In article ,
Max Power wrote:

High Angle Range WU, what is the highest acceptable design value (PR -
telescope)?

Something around 63 degrees, although possibly not in 107 seconds.

This is actually more difficult to compute than it at first seems. There
will always be a component of about 0.5 degrees due to earth rotation, but
this needs to be combined in a vector manner with the rest of the range.
For useful work units, the telescope is either locked down or is
tracking with the carriage house or Gregorian receivers. When locked down,
only the 0.5 or so degrees component will apply. When tracking with the
carriage house, one would expect an approximately zero range. When the
Gregorian is tracking, the carriage house will, typically, be swept in
an arc, but that arc will be a small part of the possible motion during
any one work unit.

When repositioning, there could be a much larger angle range. Moreover,
the motion may not be uniform. Generally such work units are not terribly
useful.

The telescope pointing data is only available at the start of each block
of data on the data logger tapes (these positions are included in the full
work unit header). The latest BOINC working version of the splitter
computes the overall angle range by summing up the great circle angles
between the starts and ends of each block:

for (j=2;jTAPE_FRAMES_PER_WU;j++) {
wugrp.data_desc.true_angle_range+=angdist(tapehead er[startframe+j-1].tels
tr,tapeheader[startframe+j].telstr);
}

The maximum possible excess angle range over that from earth rotation, in
one hop, is from minimum altitude in one direction (about 70 degrees)
to minimum in the opposite, i.e. about 40 degrees. The maximum possible
non-earth rotation component, integrated around the whole path, for a
motion that doesn't double back on itself, is probably that for a 180
degree azimuth rotation with the carriage house at 70 degrees declination
(where it is most of the time). Although one should really use spherical
trig, I think a reasonable approximation would be to assume pi/2 times
the 40 degrees for the direct motion, giving about 63 degrees. To get
a more accurate figure one needs to account for earth rotation, spherical
trig and for the measurements on the tape defining an inscribed polygon,
not the actual circular motion.

A direct, great circle, motion over the 40 degree range isn't really
possible because the Gregorian will block the carriage house, so it
would have to be done with an azimuth swing.

The real maximum may well be less as it may not be possible to slew
through 180 degrees in 107 seconds. That information is probably
available on the Arecibo web site.

Why is this not indiacted in the SETI client itself?

Probably the same reason that your spelling error isn't *indicated* by
my newsreader.

Why is the work unit not tagged as [meta-data] invalid for the PR telescope,
if such high Angle Ranges are reached?

Presumably because the client doesn't expect the splitters to create this
sort of error and, given that it does, it doesn't matter in that the bad
work unit can be rejected later, if it otherwise looks interesting.

In this case, I suspect that there was a gap in the data and the angle range
has actually be computed over several hours, rather than 107 seconds.