y3+ PAIR5+ 19 8 X3+ 7 20 100ΩPAIR6+ 7 20 SerTC+ 20 7 Terminated PAIR6- 20 7 SerTC- 8 19 Z0- PAIR7- 8 19 SerTFG-

各种接口针脚定义

RJ45 接口信号定义，以及网线连接头信号安排 以太网10/100Base-T 接口: Pin Name Description 1 TX+ Tranceive Data+ ( 发信号+) 2 TX- Tranceive Data- ( 发信号-) 3 RX+ Receive Data+ ( 收信号+) 4 n/c Not connected ( 空脚) 5 n/c Not connected ( 空脚) 6 RX- Receive Data- ( 收信号-) 7 n/c Not connected ( 空脚) 8 n/c Not connected ( 空脚) 以太网100Base-T4 接口: Pin Name Description 1 TX_D1+ Tranceive Data+ 2 TX_D1- Tranceive Data- 3 RX_D2+ Receive Data+ 4 BI_D3+ Bi-directional Data+ 5 BI_D3- Bi-directional Data- 6 RX_D2- Receive Data- 7 BI_D4+ Bi-directional Data+ 8 BI_D4- Bi-directional Data- 1white/orange 2orange/white 3white/green 4blue/white 5white/blue 6green/white 7white/brown 8brown/white 注：RJ45 接口采用差分传输方式，tx+ 、tx- 是一对双绞线，拧在一起可以减少干扰 RJ48 接口信号定义 RJ48 是用于T1/E1 等串行线路的接口。和以太网的RJ45 是一样的。对于接不同的传输，信号定义不一样。 RJ48 口呈“凸”这个形状，线序从左往右为87654321. 例如CT1/PRI-CSU (RJ-48C) 信号定义如下 RJ-48C Pin Description 1Receive Ring 2Receive Tip 4 Ring 5 Tip 对于T1/E1 Trunk and Digital Voice Port (RJ-4 Pin1 Sig nal 1RX + (in put) 2RX - (in put) 3— 4TX + (output) 5TX - (output) 6—

Cameralink协议简介

Cameralink简介 CameraLink是一种专门针对机器视觉应用领域的串行通信协议，使用低压差分信号LVDS传输。CameraLink标准在ChannelLink标准的基础上有多加了6对差分信号线，4对用于并行传输相机控制信号，其它2对用于相机和图像采集卡(或其它图像接受处理设备)之间的串行通信。CameraLink标准中，相机信号分为四种: 电源信号、视频数据信号(ChannelLink标准)、相机控制信号、串行通信信号、视频数据信号。 视频数据信号 视频数据信号部分是CameraLink的核心，该部分为其实就是Channel Link协议。主要包括5对差分信号，即X0-~X0+、X1-~X1+、X2-~X2+、X3-~X3+、Xclk-~Xclk+；视频部分发送端将28位的数据信号和1个时钟信号，按7:1的比例将数据转换成5对差分信号，接收端使用Channel Link芯片（如Channel Link转TTL/CMOS 的芯片DS90CR288A）将5对差分信号转换成28位的数据信号和1个时钟信号。28位的数据信号包括4位视频控制信号和24位图像数据信号。 4位视频控制信号 FVAL:帧同步信号。当FVAL为高时表示相机正输出一帧有效数据 LVAL:行同步信号。当FVAL为高时，LVAL为高表示相机正输出一有效的行数据。行消隐期的长短由具体的相机和工作状态有关。

DVAL:数据有效信号。当FVAL为高并且LVAL为高时，DVAL为高表示相机正输出有效的数据,该信号可用可不用，也可以作为数据传输中的校验位。 CLOCK:这一信号为图像的像素时钟信号，在行有效期内像素时钟的上升沿图像数据稳定。值得说明的是，CLOCK信号单独采用一对LVDS信号传输，不管相机是否处于工作状态，CLOCK信号应该始终有效，它是ChannelLink芯片的输入时钟，是ChannelLink芯片之所以能在4对信号线中传输28位数据，就是因为对CLOCK信号7倍频的结果。 相机控制信号 CameraLink标准定义了4对LVDS线缆用来实现相机控制，它们被定义为相机的输入信号和图像采集卡的输出信号。一般情况是这些信号命名为: CameraControl1(CC1) CameraControl2(CC2) CameraControl3(CC3) CameraControl4(CC4) 串行通信信号 CameraLink标准定义了2对LVDS线缆用来实现相机与图像采集卡之间的异步串行通信控制。相机和图像采集卡至少应该支持9600的波特率。这两个串行信号是相机: SerTFG(相机串行输出端至图像采集卡串行输入端) SerTC(图像采集卡串行输出端至相机串行输入端) 其通信格式为:1位起始位、8位数据位、1位停止位、无奇偶校验位和握手位。 相机电源并不是由CameraLink连接器提供的，而是通过一个单独的连接器提供。 视频传输模式 由于单个Camera Link芯片只有28位数据可用，有些相机为了提高传输数据的效率，需要几个Camera Link芯片。按使用要求不同，视频传输模式分为三种

