Performance of the GBT at 2.3 GHz (13 cm)

Measurements done April 19, 2004.
June 22, 2004;
Y. Kovalev, F. Ghigo

Introduction

GBT gain curve measurements in the 13 cm band (S-band) were done April 19, 2004. The gain versus elevation was measured by observing several strong calibration sources over a wide range of elevations using the peak procedure. The subreflector focus was adjusted about once per hour. The observations were done at a frequency of 2240 MHz with a bandwidth of 80 MHz. The frequency was chosen to match the VLBA S-band. A hybrid was used to produce LCP and RCP output from linearly polarized feed horns. Weather conditions were good (partly cloudy). The active surface (zero offset: ON, FEM model: OFF) and dynamical corrections were in use for this run.

Summary

Channel Maximum Gain Tsys (high elevation) Tcal
LCP 1.91 K/Jy 19.5 K 2.10 K
RCP 1.75 K/Jy 19.5 K 2.15 K

Observing Procedure

Observing of the calibration sources was done using the "peak" procedure and fitting gaussians with "GFM" (see description of this procedure here).

System Temperature Measurements

The Tsys values as calculated by "GFM" from each "peak" procedure, are plotted in Figure 1. Some of the system temperature measurements were corrupted by RFI. They are not used in our calculations.

Figure 2: system temperature vs elevation

A simple model for the atmosphere was fitted by least-squares to the data:
Tsys = T0 + Tatm (1 - exp(-Tau*A)),
in which airmass A = 1/sin(Elev), T0 is extrapolated noise temperature for airmass = 0 (sum of the receiver temperature, antenna and groud contribution, etc.), tau is the atmospheric opacity.

The fitted parameters are as follows (assuming Tatm = 260 K):
Parameter Fit to LCP data Fit to RCP data
T0 16.12 K 16.44 K
Tau 0.0091 0.0086

Tipping scan

More study of Tsys versus elevation was done by a tipping scan, driving the antenna from 8 to 60 degrees elevation at 50 degrees azimuth, and recording the system temperature five times per second. Figure 3 shows these scans and the model fits using a similar model as above. The opacities, which is depend more on the low elevation data, agree well with those derived from the "peak" data.

Figure 3: Tipping Scans.

Gain Curve Measurements

The Tant values for observed sources were calculated by "GFM" from each "peak" procedure. The following sources forming standard scale were observed: 3C48 (0134+329), 3C147 (0538+498), 3C286 (1328+307), 3C295 (1409+524). Most of the time was spend observing the last two sources. We have used Baars et al. (1977) parameters for the spectra of the sources 3C48, 3C286, 3C295. Significant change in flux of 3C147 encouradged us to use Ott et al. (1994) spectra parameters for it. The size correction factor taken from Ott et al. (1994) was applied.
Actual values (size correction factor applied) of calibrators flux density in use are as follows:
Calibrator Flux density, Jy
3C48 10.81
3C123 32.31
3C147 15.13
3C286 11.65
3C295 14.64

Gain = Tant/Sstandard is plotted on Figure 4.

Figure 4: noncorrected gain curve

A second order polynomial was least-square-fitted to the data. The maximum value of the gain was taken as the DPFU. Resulted gain curve as a function of the zenith distance (ZD=90-Elev, deg.) has the following form:
Gain[LCP,RCP] = DPFU[LCP,RCP] * (A0 + A1 * ZD + A2 * ZD^2).
The fitted parameters for not corrected gain are as follows:
A0, A1, A2 = 0.97935, +1.1154e-3, -1.7833e-05
Parameter Fit to LCP data Fit to RCP data
DPFU 1.91 K/Jy 1.75 K/Jy
Mean gain for elev. 20-87 deg. 1.89 K/Jy 1.73 K/Jy

Figure 5: corrected for atmospheric absorption gain curve

A second order polynomial was least-square-fitted to the data. The maximum value of the corrected gain was taken as the DPFU_corr. Resulted gain curve as a function of the zenith distance (ZD=90-Elev, deg.) has the following form:
Gain_corr[LCP,RCP] = DPFU_corr[LCP,RCP] * (A0_corr + A1_corr * ZD + A2_corr * ZD^2).
The fitted parameters for corrected gain are as follows:
(A0, A1, A2)_corr = 0.97423, +1.2704e-3, -1.7900e-5
Parameter Fit to LCP data Fit to RCP data
DPFU_corr 1.93 K/Jy 1.75 K/Jy
Mean gain_corr for elev. 20-87 deg. 1.91 K/Jy 1.74 K/Jy

The luck of low elevation data in this run calls for additional gain curve measurements, when appropriate period of GBT time will be available.

Efficiency Estimation

We have estimated the GBT efficiency, assuming that projected geometrical GBT area is equal to 7853.98 m^2. Error in the efficiency estimation is about 10% and comes from the uncertanty of Tcal value.

Efficiency is plotted on Figure 4.

Figure 4: efficiency