AOT 1 scans the entire SWS wavelength range at a reduced spectral resolution. The scans are effectively continuous, using small and frequent steps, and on a timescale much shorter than the selected reset interval of the detector pre-amplifiers. The integration in time translates into a smoothing of the spectrum, which prevents aliasing problems due to insufficient sampling.

Table 4.4 shows the reset intervals, dwell times, stepsize and number of up-down scans for AOT 1. The reset interval is the time between detector resets, the dwell time is the time between the grating moving and the stepsize, in LVDT, is the amount by which the grating moves every dwell time. An LVDT is an internal unit of measurement of the grating position. A movement of 1 LVDT corresponds to approximately 1/8 of a resolution element. The pipeline resolution degradation is discussed in section 4.2.4.

**Table 4.4:** AOT 1 parameters
Speed	Duration	Reset	Dwell	Stepsize	Number of	Pipeline
		interval	time	LVDT	up-down	Resolution
	sec	sec	sec		scans	degradation
1	1172	1	1/8	4	1	7
2	1944	2	1/8	2	1	7
3	3846	2	1/8	1	1	4
4	6570	2	1/4	1	1	2

For an AOT 1 measurement, where the grating moves during a reset interval, the grating position in the SWSPGPOS field of the SPD is set to the start position of the grating during the reset interval. This contrasts with the wavelength in the SWSPWAVE field of the SPD which is the effective wavelength, i.e. calculated from the middle valid sample. For example, for a one second reset interval, the grating position is that of the first sample. As samples are thrown away due to the reset pulse (see section 5.6, operation described in section 8.2.2), the wavelength is that from sample (24 - Cal-G file 3) / 2 (assuming no data contains glitches, is all within limits etc).

4.2.2 AOT SWS01 sample time-line

1. Acquisition - point aperture 1 to target
2. Dark current measurement (20 sec)
3. Single scan up, bands 1A, 1B, 3A (70 sec)
4. Single scan down, bands 1A, 1B, 3A (70 sec)
5. Dark current (20 sec)
6. Internal photometric calibration (40 sec)
7. Switch to aperture 2 (10 sec)
8. Dark current measurement (20 sec)
9. Single scan up, bands 1D, 1E, 2A, 2B, 3C, 3D (170 sec)
10. Single scan down, bands 1D, 1E, 2A, 2B, 3C, 3D (170 sec)
11. Dark current measurement (20 sec)
12. Switch to aperture 3 (10 sec)
13. Dark current (20 sec)
14. Single scan up, bands 2C, 3E, 4 (150 sec)
15. Single scan down, bands 2C, 3E, 4 (150 sec)
16. Dark current (20 sec)
17. Internal photometric calibration (40 sec)

After resetting the instrument, the AOT is composed of three SS0005 ICSs, one for each aperture. The sequence is aperture 1, 2 then 3. Each ICS 5 stars with a dark, then makes the measurement. An ICS 6 (dark) is used to make another dark current measurement after each ICS 5. A photometric check is made between the aperture 1 ICS 6 and aperture 2 ICS 5. From May 1996 an extra photometric check was added at the end of the AOT.

4.2.3 Example AOT 1 data

Figures 4.2 and 4.3 show ERD data for detectors 1, 13, 25 and 37 (bands 1, 2, 3 & 4) during an example AOT 1 observation. The signal is shown in bit values against time (ITK) and only four samples are shown out of the 24 per second for clarity. The periods when apertures 1, 2 & 3 are used, times of dark current measurements and photometric checks are indicated.

Figures 4.4 and 4.5 show the wavelengths seen by the middle detector of each band during the AOT 1 observation along with the scanner position (LVDT). The nature of up-down scans can be seen in the plots, ref section 4.6.1.

Figure 4.6 shows SPD data, in $\mu V/sec$ , for detector 1 of each band against time. Figure 4.7 shows the AAR for the same observation, now as flux (Jy) against wavelength. The spread of flux between different detectors in the same band can be seen in the band 4 data. Figure 4.8 shows what can be achieved after further processing of such data in either ISAP or IA.

