Hi,

Yesterday I noticed some inconsistency of how CCDAP reports plate solve accuracy with sub-frame or full frame. This might also be related to target lists.

Just before my testruns with CCDAP, I did a 100 point TPoint mapping run, which gave me very good all-sky goto accuracy. I tested this using TheSky6 manually.

With sub-frames on, CCDAP sometimes reported slew errors of more than 1000 arc-sec! With sub-frames off and with different target list my slews where all the time within 50 arc-sec?

Pretty weird. With reported 1000 arc-sec error my target was still almost center of the frame.

Logfiles for both good and bad accuracy from same evenign (combined into one file)

ccdap_comb.log (34 KB, 5 downloads)

Hi Petri,

This is interesting. I noticed that very few stars are in the full frame solution, 14 - 21, and 300+ are in the half-frame solution. I believe the quality of the solution depends on the number of stars. I wonder if somehow some hot pixels are being counted as stars and throwing off the solution? When you sub-frame, you may be missing some of those hot pixels. You may want to turn trace on so that successfully solved sync images are retained. You can then examine them in both Maxim and Visual PinPoint to see if this is so.

Another alternative is, given your modest FOV and high pointing accuracy, the expansion of 0.3 is causing PinPoint to look at too large a field for a solution. The next maintenance release, currently under test and to be released shortly, will allow setting the expansion to 0. This may help as well.