Next: 3.7 Dark signal subtraction Up: 3. Derive SPD level Previous: 3.5 Extraction of signals:

3.6 Reset interval correction (PHT-P, PHT-C)

synopsis: residual systematic error which depends on the difference in reset interval length selected for the sky measurement(s) and FCS measurement. In normal cases the residual error is better than 20%.

limitations and applicability:
PHT-S data are not corrected.

description:
The signal dependence on reset interval is removed by correcting all signals to one reference reset interval of 1/4 s. This correction reduces the systematic error between measurements of a given AOT.

purpose correction:
A given photocurrent yields different signals depending on the reset interval selected. This variation is fixed and has been determined in orbit from calibration measurements. To remove the effect all signals are corrected to the 1/4 s reset interval according to the relation:

$\begin{displaymath} s(RI=\frac{1}{4}~sec,~dr=1) = A_0(RI,~dr) + A_1(RI,~dr) * s(RI,~dr) ~~~~~V/s, \end{displaymath}$

(1)

where s(RI, dr) is the uncorrected signal taken with reset interval length RI and data reduction dr. The correction parameters A₀ and A₁ depend also on the detector. Since all signals of all measurements are converted, the correction becomes only relevant when comparing signals of measurements obtained with different reset intervals such as sky and FCS measurement.

uncertainty/noise introduced:
The consistency between signals of measurements in the same AOT is improved, especially for the comparison between sky and FCS measurement. The correction parameters A₀ and A₁ have uncertainties. The systematic uncertainty in the derived flux densities depends on the reset interval difference between the FCS measurement and sky measurement which can vary from 0 (=no uncertainty) to about 2⁵ (maximum uncertainty introduced by this correction). The latter could happen in case a multi-filter or multi-aperture AOT is used where the dynamic range between the first and last sky measurement is high. If for a given reset interval the dynamic range of the ramp is below 1% (0.01-0.02 V) the residual uncertainty becomes larger than 20%.

auxillary data:
Cal-G files: PxxRESETI, where xx = P1, P2, P3, C1, and C2, referring to the various detector subsystems.

keyword(s):
None

Next: 3.7 Dark signal subtraction Up: 3. Derive SPD level Previous: 3.5 Extraction of signals:

ISOPHOT Error Budgets: Derive_SPD Processing Steps, Version 1.0, SAI/98-091/Dc