Guiding the AVX what can I improve upon?

[Richard Heinsohn]

Since the forum helped solve a prior problem with guiding and have done well with my DSLR I have made a few equipment changes and hence introduced a new problem I think.

I upgraded to a CCD camera and filter wheel setup to replace my DSLR. I rebalanced my Dec to be slightly heavy on the camera end to keep the gears meshed and weight side heavy in the RA side also to keep the gears meshed. Since I changed equipment I re-did PEC training on the mount and ran a new calibration on PHD2. I was going to try PemPro for PEC to get a more accurate PEC correction factor but it does not run on Win 10 Pro. The Celestron tool will not run on a 64-bit Win 10 Pro system so I am limited to using the mounts internal PEC tool.

What I am seeing now appears to be additional oscillations in both the RA & DEC axis and after dithering the settle time can be up to two minutes. Is this normal?  The two files attached have a few Calibrations in them but number 5 along with the last guide log is the ones I'm trying to figure out. The target was M42 for more than two hours of data and one note I read mentioned to consider tracking in one direction in DEC. My stars are a little out of round and I noticed that the stacking reference of registered images were not overlapping very well and had some major cropping to do on M42.

What could I look at to improve tracking parameters for this mount. Some days it is fairly descent and days like this log are all over the place. I have my eyes and wallet set for a much better imaging mount in the very near future.

Guide log is attached. The Debug log is to large to attach to this post. I don't use drop box.

Thanks,

Richard

[randy....@gmail.com]

Hi Richard,

I could have written this post a few weeks ago. I found the following thread which helped me immensely: https://groups.google.com/forum/#!topic/open-phd-guiding/Uoo1ZIpuRlE

I did not replace my spur gears, but I did remove as much backlash as I could. I concentrated on the DEC axis and found my set screws loose on one of the spur gears causing quite a bit of slop and way too much backlash. While I was at it, I tightened the mesh between the worm and main Dec gear, by slightly loosening the two large set screws and tightening the central set screw about 1/8 turn. After testing it at all speeds, I was convinced everything was much tighter, but not binding.

The next thing I did was load PECtool. I have a 64 bit Windows 10 laptop and it loaded fine. The only problem is you must use ST4 guiding rather than ASCOM, but PE is PE either way. I ran set of 5 sequences and averaged them in the tool and loaded them into my mount. I also changed several parameters in PHD2. I find my mount guides better with 1.5 sec exposures and I set the max durations to 5,000. I always balance my scope slightly east heavy and camera end heavy. However, I changed that to slightly east heavy and balance my altitude axis with no bias at all. I use backlash compensation in PHD2 and find the software figures out the optimal compensation setting after a few minutes guiding.

The results of these changes has been to lower my total PHD2 error from about 1.5-2.25 arcsec depending on seeing to .85 - 1.25 except one night when the seeing was pretty bad. I believe that is about all the guiding performance we can expect from this class of mount and I am very pleased with this imaging with a 80mm Apo with a 1.03 arcsec image scale.

One other thing is Polar alignment. I use SharpCap and get it to 1 arcmin or below. Polar alignment is not a problem with SharpCap. Before making the changes, I always guided in one direction only. Now I use auto and get good results.

Good luck with your mount. A bit of patience goes a long way in getting the most out of the AVX. It also helps to have reasonable expectations for what it can and cannot do.

Randy

[Andy Galasso]

Hi Richard,

The long settling time seems to be caused by ordinary declination backlash.  You could try using the Spiral Dither option and that will reduce the number of dec reversals caused by dither. Or, if you want to go a step further you could try dithering in RA only.

My stars are a little out of round

Your RA and Dec RMS guide error are comparable magnitude, and your guider error scatter plot looks pretty round, so I would not expect that your guiding is the source of the star elongation. I'd be more suspicious of differential flexure.

 

I noticed that the stacking reference of registered images were not overlapping very well and had some major cropping to do on M42.

