[evlatests] EVLA and VLA L-band tuning

[Dan Mertely]

Those are some very odd VLA results!  For the VLA,
the sky frequency is up-converted to C-band via a
fixed, 3200 MHz LO in the F2 up-converter module.
The L6 LO is then mixed-in to convert to the 1025
MHz IF.  This down-conversion is accomplished at
the F4.  According to the standard, VLA block diagram,
the F4 has a "3750/630 MHz BPF" in-line with
the L6, LO input.  the filter should really kill the
LO power much outside of the 3435 - 4065 MHz range.
Using the closest L6 lock-to frequency on the low end
of 3440 produces a center of sky frequency value of
(3440 - 3200 + 1025) +/- 25 MHz, or 1265 +/- 25 MHz,
indicating that the lowest frequency the VLA should
be able to tune to is 1240-ish.

On the high end, the last L6 lock-to frequency that
can make it through the F4 LO input filter would be
4060, which produces a center of sky frequency value of
(4060 - 3200 + 1025) +/- 25 MHz, or 1885 +/- 25 MHz,
indicating that the highest frequency the VLA should
be able to tune to is 1910-ish.

If the block diagram is correct about the F4 LO input
filter, I can't see how VLA antennas can produce any
output much below 1240 MHz, *maybe* 30 MHz below that
at the absolute minimum.

-Mert


Rick Perley wrote:
>     Ken and I have done a number of tests over the past two days to 
> explore the tuning ranges, and the sensitivities of the EVLA and VLA's 
> L-band receivers.  The results are quite astonishing. 
>     The complete frequency range we explored was from 1070 to 2010 MHz, 
> spaced by 10 MHz at the ends, and by 50 MHz in the middle (where the 
> characteristics are already well established).  Observations were loaded 
> as correlation coefficients, so a 'CALIB' provides a direct measure of 
> antenna sensitivity.  We used the spectral line mode '4', with either 
> 6.25 or 12.5 MHz BW. 
> 
>     1) VLA tuning ranges and sensitivites. 
> 
>     Remarkably, some VLA antennas gave sensible fringes as low as 1110 
> MHz.  The sensitivities at this frequency are typically a factor of four 
> worse than at band center.  All VLA antennas gave fringes at 1120 MHz, 
> and higher.  However, the sensitivities are a sharp function of 
> frequency, due to various resonances in the VLA's polarizer.  See the 
> plot on page 9 of the 'VLA Observational Status Summary' which shows the 
> array sensitivity from 1255 to 1145 MHz.  The 'good regions' shown in 
> that plot are confirmed in our measurements.   In summary, these are:
>        a) above 1220 MHz.
>        b)  from 1190 to 1205 MHz
>        c) from 1165 to 1175 MHz
>        d) from 1150 to 1157 MHz. 
>     At the high frequency end, the VLA antennas operate very well up to 
> 1900 MHz.  There is little loss in sensitivity for VLA antennas up to 
> this frequency.  Limitations of the L6 synthesizer make it impossible to 
> tune the VLA to any higher frequency. 
>     There is a notable loss in VLA sensitivity -- by a factor of two to 
> four -- in the range 1800 -- 1820 MHz.   Above this, the sensitivities 
> return to 'band-central' values, up to the tuning limit of 1900 MHz. 
> 
>     2) EVLA tuning ranges and sensitivites.
> 
>     It seems that the EVLA antennas will tune, with good sensitivity (!) 
> down to 1070 MHz.  This is a qualified statement, as these lowest 
> frequencies were the first in the file, and not all antennas got to 
> source in time (or -- they didn't lock up -- more tests will be needed 
> to determine the difference). 
>     With the exception of ea13, the sensitivities of EVLA antennas is 
> flat between ~1150 and 2010 MHz, although there is considerable 
> variation in the measurements below 1200 MHz between the two IFs.  The 
> resonances in the VLA antennas are absent for the EVLA antennas, as they 
> employ a different style of quadrature phase shifter. 
>     Most remarkably -- all EVLA antennas are as sensitive at 2010 MHz as 
> they are at the center of the band.  We do not know the top end of the 
> tuning range -- 2010 MHz is as high as we tried.  Further tests will be 
> run to see how high we can go. 
> _______________________________________________
> evlatests mailing list
> evlatests at listmgr.cv.nrao.edu
> http://listmgr.cv.nrao.edu/mailman/listinfo/evlatests