MyT Guiding updates - RA excursions

[Clayton Yendrey]

Guide logs here: https://openphdguiding.org/logs/dl/PHD2_logs_vwN5.zip

I ran an imaging session last week with my Paramount MyT, this is my third outing with it (it is mounted on a permanent pier).

The guiding varied from very good (.4" ttl rms) to barely OK (1"+ ttl rms).  I was seeing some largish RA excursion that appeared to have a regular-ish frequency, at least that was my impression.

I looked at the logs in PHD2 viewer; the frequency analysis is indistinguishable from the first light session.

I did not run Guide Assistant but I did run a Calibration since I move the OTA on/off the mount since the pier is unprotected (I cover the mount with a TeleGizmo Solar Cover when not in use).

I think the issue was seeing related - the only changes I made during the guiding sessions was to the guide exposure length.  Guiding did have issues after moonrise, so had to stop/let PHD2 select new guidestars which seeme to resolve this).  

Meridian flip was accomplished without problems since I now have DEC reversal box checked in PHD2 (I had forgotten to do this on my first light outing even though I knew it was required for Paramounts).

The raw PE of the mount was measured at 2.9" after the Tpoint model was implemented (80 stars).  It was/ is 1.9" after the PEC was implemented.  The PEC was to the minimum time recommended by Software Bisque (20 minutes) for the MyT.  My plan is to let the mount 'run-in' for a few more sessions, then start with a fresh T-Point, check the uncorrected PE for any shifts, then run at least an hour PEC recording.   My thinking is that will filter out noise/seeing influences that might still be present in the current PEC due to its brevity.

I'd like some input /review of the lastest logs to see if there is something else I should do tuning related or ???  My goal is to not have the RA excursions or least mitigate them to keep the total RMS error from jumping up so much.  ???

[Brian Valente]

Hi Clayton

Looking at your log files, i don't think i would call those excursions - they seem more like instant bumps that have an oscillation to them

here's a typical example, where it immediately drops nearly 3" and when corrected, it oscillates

This is happening outside of PHD, so I don't know if there's a lot i suggest within PHD

What i do think about is PEC and TPoint model. those can cause these kinds of behaviors if they aren't setup and tuned properly (I only have experience with AP mounts and APCC pointing and tracking models). 

i suggest you try some guiding without those enabled, and then add them one at a time (PEC and then tpoint) and see if one or both of them cause this behavior

Brian

--
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/1ed23947-524b-4cc4-921c-1d8ac531f997n%40googlegroups.com.

--

Brian 

Brian Valente

portfolio brianvalentephotography.com 

[Clayton Yendrey]

Brian,

Thanks for the quick response.  I'm going to try to approach this somewhat incrementally but I'm a little loathe to abandon the Tpoint/PEC since the unguided results appear to be quite good.  What I am suspicious of is ProTrack since it is putting its own input into the tracking for RA/DEC.  It input is based on the Tpoint model results so some feedback noise into the response seems possible.  It seems there could very easily be 'conflicts' between ProTrack and another external RA/DEC input source like PHD2.  So I think I will start there first with ProTrack disabled when using PHD2 (it may be a week or two given current forecasts...).

[Brian Valente]

>>> I'm going to try to approach this somewhat incrementally but I'm a little loathe to abandon the Tpoint/PEC since the unguided results appear to be quite good. 

i definitely agree - do not abandon TPoint/PEC!!

i was only describing an incremental approach to finding the source of that oscillation

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/c765a020-6555-418f-8dd7-a324544f2779n%40googlegroups.com.

[Clayton Yendrey]

Brian,

Thanks.  Software Bisque has asked to see the Tpoint model after sharing our discussion with them.  I'm hoping they'll say I missed something in the setup that is easily remedied.  ;-)

[Brian Valente]

That's good to hear. 

Speaking from personal experience i've messed up plenty of models, so i'm hopeful for you

B

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/ba70a501-71fe-4c12-a16a-b3ae6abe443bn%40googlegroups.com.