Does the registration offset appear to be due to translation or rotation?  If rotation, then the source would be polar alignment error.

 

What could I look at to improve tracking parameters for this mount.

I don't see anything guiding parameter-related here.  There's the dec backlash (mechanical) leading to the longer settling times. You could try enabling PHD2's backlash compensation via the Guiding Assistant Measure Dec Backlash option to recover more quickly from the dec reversals.

 

Some days it is fairly descent and days like this log are all over the place.

Don't under-estimate the impact of seeing conditions. Seeing varies dramatically from night to night and even over the course of a night. If seeing is bad, there's really nothing you can do about it--your guiding is going to be "all over the place". The only thing you can do is back off on your guiding aggressiveness (i.e. increase the min-motion settings) so as not to have the guider chasing the bad seeing.

Here's a relatively calm section from your log:

We can learn a lot from this sample.

Declination (red)

  - there's a slow downward trend due to polar alignment error--not necessarily a bad thing unless it is causing too much field rotation.

  - phd2 is making "upward" corrections and the guide star is not moving in response -- this is the signature of dec backlash. notice the declination correction direction reversal at the beginning of the sample. The gears reversed direction and the slack is being taken up.

  - Since the dec corrections are not really doing anything, the "spikiness" of the dec trace is pretty much a direct indication of the seeing conditions. Guide star motion of up to 1 arc-sec or more = not great seeing.

Right Ascension (blue):

  - here we see the ~1" seeing spikes, plus some additional error introduced by the mount's RA tracking error. On a night of good seeing I suspect this would become your limiting factor for guiding. Running the guiding assistant could give you a better picture of the overall RA tracking behavior, and whether you might need to use PEC and/or PPEC.

One more note: in this sample your Dec RMS error was about 0.78" almost completely due to seeing.  Even with perfect RA tracking you would expect your RA RMS error to be comparable magnitude (0.78"), and your total RMS error would be ~ 1.1" (RA and dec errors add in quadrature).  Your actual combined RMS error in this sample was 1.39", not dramatically worse than the "ideal" 1.1" you would expect to get with these seeing conditions.

I wanted to drill into this example to highlight just how dramatic seeing can be on guiding results.  Since seeing is essentially random, it makes it really hard to compare results from night to night. One might easily jump to a causal conclusion ("I changed X and my guiding got worse") when in fact it is just random. (Sorry for getting a little off-topic from your original post.)

Andy

[Richard Heinsohn]

Hi Randy,

Thanks for the info.

I has tried installing the PECtool from Celestron back a couple months ago and the file will not install on none of my Win 10 Pro machines (2 laptops, 1 desktop) so the Celestron tool is useless to me. The install aborts and reports an error message report that it will not install on Win10 6.x. I'm an old hand (Retired IT professional) at software programs and installation so I know my way around a PC hardware and SW.

I have already cleaned, lubed and set the DEC and RA gear sets with some high end Lithium grease  but I'll look into what else you have written here. I believe too that most of my problem is in the DEC axis.

I will be upgrading to a much better mount by the end of this month so I know that tracking should be better. As for now I live with the limitations of the AVX and just try and squeeze the most out of it.

-Richard

[Richard Heinsohn]

Andy,

I had a feeling that this was going to boil down to DEC and seeing conditions. The night I imaged this sequence seemed to be very clear and had the best seeing conditions over a 14 day period for several hours. I did run the GA once that night and accepted the settings except backlash parameters.  I also ran PEC Record from the handheld paddle before running the GA in PHD2. When I get another clear night I'll look at the rest of your suggestions to try out and watch for improvement.

Thanks,

-Richard

[1CM69]

Pretty sure that the PECtool has to be installed & run as Admin on WIN10, vague memories of having to do this myself a while back

[Richard Heinsohn]

I run my machines in Admin mode already so that's not the problem when I receive the error code. Also I tried installing the setup file from compatibility mode and does not work either. I also disabled the anti-virus program and tried a vanilla install of the setup file. No

 go!