**Figure 4.3:** ERD band 3 & 4 data for AOT 1
$\begin{figure} \begin{tabular}{c} \centerline{\epsfig{file={sws01_erd2.eps},width=15.0cm}}\\ \end{tabular}\end{figure}$

**Figure 4.4:** Wavelengths seen by bands 1 & 2 during an AOT 1
$\begin{figure} \begin{tabular}{c} \centerline{\epsfig{file={sws01_wave1.eps},width=15.0cm}}\\ \end{tabular}\end{figure}$

**Figure 4.5:** Wavelengths seen by bands 3 & 4 during an AOT 1
$\begin{figure} \begin{tabular}{c} \centerline{\epsfig{file={sws01_wave2.eps},width=15.0cm}}\\ \end{tabular}\end{figure}$

**Figure 4.6:** SPD data from the first detector of each band during an AOT 1. The signals have been arbitrarily scaled.
$\begin{figure} \begin{tabular}{c} \centerline{\epsfig{file={sws01_spd.eps},width=15.0cm}}\\ \end{tabular}\end{figure}$

**Figure 4.7:** AAR data from all detectors against wavelength
$\begin{figure} \begin{tabular}{c} \centerline{\epsfig{file={sws01_aar.eps},width=15.0cm}}\\ \end{tabular}\end{figure}$

**Figure 4.8:** An example of what can be achieved by the further processing of OLP data
$\begin{figure} \begin{tabular}{c} \centerline{\epsfig{file={sws01_ia.eps},width=15.0cm}}\\ \end{tabular}\end{figure}$

4.2.4 AOT 1 Spectral Resolution

The pipeline resolution degradation, given in table 4.4 is the amount by which the resolution of the spectrometer is degraded after the data has been processed by the pipeline. The pipeline processes data for blocks of length of the reset interval (1 or 2 seconds), during which the grating will have moved several times. The pipeline effectively smears out the resolution. Both theory and observations can be used to characterise the amount of degradation.

4.2.4.1 Theory

$\begin{displaymath} {Resolution\ degradation} = {{reset\ time} \over {dwell\ time} } \times { stepsize \over {steps\ per\ res.\ element}} \end{displaymath}$

For a source filling the slit, the number of steps per resolution element is about 4.5, except for bands 3E and 4 , where it is about 7.0.

Hence the effective resolution of AOT 1 data processed by the pipeline is between a factor of 2 and 7 worse than that expected from an AOT 2 pipeline product.

4.2.4.2 Observations

Preliminary measurements of the resolution of AOT 01 scans at all four speeds were carried out on observation of the PNe NGC 6543 and a speed 4 observations was carried out on the PNe NGC 7027. The assumption was made that the lines from these objects would not be resolved by the SWS01 scans. NGC 7027 contained by far the most lines of the two objects.

**Table 4.5:** NGC 7027 speed 4 AOT 1 resolution
Wavelength	Resolution	Wavelength	Resolution
$\mu m$	$\lambda/FWHM$	$\mu m$	$\lambda/FWHM$
2.407	1388	7.902	1123
2.626	1216	9.666	1550
3.092	1458	10.51	1697
3.741	1143	12.81	1149
4.052	1236	13.10	1186
4.487	1366	13.52	1176
4.530	1347	14.32	1357
4.654	1422	15.55	1411
5.610	818	18.71	1663
6.986	1144	24.32	1004
7.319	1149	25.89	1049
7.460	1031	34.81	1619
7.653	1227	36.02	1459

**Table 4.6:** NGC 6543 all speeds AOT 1 resolution
Wavelength	Speed 4	Speed 3	Speed 2	Speed 1
$\mu m$	Resolution	Resolution	Resolution	Resolution
	$\lambda/FWHM$	$\lambda/FWHM$	$\lambda/FWHM$	$\lambda/FWHM$
7.46	983	0	0	0
8.99	1332	771	388	393
10.51	1670	0	463	538
15.55	1344	834	0	526
18.71	1718	1060	529	585
33.48	1323	677	0	560
36.00	1426	886	0	603

The measurements on NGC 7027 give more reliable results than those on NGC 6543. No systematic difference between the speed 4 resolutions of the two objects could be found.

4.2.4.3 AOT 1 spectral resolution conclusions

The speed 4 resolutions vary between approximately 800 and 1700, increasing with wavelength per order. This is approximately half of the AOT 2 resolution, matching expectations. The FWHM remains nearly constant.

We can compare the resolutions of the different speeds using the NGC 6543 observations. The results, which match expectations, are: