SPARO CSO Tests

  • 4He Hold Time

    • The hold time for the reservoir with all of the optics and electronics installed and the JFET heaters warming the JFETS to ~135 K (27 Volts each) was about 3 days(loss rate=5 inches/day).

  • Pixel Separation

    • The nearest neighbor pixel separation in the 9 element configuration was calculated to be about 31 arcseconds.

  • Noise Equivalent Flux Density (NEFD)

    • The NEFD was determined to be 15 Jy/Hz1/2

  • Sources Observed

    • The Moon(calibration)
    Jupiter(calibration)
    M42.1(Orion)
    W51
    W49
    KL
    KHW
    Sgr B2

  • Polarization in Sgr B2 and Contribution from the Instrument

    • The instrumental (SPARO + telescope) polarization was measured in two different way. The first(and more crude) measurement was done using the observations of Sgr B2. As an alt-az mount telescope tracks an object over the course of a night, the object will appear to rotate with respect to the "stationary" instrument. The effect of this rotation is to rotate the polarization vector from the point of view of the detectors in sparo. Since SPARO is designed for deployment at the south pole(where an alt-az telescope is an equatorial telescope and no relative rotation occurs), the cryostat is not designed to rotate with the sky. Thus, during the observations of Sgr B2, SPARO was held fixed with respect to the telescope. Therefore, given a sufficient amount of sky rotation, it is possible to do a fit to separate the contribution of the instrument from that of the source. The resulting instrumental polarization from this measurement was P=0.416 ± 0.539, phi=170.46 ± 37.14. The errors are large because the sky rotation was only about 27 degrees between the two nights of Sgr B2 observations. Note also that this analysis was done using pixel 9. The resulting Sgr B2 measured polarization was P=0.960 ± 0.506, phi=143.01 ± 15.17.

    The second test involved our Jupiter observations. For this we used the routine "rougher" as described in the Data Analysis section of the web page. However, we used this code in a slightly nonstandard way. Normally, rougher is given an instrumental polarization and is used to calculate a source polarization. In our case, we input a null instrument polarization and looked at an unpolarized source (Jupiter) and any source polarization we measured, we assumed to be due to the instrument. The resulting instrument polarization measurment was P=0.82 ± 0.05, phi= 152.8 ±1.9. These are in agreement with the measurements made with the other technique. There is a problem with using rougher in this fashion. Rougher takes into account sky rotation when calculating the polarization of a source whereas, the instrumental polarization does not rotate. Since the sky did rotate through the course of these observations by about 11 degrees, the errors will be slightly greater than those listed above. Once the instrument polarization was determined, the Sgr B2 polarization was calculated(again using rougher) to be P=0.47 ± 0.25, phi=140.6 ±15.1. This result is also in agreement with the above result.

    The edge effect for pixel 9 was also measured using observations of Jupiter. The basic idea was to take three polarization measurments: one with Jupiter in the middle of pixel 9, one with Jupiter on the inner radial edge of the detector, and one with Jupiter on the outer radial edge of the detector. The results are summarized as follows:

    Position of Jupiter Polarization(%) Polarization Angle(degrees)
    center 0.82±0.05 152.8± 1.9
    inner edge 0.83±.18 146.4±6.4
    outer edge 1.24±0.23 101.3±5.5


    As the above data show, there is an edge effect on the polarization as the object moves to the outer edge of the field of view.

  • Evaluation of Zero Angle Chi

    • According to Platt, et al, the zero angle of the dewar is related to the phase of the polarization angle, delta, buy

    delta= 1/2(phi-alpha+gamma-chi)

    Here, phi is the angle of polarization on the sky, alpha is the sky rotation, and gamma is the dewar rotation. To measure this angle, we used a large polarizing grid placed on the telescope with its plane of polarization oriented at 90 degrees with respect to the vertical. This defines phi=90+alpha. Using rougher once again, we were able to get a measurment of phi-chi. From these measurments and for those with the plane of polarization at 30 degrees, we obtained an average chi of 59.04±2.08 degrees.
    Note: These are preliminary results and are currently being checked.