Trying to interpret guide log results from my EQ8-R

[Jeff Nibler]

I had a guiding issue I wasn't able to solve with this mount last year and I put it to the side as a result. I need to get it working now so am picking back up from where I left off. 

https://openphdguiding.org/logs/dl/PHD2_logs_2cFQ.zip

The main issue is Dec axis behavior. I'm aware that the worm gears are not spring-loaded on these mounts which causes a lot of issues for people but I'm not sure if what I'm seeing in the guide logs is pointing to too much Dec backlash or Dec axis binding. 

From the guiding assistant runs I performed, it does seem that there is ~3,800ms of Dec backlash which is a lot. I am happy to adjust that manually on the mount if I need to but I want to be sure that's the problem first. 

Note that in the attached log, between midnight and around 1am or 1:30am I was remotely adjusting settings so if you see changes around that time, that is why. One of the last things I did was configure Dither to RA-only around that time. That helped mask the issue but I still see things in guiding that don't make sense. 

Given that guiding assistant showed a lot of Dec backlash, what is confusing me is that later in the night after my adjustments (one of the main adjustments I did was to switch dither to RA-only but I also played around with setting Dec Guide Mode to North or South, but ultimately left it on Auto), I see instances where the guide pulses could not correct sudden and substantial Dec drift. I understand that is exactly what backlash looks like but in the log at the time, the dec pulses were always in the same direction, and the same-direction pulses looked like they were working fine to correct Dec drift until they suddenly were not and no amount of pulses in the same direction seemed to correct it for six whole minutes (example 2:48am). When I installed the mount last night, I set the PA so it had 1-2 minutes of PA error so a bit of Dec drift would help with backlash. 

Thanks in advance for any help interpreting these. 

Jeff

[steve]

Hi
If it is backlash, you should easily be able to detect 3 seconds worth mechanically by rocking DEC clockwise and anticlockwise whilst the clutch is engaged. Stiction and tight mesh manifests itself more by overshooting at the end of a guide pulse sequence. I don't see much overshoot on your runs. 
The adjustment is quite straightforward, but note that most guides forget to tell you that the easy way to loosen the dec block is by loosening the motor to gain access to the remaining -difficult access-  bolt.
HTH

--
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 visit https://groups.google.com/d/msgid/open-phd-guiding/4be2614a-4ed1-423d-9bc6-ed4db5bb5212n%40googlegroups.com.

[Jeff Nibler]

Thanks for the reply Steve. I did do that mechanical test and there is backlash on the Dec axis for sure. I've read up on how to adjust the Dec worm mesh and feel comfortable doing it but looking at the guide logs, note 2:48am - you see how it corrects dec in the same direction continuously as it should since Dec is constantly, slowly drifting in one direction due to the small PA alignment error I set, guide pulses are all in the same direction, yet Dec just starts drifting out. Looking at all those dec pulses in the guide log, I would expect to see exactly this for high backlash but only in a situation where the worm gear was unloaded and it changed directions. I don't think it changed directions in this case. Am I interpreting what I'm seeing in the guide log incorrectly? I'm definitely not an expert in interpreting these. 

Thanks again,

Jeff

You received this message because you are subscribed to a topic in the Google Groups "Open PHD Guiding" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/open-phd-guiding/F232lUtg9wE/unsubscribe.
To unsubscribe from this group and all its topics, send an email to open-phd-guidi...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/open-phd-guiding/9b963255-42bf-4dbf-b860-823d4735cf1e%40gmail.com.

[steve]

Hi Jeff. 

We just go by the old adage: the mount has to be better than perfect for the software to be effective. If you can feel backlash, fix that first. You'll then be in a much better position to assess the guide log. Many eq8rs are unusable out of the box, usually due to the Dec worm adjustment. Ours were no exception. We speak from bitter experience!

Sorry can't interpret your issue at 2:48 but to my untrained eyes it looks like it could be runout on the crown and you've hit a low zone. I'm sure the software gurus will have a more specific advice. 

Good luck and get those hex wrenches clean and ready.

To view this discussion visit https://groups.google.com/d/msgid/open-phd-guiding/CAM_xTkcd3PsKjWAsfnoP%2B1B72SL0WGhpK8f5ea3C0Fycum0DZw%40mail.gmail.com.

[Bruce Waddington]

Hi Jeff.  I think the presence of backlash is pretty clear - here's a graph of the test results in case you didn't look at them:

The red points are the 16 consecutive guide pulses to the north and the green points show what happens when the direction of the guide pulses is changed to south.  You can see the flat area on the curve where the Dec axis isn't moving at all.  So this is where the estimate of a 3.2 sec backlash comes from.  I agree with Steve's advice that it would be a good idea to improve on this mechanically, at least to the point where the PHD2 Dec backlash compensation feature is likely to produce good results (roughly 2 seconds of backlash or less).  That said, I'll also point out that your overall tracking and guiding in Dec is roughly equal to what you get in RA so I don't think you're being strongly limited by this.

With regard to what happened at the 02:48 mark, it looks like a bit of an anomaly (Dec in green):

This again shows the axis was unresponsive for several minutes even in the absence of any change in direction.  This could come from something happening with the payload rather than the mount - a cable drag, a loose clutch, a large weight imbalance, or some other form of mechanical resistance outside of the mount drive.  As you can see, the situation resolved itself.  There are several other cases where this kind of problem is suggested:

These are huge excursions unrelated to guiding, so I think they are also something that needs to be tracked down.  As a final note, you should update to the current release (2.6.14) and start using the Calibration Assistant for all your calibrations.  That tool is covered fairly thoroughly in the User Guide.

Good luck,

Bruce

[Jeff Nibler]

Thanks for the reply Bruce. Did the upgrade tonight and have been using guiding assistant to measure the dec backlash at various points on the dec gear tonight until moments ago when clouds rolled in. So far it's consistent at ~ 3800ms +- 200ms at tonight's temperature in all positions except one - CW horizontal, West side of pier, Dec 0; there is no backlash there at all. I did the test twice to confirm. That seems like a mechanical problem with the mount. 

So everything you say makes sense, and I can re-mesh the gears to improve it, while hoping doing so doesn't result in binding at that single point where PHD2 measured no backlash. 

The thing that still puzzles me are the excursions. This OTA has it's cables connected to the ports in the saddle so there are no hanging cables or wires at all. I balanced it quite well in Dec too. It doesn't move with the clutch loose and a gentle nudge in either direction with clutch loose causes about the same amount of movement. It wasn't windy last night but there may have been a gentle breeze here and there. That's the only thing I can think of because in the example I showed, that you pasted in green above, it was definitely pulsing consistently North prior to the excursion, then the excursion happened and the pulses didn't correct it. I know it's not a PHD2 thing or a guiding thing but I'm puzzled to try to understand what might have caused it beyond wind. 

For now I will use Lowpass2 with Dec mode set to North only, until I can get back to the mount and tighten the worm gear. 

Thanks again to both of you,

Jeff

To view this discussion visit https://groups.google.com/d/msgid/open-phd-guiding/1955298d-8f65-469d-80e9-b62a06272e9fn%40googlegroups.com.

[Bruce Waddington]

Keep in mind that we're making measurements at the micron level so even very tiny unwanted movements can cause large apparent guiding excursions.  With regard to using uni-directional Dec guiding, don't forget that the direction needed for guiding depends on the pointing position.  So at some point, you will have to change from north-only to south-only.  Where in the sky that happens depends on the details of the polar alignment error.

Bruce