[Clayton Yendrey]

Hahaha!!!  If you don't break it, you can't learn to fix it!!!  ;-)

[Brian Valente]

If that's the case I must have learned to fix a lot of stuff. so very..... very many things :)

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/a0cb7891-3979-4312-a088-e93bd9a4ddccn%40googlegroups.com.

[bw_msgboard]

Hi Clayton. Can you tell us what your guiding setup is?    Judging from the image scale, it looks like you’re using a separate guide scope assembly.  If so, these are the kinds of excursions we often see with those things – something loose or moving around on the guiding assembly or a brief cable snag that can cause these apparent oscillations.  Having a very accurate mount doesn’t exempt you from these kinds of problems and I don’t think you can assume this has anything to do with the mount.

 

Good luck,

Bruce

 

---

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/CAJa45i7-1GS%3DUR-pvWtaMmbh4VTDh9FqWs81HDpvkixHVEZ4Ew%40mail.gmail.com.

[Clayton Yendrey]

I believe to very solid, much more rigid that the original setup that I had.  Guide camera is a Altair 60GS 60mm, camera a ZWO ASI120mm-s.  There are no cables leaving the OTA except a power and USB connection to the UPBv2 and those travel internally, through the mount, not externally.

[Clayton Yendrey]

I just went through all the fastening points again.  No slack in the guidescope v-d plate attachment.  I did get about a quarter turn more on the top and bottom D plate fasteners.  When the weather clears up, I'll uncover the pier/mount and check the Versaplate attachment bolts on the mount.

[Clayton Yendrey]

This is the rig on the mount on the pier.

[bw_msgboard]

Hi Clayton.  In my experience, nearly everyone who uses a rig like this thinks there’s no way it can move around – and they’re usually mistaken.  There are a couple of things to not like about this arrangement:

 

1. It’s full of thumb-screw fasteners, no doubt Delrin-tipped to prevent marring the finish on the guide scope.  Those don’t stay tight through temperature changes and changes in the gravitational load on the assembly as the scope moves through meridian flips
2. I would say there is too much separation between the axes of the two scopes.  Start counting up the mechanical interfaces here, places where different pieces are mated and fastened.  And think about the moment arm you’ve created.  These “big” 3 arc-sec excursions you’re seeing equate to a shift in the guide camera by about 1 pixel or 3.8 microns.  That’s less than 1/10^th the thickness of a human hair.  And if the part that’s moving is actually down close to the main scope, the amount of movement there might be very much smaller but it’s magnified by the large moment arm you’ve created.

 

Finally, through-the-mount cabling is no guarantee of freedom from cable problems.  The cables can intertwine in there and they won’t typically have the same lengths.  So the shortest cable may end up pulling on the rig at various pointing positions.  The best guide scope arrangements I’ve seen use clam-shell type fasteners rather than thumb-screws and keep the optical axes of the two scopes as close together as possible.  They also typically have a mounting point closer to the guide camera to reduce sag in the focuser.

 

None of these may be problems for you.  But if I was the mount rep, I’d ask you to guide through the main scope and demonstrate the problem that way.  And just personally, I’d never settle for anything less than an OAG for a mount as expensive as this.  That’s just a personal opinion, I don’t want to offend you, and obviously you need to apply your own judgments on this.  Perhaps doing some guiding tests through the main scope will shed light on the situation.

 

Regards,

Bruce

---

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/e77c1740-89f6-4b10-8b3c-c687a7b20eabn%40googlegroups.com.

