[evlatests] Phase vs. elevation: state of play
Vivek Dhawan
vdhawan at nrao.edu
Fri Aug 15 19:16:28 EDT 2008
Recent tests are summarized. I hesitated to send another complicated
email with partial progress, but then decided to write it down before
I forgot it all.
Some progress: L302's show a marked phase change with elevation, ~60deg
at 13GHz. This is tentatively attributed to the 128 MHz reference
from the L305, not the L302 modules themselves, because all L302's
on a given antenna show it to the same degree, other antennas show
very little.
The 512 MHz reference (and multiples derived from it) are clean in
some tests, but not all.
More next week: Tests to a) Clarify the diagnosis and
b) Isolate the sensitive components in the L305.
A bit more detail: The measurement type in the table below refers to
A = astronomical fringe phase.
For the next 3 types the antenna is tipped up and down while
monitoring the phase of a test signal at 512 MHz.
R = Round-trip LO phase @ 512 MHz on the antenna's L352 meter.
D = Phase of test signal divided to 512, measured on spare L352.
M = Phase of test signal multiplied to 512, measured on spare L352.
# type Details
1. A Baseline run, at X & L band. X band flat vs. elevation. L changes
~60 deg with elevation on some antennas, all 4 IF's identical.
L band uses the L301 and L302, at 13 and 14.4 GHz respectively.
X band uses only the L302 at 11.4 GHz. The L301 is made from
(N*512 + 128) so this implicates the L301. Other tests below
exonerate the 512*N. Need to resolve this!
2. R 512 reference to antenna. Clean (factor ~8 less than #1.)
3. D 4096 = 512*8 from L300 comb generator. As clean #2.
4. D L302 locked to 128 reference. Set to 12288 = 512*24.
Shows elevation change 60deg at 13GHz, like #1.
5. D Same L302 locked to 512 reference. As clean as #2.
plan to do one or more of these next week:
6. M 128 from L300. Measure 128*4
7. - 512 from L300 - compare with #1 and #6.
8. Swap L305 of bad/good antennas, do L302's follow L305?
-------------------------------------------------------------------------
More information about the evlatests
mailing list