Introduction

Over the weekend 24-25 June, 2000, I performed a series of stability tests with the RTTF. The aim was to establish how the TTF behaves with temperature and with slew angle.

The historical context for these tests are as follows. We have been suffering from erratic TTF behaviour over the past year and unable to establish well defined relationships between the spectral response of the TTF and external factors. The present tests were performed after extensive electronic debugging by EJP who established that a number of op-amps in the old CS100 (c. 1985) had corroded. But note that these tests were performed ahead of the broad-narrow upgrade in order that we can establish the ``normal'' response of the TTF.

TAURUS/TTF was mounted at f/8 Cass while 2dF was mounted at the Prime Focus. The ccd3_cass line was used with the ccd1 controller. For the tests, a single isolated Ne line (659 nm) was monitored with the R0 filter. At commencement, TAURUS/TTF had been powered up and running for more than 12 hours, with the TAURUS chamber continually flushed with dry N. The dome and TAURUS chamber were monitored for temperature changes independently. The following system constants were used:

$$\left( \begin{array}{ccc} X_c & X_f & X_s \\ Y_c & Y_f & Y... ...9 & 0 \\ 0.1 & 7.3 & 0 \\ 5 & 0 & ... \end{array} \right)$$ (1)

It was clear from the outset that the electronic upgrades had improved the system performance. In particular, for the first time in two years, we were able to operate at the original (and optimal) CS100 control settings for the TTF, i.e. gain = 8 (SMS=3) and time constant = 256 ms (SMS=8).

 

Figure 1: The top plot shows temperature change within the TAURUS chamber (points with fits) and dome (curves) during the course of the weekend. The dashed line is 24.6.00 and the dotted line is 25.6.00. The bottom plot shows the drift of the TTF response to a Ne line (659 nm) due to various external effects -see text. The fit equations are also included. The data sets A and B were obtained on 24.6.00; data set C was obtained on 25.6.00. Set A shows variations due to ambient temperature and an improperly tuned CS100 which produced an asymmetric line profile. B and C show well behaved temperature drift for the initial sets of points. The points which depart from the fitted slopes result from slewing the telescope.