PHD2, PA and mount ticks optimization (mesu-200 friction mount)

[Panagiotis Papadakos]

Dear all,

I am trying to optimize my mesu-200, a friction drive mount which I converted to OnStepX (from the original argonavis / servocat) combo. Currently, I am playing again with the ticks of my mount since I think that I can improve my guiding especially in RA. Both axes are using HEDM-5500-B02 encoders with a ratio of 1:2,000 so in total I should have around 8,000,000 ticks / 22,222.222 ticks/degree. However there are some variations in the gears so the number of ticks must be fine-tuned. Right now the values that I am using are  7,977,460 total ticks or 22,160 ticks/degree  in the RA and 8,016,110 total ticks or 22,267  ticks/degree. These values were gathered by slowly rotating the two axes and getting the values of the encoders. However it might be the case that the numbers above are not accurate due to some slippage.

In this thread however, another approach is discussed by measuring the distance that a star trails in the sky in a time window of around 10min. So I have been wondering about some things:

a) I think the discussed approach depends on the PA of the mount. You need an accurate PA to be able to measure the drift. Else how can you say whether the drift in the RA is due to bad PA or inaccurate speed? It seems a chicken-egg problem.

b) PHD2 offers a calibration assistant that also reports drifts. Is the technique discussed in the thread the same as the drift measured by PHD2 in the calibration assistant?

c) I have issues using any of the automated tools for PA either in NINA or PHD2. The best PA I ever had allowing even 4 min unguided imaging with my RC8 was done manually using the drift PA. I have used both drift PA and drift Polat PA in PHD2.

d) How accurate should the reported focal length of the scope be for such tools to work accurately enough? The issue is that I have an RC OTA and the collimation of the mirrors alters the focal length. Using astrometry.net I have computed the fl of my RC to 1619mm, which is a bit different than the value of 1630 I used in both NINA/PHD2 settings.

Best regards

Panagiotis

[Bruce Waddington]

I think you are right to be skeptical of this whole business.  As you have realized, the PHD2 calibration is subject to various measurement uncertainties, notably uncorrected periodic error in RA and polar alignment error.  Both of these contribute to what we report as orthogonality error which is essentially an uncertainty about the exact orientation of the RA and Dec axes.  Within reason, these things don't materially degrade guiding.  It seems to me if you want to more precisely measure tracking errors only in RA, you should do it via plate-solving.  Start an unguided session near the horizon and continue it to the central meridian, periodically taking full-frame exposures and plate-solving them to compute RA and Dec coordinates.  Then you should repeat the exercise on the other side of the pier.  The reason I suggest doing it this way is that, in my experience, "one-sided" guiding in RA is more likely to be caused by flexure of the payload than any substantial error in sidereal tracking rates.  As a result, I would expect the experiment I've described will show a non-linear rate of tracking error depending on scope pointing position.  I must say, I was taken by the small photograph of the scope setup in the thread you referenced - a large Newtonian scope *using mounting rings* with a large refractor on top that we have to hope was not being used as a guide scope.  To my eye, this has "flexure" written all over it - but I could be wrong of course. :-)

To wrap this up, one-sided RA guiding can be caused by a number of factors unrelated to native tracking accuracy including flexure, payload imbalance, clutch looseness, or guiding aggressiveness.  And none of this is necessarily a problem in itself - in the end, what matters is the star sizes and star eccentricity `in your final images.

Regards,

Bruce 

[Panagiotis Papadakos]

Thank you for the detailed reply Bruce!

I will follow your advice and check out the tracking of the mount using plate solve across the pier sides.

Also it seems that my RA has a small calibration issue of the friction plates which might be the root cause of the various issues I am facing.

Best 

Panagiotis