
-4-
y-directions
2
) from desired position
3. Calculate the duration of slews (x- and y-
directions) required at the current slew
(guiding) rate to return the star to the desired
position.
4. Issue command(s) to the telescope to slew in
the direction(s) required
5. Stop the slew(s) after the calculated
duration(s).
6. Repeat the process.
The duration of the slew may be altered (usually by
an “aggressiveness” setting) to over- or
undershoot the required slew in steps 4 and 5.
Generally, the position errors induced by improper
tracking (periodic error, inaccurate sidereal rate, or
poor polar alignment) are much smaller than the
sidereal rate. In other words, the errors that must
be corrected are much less than 15 arcseconds per
second. At a guiding frequency of 1 Hz, the errors
are typically a few arcseconds or so (depending on
seeing).
Typically in the past, MTSCP autoguiding for
LX200 telescopes has attempted to use the :Md#
and :Qd# commands (where d is the direction of
slew). For a LX200GPS using a guiding rate
setting of 66% (about 10 arcseconds per second)
and a guiding frequency of 1 Hz. This means that
the required slewing movements (at the guide rate)
are typically 300 milliseconds or less. The software
would calculate the duration of the slews (both x-
and y- directions), issue :Ms# and :Me#
commands (for a guide star that had moved
southeast of the desired position), then issue :Qs#
and :Qe# commands at the proper times.
In firmware version 2.0i and below, the :Qe#
command is followed by a pause of the sidereal
tracking, variously reported as 200 to 500
milliseconds. Imagine the effect this has on the
process. The star has just been moved back on
center (assuming an aggressiveness setting of
100%), but it is now allowed to drift eastward by
three to eight arcseconds, some of which may occur
while the guide camera is acquiring the image that will be
used to calculate the next corrections. This drift has not
been anticipated by the guiding software. When
the next frame from the guide camera is analyzed,
a large position error will be calculated and
appropriate commands issued to return the star to
the desired position. If the aggressiveness is set at
100% or higher, and if all of the tracking pause
occurred before the guide camera image for the
next iteration has been captured, the star will be
moved back to the desired position, only to be
once again immediately “knocked off position.”
The result will be oscillation of the guide star back
and forth with each guiding iteration. However, if
the aggressiveness is set to less than 100%, or if
tracking pause has not been completed before the
next frame is acquired by the guide camera, then
the guiding software will not be able to fully move
the star back to center. With each iteration, the
star will move farther east and will quickly leave
the frame.
This latter scenario is exactly what users of
K3CCDTools2 v. 2.1.3 and below have
experienced, as well as GuideDog users running
versions 1.0.4 and below. The software was
working correctly, but the telescope was not doing
as commanded. (Ironically, this problem does not
seem to have affected LX200 classic users.)
Neither Peter Katreniak (author of K3CCDTools)
nor Steve Barkes (author of GuideDog) have
LX200GPS telescopes (although Steve has a
LX200 classic), so they are both dependent on
user reports to assess the performance of their
software on our telescopes. Both Steve and Peter
are aware of the LX200GPS problems, and have
taken measures to address them (as described
below).
But the plot thickens.....
Undocumented Command
In an attempt to figure out what the heck is going
on with LX200GPS autoguiding, Web Dove used
PortMon to log the commands that were being
issued by autoguiding software through the serial
port of his computer to his LX200GPS. He did
this using Meade’s Autostar Suite and GuideDog
version 1.04. In doing so, he made an important
discovery. GuideDog was using :Md# and :Qd#
commands as described above to guide the scope.
However, he discovered that Meade’s Autostar
Suite uses an undocumented command.
2
Some autoguidng software defines “x” and “y” axes.
Normally, the x-axis is aligned with the RA axis, and the y-
axis is aligned with the declination axis.
Comentarios a estos manuales