[Clayton Yendrey]

Bruce, 

I appreciate the input but until I have an OTA with a much longer FL than currently, no OAG.  They introduce their own set of problems and many of the more 'popular' brands seem to have issues with a *lot* of flexure issues that require constant attention to the guide camera attachment/focus point.  That is demonstrable from the discussions in this group, and on other discords.  I'd love to have a clamp ring for the guide scope like that for the OTA, but haven't spent any time to look for one.  YMMV

With my rich field FoV (don't really consider 550mm fl wide field IMO), I've demonstrated through my previous mount failures that this OTA/Camera setup can get round stars with 5 minute exposures as long as the ttl RMS error doesn't exceed around 1.8as.  In looking at the RA through the three long guide sequences, there does appear to be a 1 minute-ish period with the RA - its visibility/magnitude comes and goes, but does keep showing up.  It also appears that DEC really doesn't like the dither inputs, so I'm going to go to RA only dithers and may reduce the frequency as well.  Some of the worst RA movements seem to occur with equally extreme DEC movements but in the opposite direction (on the PHD2 Log Viewer graph) but I don't know what this may indicate or if it is expected.  I am going to disable DEC backlash compensation next time - this was the first time it was enabled for this mount; the last GA session was the first time GA gave any indication of DEC backlash and the amount was very small, so I'm not sure that it is either 'real' or enough a concern that compensation is beneficial rather than a complication.  Software Bisque support has indicated that it shouldn't be necessary from reading other discussions on their support site.

I don't see anything in the log trends that indicates the RA movement is associated with a specific position - it seems to occur through out the night to varying degrees and there was only a single target being tracked for the entire session.  I don't see anything that seems to show the RA movement was worse or better before or after the meridian flip, an event which should induce a noticeable response (good or bad) if it 'slack' existed or cables were too tight.  Dithers do seem to unsettle things (especially the DEC) which may point to some algorithm tuning issues.

There are two primary differences between this session and the first light session (after I got through all the setup and Median flip configuration) - in this session DEC backlash compensation in PHD2 was enabled, and ProTrack was enabled in TSX.  Neither was the case in the previous session which did not have as large RA movement although it was present.  That is why I'm going to drop back to that 'startup' configuration and wait to see what SB has to say about the Tpoint model.  Just as an FYI, I know of one other MyT user that feels that PHD2 and ProTrack do not play well together; however he has an 11" SCT so there isn't a direct comparison relative to my setup for guiding (and he does use an OAG).

[Clayton Yendrey]

Brian,

Thought I'd post an update.  Software Bisque/TSX/MyT support didn't see any issues per se with my model or PE/PEC.  However, they said the PE/PEC  could be improved by selecting a star that was not saturated (I missed that when I was recording the PE).  There is also some 'confusion' in the PE/PEC logs because TSX/TSC captures the manually entered guide scope/camera configuration instead of the 'live' information from the camera/OTA actually being used at the time the  PE is recorded.  I'm almost completely certain that I made the switch to the primary OTA/Imaging camera before recording the PE, but that cannot be verified from the logs.

So my 'answer' at the moment is to record a new PE/PEC after insuring the guide star is only in 50% range of the well depth for the gain I have the camera operating at.  At unity gain, the 2600MC-P available well depth is getting close to around 50% of its advertised well depth.  Which may have been where I messed up on the first PE, but I just don't remember checking the ADU.  I've removed all references to the guide scope/camera from TSX.  I think I'm going to create a new Tpoint model as well, going larger this time (instead of 80, looking to go to 150ish stars).

I've also been thinking about my guide camera/guide scope - this setup was decided on when the ASI294MC-P was the primary camera.  I never went back and looked at this when I changed the primary camera to an ASI2600MC-P, but with the original configuration I was just barely inside the 30% ratio (per the  white paper on guide scopes/cameras found on Agena Astro) at a ratio of 1:1.27. With the ASI2600MC-P as the primary camera (with pixels almost 1u smaller than the 294MC), the px resolution of the guide scope/camera combination I have is well outside the 30% range relative to the primary OTA/Camera a (as calculated on Astronomy tools it is 1:2.44 - over a 200% ratio).   I just overlooked this when I made the primary imager change.  With the upgrade(s) I have planned (Williams Optics Guidestar 61/ASI290MM), it gets the relative ratio inside 20% (1:1.18)

I don't know for certain how much improvement or help this will be (if any) or how PHD2 + MyT will respond, but it will be interesting to find out. ;-)