各种接口引脚的定义-USB接口,主板接口,PS2接口

USB的全称是Universal Serial Bus，USB支持热插拔，即插即用的优点，所以USB接口已经成为MP3的最主要的接口方式。USB有两个规范，即USB1.1和USB2.0。 USB1.1是目前较为普遍的USB规范，其高速方式的传输速率为12Mbps，低速方式的传输速率为1.5Mbps（b是Bit的意思），1MB/s（兆字节/秒）=8MBPS （兆位/秒），12Mbps=1.5MB/s。目前，大部分MP3为此类接口类型。 USB2.0规范是由USB1.1规范演变而来的。它的传输速率达到了480Mbps，折算为MB为60MB/s，足以满足大多数外设的速率要求。USB 2.0中的“增强主机控制器接口”（EHCI）定义了一个与USB 1.1相兼容的架构。它可以用USB 2.0的驱动程序驱动USB 1.1设备。也就是说，所有支持USB 1.1的设备都可以直接在USB 2.0的接口上使用而不必担心兼容性问题，而且像USB 线、插头等等附件也都可以直接使用。 使用USB为打印机应用带来的变化则是速度的大幅度提升，USB接口提供了12Mbps的连接速度，相比并口速度提高达到10倍以上，在这个速度之下打印文件传输时间大大缩减。USB 2.0标准进一步将接口速度提高到480Mbps，是普通USB速度的20倍，更大幅度降低了打印文件的传输时间。 右图为USB接口连线定义 USB是一种常用的pc接口,他只有4根线，两根电源两根信号,如下图.故信号是串行传输的,usb接口也称为串行口，usb2.0的速度可以达到480Mbps。可以满足各种工业和民用需要.USB接口的输出电压和电流是： +5V 500mA 实际上有误差，最大不能超过+/-0.2V 也就是4.8-5.2V 。usb接口的4根线一般是下面这样分配的，需要注意的是千万不要把正负极弄反了，否则会烧掉usb设备或者电脑的南桥芯片：黑线：gnd 红线：vcc 绿线：data+ 白线：data- -------------------------------------------------------- USB接口定义颜色 一般的排列方式是：红白绿黑从左到右 定义： 红色－USB电源：标有－VCC、Power、5V、5VSB字样 绿色－USB数据线：（正）－DATA+、USBD+、PD+、USBDT+ 白色－USB数据线：（负）－DATA-、USBD-、PD-、USBDT+ 黑色－地线： GND、Ground

CameraLink电缆规范

Camera Link Cable Assembly Camera Link ? Cable Assembly ￥s t CCD Camera Frame Grabber [ Camera Link Standard ] a Machine Vision ?p Camera Link ? CCD Camera ?M Frame Grabber a ¤, ¨b2000|10¤G D n F, §s o e,-C s o ,￥ Camera Link cable §i e t T, CCD Camera ± T. [ Camera Link Cable 14B26 series à ] z Cable , 3A t e. Connector ¨ 1.27mm pitch MDR z SMT type 3s, ￥i T t, ¨T L [ Camera Link Cable a ] s TRM-CL-3M 3| TRM-CL-5M 5| TRM-CL-10M 10|

High Speed MDR Digital Data 3M calls this product a High Speed MDR Digital Data Transmission System, not just a cable assembly, because it includes a complete solution, from board to board, for 3M worked closely with the Camera Link consortium companies and with chip maker National Semiconductor to align this system with the emerging standards for the digital camera-to-frame-grabber interconnect. The result was the first product to meet the Industrial Camera Link Standards. The High Speed MDR Digital Data Transmission System’s robust ribbon leaf contact and positive locking shell provide the reliable, durable connection for industrial use. It’s dual shielding, with each pair shielded as well as the overall cable, offers the optimal performance needed for high speed transmission. The product is appropriate for use in any digital Semiconductors and electronics manufacture ? Pharmaceutical and medical supply manufacture 3M’s MDR cable and connector system has been widely used in LVDS technology and has proven itself in rugged performance and reliability in a wide array of applications. ? 26 position connector, with 11 shielded pairs and ? Positive locking thumbscrew shells for a reliable Shielding for each pair as well as for overall cable Available in cable lengths from one to 10 meters