From erik Wed Jan 20 13:01:15 1999 Subject: CUO_56.txt, NBP dichroic To: erik (Erik Hoeg) Date: Wed, 20 Jan 1999 13:01:15 +0100 (MET) X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 6203 Status: RO Narrow-band photometer with dichroic filter =========================================== E. Hoeg, C. Fabricius, J. Knude, V.V. Makarov 20 Jan. 1999 SAG_CUO_56 ABSTRACT: Design of the narrow-band photometer (NBP) with dichroic filter is described. The resulting improved precision is, e.g., 0.009 mag compared with 0.013 at V=18.0 for the y-band. In addition, about four red bands may be introduced. Dichroic filter --------------- The performance of the NBP can be significantly improved by increasing the available area for filters from the present 1*1 sq.deg up to 3*1 sq.deg. This is possible by means of a dichroic filter in front of the present focal plane. The incidence angle should perhaps be 25 deg as in the Tycho experiment (see SP-1200, Volume 2, Sect. 2.6 and Tables 2.3-4) in order to minimize polarization effect. The astigmatism introduced by the transmittance through glass should be quantitatively evaluated. The astigmatic elongation should be oriented perpendicular to the scan, and the length of the sample should be optimized accordingly. The transmission loss will be about 10 per cent according to the Tycho experience, which is small enough to neglect in the following discussion since it corresponds to the loss of 0.10 mag. The separation wavelength might be 700 nm. The field with the longer wavelengths is used for the RVS and some TBD red bands. The field with shorter wavelength has an area of 2*1 deg and could be used for the Stromvil bands and TBD other bands. The precision in Stroemgren y from a 5 s integration instead of 2.2 s is 0.009 mag instead of 0.013 at V=18.0, according to Table 3 of CUO_55 (Columns 4 and 5) where only photon noise from the average of 100 observations is taken into account. The precision in u from 2 integration times of 6 s is 0.019 mag for a G2V star. The integration times imply that a sample of 5 pixels across scan is always required, instead of sometimes 4 pixels in CUO_55. The impact on crowding is negligible. The questions of crowding and of observation with ground-based telescope of areas with high star density and high surface brightness have been discussed in CUO_55 and apply to this design without any changes. We prefer to increase the integration time on each band compared with those given in Table 2 of CUO_55 so that the data rate with the 3 times larger field will not be 3 times higher. The estimated data rate increases from 264 kbits/s in CUO_55 to 332+162=494 kbits/s for 4 Stromvil bands plus 4 red bands. Data rates for Stromvil photometry ---------------------------------- The MMS final report Figure 6.7/2 assumes: 75 stars/s entering the SSM, field height 1.0 deg, Vlim=17. 64 bits per band, incl. SSM --> 7 bands *64=448 bits/star --> 33.6 kbits/s We assume for 7 band Stromvil photometry: Vlim=19.5 --> 500 mio stars on the sky --> 12 000 star/sq.deg --> 400 stars/s. 64 bits for SSM1 and 16*16 bits per band for the patch, compression ratio 3.0 --> 85 bits/band, SSM1 + 7+2 bands/star --> 829 bits/star (2 bands are repeated) 400 stars/s or 500 million stars in total --> 332 kbits/s for Vlim=19.5 and 7 Stromvil bands. =============== Data from the red part will be added, but the bands are yet undefined. If we assume 4 bands, each with 7 seconds integration we obtain SSM2 + 4 bands/star --> 404 bits/star, 400 stars/s TBC --> 162 kbits/s for Vlim=19.5 and 4 red bands. =============== The rate could perhaps be decreased by use of shorter patches. The length of 16 samples/patch was chosen for the reason mentionend above, of finding all disturbing stars within 3 arcsec. But this requirement may be relaxed in most of the sky where the star density is low. The patches could therefore be shorter, perhaps 10 samples. But probably a few bands, e.g. u, y and S should always obtain 16 samples/patch. ------------------------------------------------------------------ Table 1 The assumed CCD arrangement in the three fields (a) the reflected light from the dichroic filter <700 nm for Stromvil photometry, (b) preceding field of transmitted light >700 nm (c) following field of transmitted light, after the radial velocity spectrometer. (a) reflected light: Vw 1.0s detection in SSM1, flux in 400-700 nm = Vw 1s =2.4mm spacing of the CCDs y 5.0s flux, at 547 nm, FWHM=23 nm, Tmax=0.80 1s b 5.0s flux, at 467 nm, FWHM=18 nm, Tmax=0.70 1s Z 5.0s flux, at 516 nm, FWHM=21 nm, Tmax=0.80 1s S 5.0s flux, at 656 nm, FWHM=20 nm, Tmax=0.80 1s v 6.0s flux, at 411 nm, FWHM=19 nm, Tmax=0.60 1s u 6.0s flux, at 350 nm, FWHM=30 nm, Tmax=0.40 1s u 6.0s flux, at 350 nm, FWHM=30 nm, Tmax=0.40 1s P 6.0s flux, at 374 nm, FWHM=26 nm, Tmax=0.42 1s P 6.0s flux, at 374 nm, FWHM=26 nm, Tmax=0.42 60.0s total= 2.00 degree as nominally available Transmitted light, preceding and followings fields: (b) PFOV: Iw 1.0s detection in SSM2, flux in 700-1000 nm = Iw 1s red bands to be defined 15.0s =total (c) FFOV: red bands to be defined 15.0s =total 30.0s =total of PFOV + FFOV Notes: (1) Two Spectro Sky Mappers (SSM1 and SSM2) are assumed each covering one wavelength range and with sensitivity in the wide bands Vw and Iw, respectively. This will best ensure that only the stars bright enough to obtain a useful accuracy in photometry and radial velocities are observed thus avoiding useless data transmission. An integration time of 1.0s is assumed for the SSMs in order to achieve a limiting magnitude for detection of V=19 to 20 (cf. Column 2, Table 3 of CUO_55) as required for photometry. A limit of V=17 mag has been required for radial velocities, but this limit should better be given as Iw=TBD since the radial velocities are measured in the red part of the spectrum. (2) The final distribution of integration times on the bands must result from further astrophysical discussions. We should probably optimize the Stromvil bands for the F and G stars because these stars represent the complete range of metallicities and ages. ------------------------------------------------------------------