Subsystem PHT-P was a multi-filter, multi-aperture photopolarimeter with single element detectors and wide beam (up to 3arcmin) capability for faint extended sources. The covered wavelength range was $\approx$ 3-130 ${\mu}$ m. The PHT-P detectors were each located inside an integrating cavity and had a dimension of 1 ${\times}$ 1 ${\times}$ 1mm $^{3}$ . This subsystem was designed for sensitive, high precision photometry and polarimetry, using three different detector types (in parenthesis the detector material is given):

**Table 2.2:** PHT-P filter characteristics. $\lambda_{ref}$ is the PHT reference wavelength (Moneti, Metcalfe & Schulz 1997, [39]), to be used for colour corrected monochromatic fluxes, $\lambda_c$ is the central wavelength, $\Delta \lambda$ is the width and $R_{mean}$ the average relative system response derived from the bandpasses (see Section 2.9 and Appendix A.2). $d_{Airy}$ is the diameter of the Airy disc taken at $\lambda_c$ . The minimum aperture was the smallest one recommended; it was the minimum aperture which completely covered the Airy disc.
No.	Filter	$\lambda_{ref}$	$\lambda_c$	$\Delta \lambda$	$R_{mean}$	$d_{Airy}$	min. aper.	scientific objective
#		[ $\mu$ m]	[ $\mu$ m]	[ $\mu$ m]		[]	[]
	P1
1	P3.29	3.3	3.30	0.22	0.10	2.8	5	PAH
2	P3.6	3.6	3.59	0.99	0.14	3.0	5	cosmological gap,
								common to ISOCAM
3	P4.85	4.8	4.86	1.55	0.16	4.1	5	continuum to #1, #4 and #5
4	P7.3	7.3	7.41	3.38	0.28	6.2	7.6	6.2, 7.7, 8.6 $\mu$ m PAH complex
5	P7.7	7.7	7.66	0.82	0.25	6.4	7.6	PAH
6	P10	10.0	10.00	1.80	0.35	8.4	10	silicate feature
7	P11.3	11.3	11.36	0.81	0.29	9.5	10	PAH
8	P11.5	12.0	11.88	6.53	0.48	10.0	10	IRAS 12 $\mu$ m band,
								common to ISOCAM
9	P12.8	12.8	12.82	2.31	0.52	10.8	10	continuum to #7
10	P16	15.0	15.16	2.84	0.35	12.7	13.8	general purpose
	P2
11	P20	20.0	21.03	9.03	0.32	17.7	18	close to standard Q band
12	P25	25.0	23.80	9.12	0.38	20.0	23	IRAS 25 $\mu$ m band
	P3
13	P60	60.0	60.85	25.89	0.11	50.3	52	IRAS 60 $\mu$ m band
14	P100	100.0	102.44	39.55	0.31	83.9	79	IRAS 100 $\mu$ m band

Any combination of apertures and filters was allowed. Polarisers were only used with the P2 25 $\mu$ m filter and the

aperture. At long wavelengths the Airy disc is significantly larger than the smallest aperture available. The diameter of the Airy disc is $\rm d_{Airy}['']$ =0.84 $\cdot$ $\lambda$ [ $\mu$ m]. The selection of apertures much smaller than $\rm d_{Airy}$ was not recommended (see Table 2.2). Section 2.9 gives more details about the calculation of $\lambda_{c}$ , $\Delta \lambda$ and $R_{mean}$ .

**Table 2.3:** *Intensity fraction $\rm f_{PSF}$ of a point source entering the apertures of PHT-P. A theoretical monochromatic point spread function centred in the aperture has been adopted.*
	PHT-P apertures
filter	5	7.6	10	13.8	18	23	20 ${\times}32''$	52	79	99	120	127 ${\times}127''$	180
P1
3.3	0.85	0.88	0.91	0.92	0.95	0.96	0.96	0.99	0.99	0.99	1.00	1.00	1.00
3.6	0.81	0.88	0.91	0.92	0.94	0.95	0.96	0.98	0.99	0.99	1.00	1.00	1.00
4.8	0.69	0.86	0.88	0.90	0.92	0.93	0.94	0.98	0.99	0.99	0.99	0.99	1.00
7.3	0.64	0.76	0.85	0.87	0.89	0.91	0.92	0.96	0.97	0.98	0.99	0.99	0.99
7.7	0.64	0.73	0.85	0.87	0.88	0.90	0.91	0.96	0.97	0.98	0.99	0.99	0.99
10	0.54	0.66	0.73	0.84	0.87	0.88	0.89	0.94	0.96	0.97	0.98	0.98	0.99
11.3	0.48	0.66	0.68	0.79	0.86	0.87	0.89	0.93	0.96	0.97	0.97	0.98	0.99
12	0.48	0.65	0.70	0.78	0.86	0.87	0.88	0.93	0.95	0.97	0.97	0.98	0.99
12.8	0.42	0.65	0.66	0.73	0.86	0.87	0.88	0.92	0.95	0.96	0.97	0.98	0.98
15	0.33	0.61	0.66	0.67	0.80	0.86	0.85	0.92	0.94	0.95	0.96	0.97	0.98
P2
20	0.22	0.47	0.59	0.65	0.69	0.75	0.79	0.89	0.92	0.93	0.95	0.95	0.97
25	0.17	0.41	0.54	0.63	0.66	0.70	0.76	0.88	0.91	0.93	0.94	0.95	0.96
P3
60	0.04	0.09	0.15	0.23	0.36	0.46	0.51	0.68	0.81	0.86	0.87	0.88	0.90
100	0.01	0.04	0.05	0.09	0.15	0.22	0.27	0.59	0.66	0.70	0.77	0.83	0.87

The precision of the ISO pointing had implications on the achieved photometric accuracy. When observing with the smallest PHT-P apertures (5.0

and 7.6

), the source might have been observed strongly off-centre, see `ISO Handbook, Vol. I: ISO - Mission & Satellite Overview', [20] for details about the ISO pointing accuracy.

