Atlas EQ-G Guiding Error Analysis
Ethan C
bw_msgboard
Hi Ethan. I don’t know what your original mount looked like, but I think the GA gave you a good picture of this one. There’s a considerable amount of uncorrected periodic error, something like 30 arc-sec peak to peak. The largest frequency component is around 516 seconds which may be the native worm period (RA in red):
There is also a large amount of Dec backlash – it took over 5 seconds for the Dec axis to reverse during the backlash test. You could reduce this by almost 2x if you increase the guide speed setting in the EQMOD to 0.9x sidereal. Of course, a better approach might be to improve the mechanics if that’s practical.
The small difference is drift/polar alignment estimates isn’t relevant. When you were imaging, you were pointing clear up at Dec=68 degrees, so it’s not surprising the drift amount would be somewhat different. Hyper-tuning the polar alignment isn’t really a useful exercise once you’re down below 5 arc-min or so of misalignment error.
Hope this helps,
Bruce
--
You received this message because you are subscribed to the Google Groups
"Open PHD Guiding" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to open-phd-guidi...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/046ff719-a8e7-416d-9912-c4abba109a58%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
Ethan C
Thank you very much for your insight! The periodic error definitely helps explain why my RA had so much more error on this new mount. According to the PHD2 Github, the worm period of the Atlas is 479s. My old mount stayed around an RMS of .6"-.7", and at its worst around 1.1" (which is still better than this mount). The debug log is below! I'll give increasing the guide rate a shot on the next clear night. Thank you again for all of your assistance!
Debug Log: https://openphdguiding.org/logs/dl/PHD2_logs_XbhK.zip
Ethan
To unsubscribe from this group and stop receiving emails from it, send an email to open-phd...@googlegroups.com.
bw_msgboard
Hi Ethan. The reason I asked for the debug log file is that I didn’t understand why the GA recommended such large min-move values – those were almost certainly too large for a typical setup like yours. But when I looked more closely at the actual GA measurements, I saw this Dec behavior:
Remember, the Dec motor isn’t running at all during this time, so this is strange behavior. We normally expect to see the Dec values drifting north or south because of polar alignment error, but we never expect to see this sort of oscillation. And this is a substantial amount of movement, maybe 6 arc-sec peak-peak. The usual explanation is that the guide camera was rotated sometime after the calibration was done, but in your case the calibration was done immediately before the GA run was started. This is also an unusually long GA run, about 35 minutes. So it’s possible the guide camera moved (rotated) during the 35 minute period. This large movement explains why the GA made the min-move recommendations it did. So I think you should take a look at things and make sure that nothing in the guiding assembly can move around like this because it will definitely impair the guiding. If you see GA recommendations like this that don’t make sense, it’s probably an indication that something went wrong with the measurement and, usually, something went wrong with how the guiding assembly/mount behaved during the GA run..
Good luck,
To unsubscribe from this group and stop receiving emails from it, send an email
to open-phd-guidi...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/9447e79b-9173-4fc1-b69a-2a1e14411768%40googlegroups.com.
Ethan C
I apologize for the late reply. I checked to see if the camera could easily rotate and it didn't seem like either it or the guidescope itself could move at all. I do remember there being light wind on the night that I took these measurements, so could that be the cause? I also increased the declination guide rate to .9x and left the RA guide rate at .5x which seemed to help the guiding itself, but I couldn't measure the declination backlash using the guiding assistant, as the test just failed. Below is the guide log. During the calibration, the mound barely moves South (I'm assuming as a result of the backlash, it also advises I re-calibrate because of this) and then when I run the Guiding Assistant at 22:04:58, the backlash measurement test fails.
Guide Log and Debug Log: https://openphdguiding.org/logs/dl/PHD2_logs_ZAdM.zip
Thank you,
Ethan
bw_msgboard
Hi Ethan, sorry you’re continuing to have trouble. I think there are some significant problems with your Dec axis that will need to be fixed mechanically. Dec guiding is definitely the limiting condition for your overall results. In the last session on 6/22, the Dec rms was 1.03 arc-sec while the RA was 0.6 arc-sec. From an earlier test you ran, we know the mount has a very large amount of Dec backlash – it takes nearly 5 seconds to reverse direction even at a mount guide speed of 0.9x sidereal. Beyond that, I think there’s also some stiction (static friction) on the axis that shows up when larger guide pulses are issued during normal guiding. Here’s an example:
The red arrows show a pattern where the first guide pulse doesn’t really accomplish anything, yet the second guide pulse creates an over-shoot. This is usually an indication of backlash (the first pulse) combined with stiction (the second pulse and the over-shoot).
I am assuming, of course, that you have followed all of the advice regarding EQMOD settings from this document: https://github.com/OpenPHDGuiding/phd2/wiki/EQASCOM-Settings
Unless you can make mechanical improvements, I think you will be best off if you guide in only one direction for Dec (uni-directional guiding). There are instructions for doing that in the PHD2 help guide. This can result in perfectly good imaging results, lots of others image this way. You might want to intentionally degrade your polar alignment a bit to make it easier to do uni-directional guiding – maybe something in the range of 5 to 10 arc-min of alignment error.
Cheers,
To unsubscribe from this group and stop receiving emails from it, send an email
to open-phd-guidi...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/a4eccb37-e951-49c9-9b0d-7a1b3ecf056f%40googlegroups.com.