2.6.2 PHT-C

PHT C100 was a $3{\times}3$ array of Ge:Ga with 0.7 ${\times}$ 0.7 ${\times}$ 1mm $^{3}$ elements. Increased photon absorption was achieved by total reflection of a 30 $^{\circ}$ wedged pixel surface and an integrating cavity. The telescope beam was fed into these cavities by $1.9{\times}1.9~$ mm $^{2}$ anti-reflection coated germanium fabry lenses mounted with 100 $\mu$ m spacing which resulted in an optical fill factor of 93%. The effective size of the pixels on the sky was 43.5 ${\times}$ 43.5arcsec $^{2}$ , the distance between the pixel centers (`pitch') was 46.0arcsec.

Throughout this manual the individual C100 pixels/detectors are labelled as in Table 2.4 where the +Z direction is upwards, and +Y is to the left. This corresponds to the projection of the array on the sky. The numbers (1...9) in Table 2.4 refer to the labelling and array counting in ERD, SPD and AAR products.

PHT C200 consisted of 4 pixels arranged in a $2{\times}2$ matrix. The telescope beam was concentrated on the four detector pixels by anti-reflection coated germanium Fabry lenses of 3.9 ${\times}$ 3.9mm $^{2}$ . The detector crystals were mounted in an integrating cavity and had prismatic shape to increase photon absorption. The sizes of the detector pixels themselves were approximately 1mm

. The effective size of the pixels on the sky was 89.4 ${\times}$ 89.4arcsec $^{2}$ , the distance between the pixel centers was 92.0arcsec.

The pixels were stressed by individual screws (Wolf, Grözinger & Lemke 1995, [58]); the stress was maximized in order to give a cut-off wavelength of 240 $\mu$ m. The labelling of the individual C200 pixels is given in Tables 2.5 and 2.6 for the ERD and SPD/AAR convention.

There were 6 filters for C100 and 5 for C200 available. Table 2.7 gives a list of these filters including the reference and central wavelength, the widths, their resolution and the transmission. Section 2.9 gives more details about the calculation of $\lambda_c$ , $\Delta \lambda$ and $R_{mean}$ .

**Table 2.7:** PHT-C filter characteristics. $\lambda_{ref}$ is the PHT reference wavelength (Moneti, Metcalfe & Schulz 1997, [39]), to be used for colour corrected monochromatic fluxes, $\lambda_c$ is the central wavelength, $\Delta \lambda$ is the width and $R_{mean}$ the average relative system response derived from the bandpasses (see Section 2.9). $d_{\rm Airy}$ is the diameter of the Airy disc at $\lambda_c$ .
Filter	$\lambda_{ref}$	$\lambda_c$	$\Delta \lambda$	$R_{mean}$	$d_{\rm Airy}$
	[ $\mu$ m]	[ $\mu$ m]	[ $\mu$ m]		[]
C100
C50	65	68.7	60.8	0.04	42
C60	60	61.8	24.6	0.13	50
C70	80	80.7	48.4	0.12	59
C90	90	95.2	56.4	0.30	76
C100	100	102.6	47.1	0.27	84
C105	105	107.2	38.4	0.24	88
C200
C120	120	118.7	49.5	0.13	101
C135	150	155.1	81.2	0.26	113
C160	170	174.3	89.9	0.43	134
C180	180	181.0	68.8	0.33	151
C200	200	202.1	56.9	0.22	168

**Table 2.8:** *Intensity fraction $f_{PSF}$ of a point source falling on one pixel of the C100 or C200 detector, respectively. A theoretical point spread function centred on the pixel has been adopted.*
filter	$f_{PSF}$	filter	$f_{PSF}$
C100		C200
C_50	0.656	C_120	0.678
C_60	0.667	C_135	0.641
C_70	0.629	C_160	0.620
C_90	0.586	C_180	0.609
C_100	0.558	C_200	0.568
C_105	0.540

2.6.3 PHT-S

PHT-S consisted of a dual grating spectrometer with resolving power of order 90 in two wavelength bands. Band SS covered the wavelength range 2.5-4.9 $\mu$ m and band SL covered the range 5.8-11.6 $\mu$ m. Each spectrometer used a linear array of 64 element Si:Ga detectors with dimensions of $0.31{\times}0.37{\times}1.80$ mm $^{3}$ per element. The arrays, hence the dispersion direction, were oriented in the spacecraft Z-direction. PHT-S had one square entrance aperture with dimensions $24''{\times}24''$ ; this aperture was imaged onto each detector pixel.

In dispersion direction this resulted in a triangular spectral bandpass with a spectral range (Full Width at Half Maximum) for a single detector of 42.4nm (=3500 kms $^{-1}$ ) for PHT-SS and 96.6nm (=3200 kms $^{-1}$ ) for PHT-SL. The resolution was about 85 for PHT-SS and about 95 for PHT-SL, respectively. The spectra could be fully sampled at half of the resolution, i.e. with 89nm (PHT-SS) and 189.8nm (PHT-SL). PHT-S could be used with the chopper.

Both gratings were operated in first order. The wavelength scale was established in-orbit against celestial sources which emit narrow lines by fitting a 2nd order polynomial through the measured line centers.

2.6 Detector Subsystems

2.6.1 PHT-P

2.6.2 PHT-C

2.6.3 PHT-S