Settling Failed

[Bill Richards]

I've recently begun getting "Settling failed" warnings following some of my dithers which I never got before.  My Dither settings are:

{"pixels":0.5,"time":10,"timeout":45}

The PHD2 Log Viewer shows that immediately following the dither, the mount is sent a series of high amplitude corrections which pull both DEC and RA back to the X axis (or close to it.  In most of the "Settling failed" cases, about 40 seconds after the dither starts, one or both axes experience a sudden displacement for no apparent reason.  These anomalies are not present when settling succeeds.  What could be causing this?  Are there some dithering settings that I should change?

PHD2 Debug Log is here.

[Bruce Waddington]

We need to see both the guide and debug log files for one of these sessions.  What you sent was a guide log, not a debug log.  Please use the Help/Upload Logs feature in PHD2 to do the uploads.

https://openphdguiding.org/getting-help/

Regards,

Bruce

[Bill Richards]

Yes, that was a typo.  Both logs have been uploaded here.

 

- Bill

--
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/Iwj_lr70VUk/unsubscribe.
To unsubscribe from this group and all its topics, 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/6795ebb1-86dc-42b1-b8a8-bb93c9655cd3n%40googlegroups.com.

[Bruce Waddington]

Hi Bill.  I think the settling failures may be incidental to the underlying problem, they're just bringing it to your attention.  If we ignore the dithering activity, we can see there are numerous other situations where you get a sudden excursion in Dec that triggers a large Dec correction - quite often a reversal in Dec that includes a backlash compensation pulse (RA in red):

These Dec events happen at times when the Dec motor isn't running.  The large Dec guide pulse then triggers a corresponding excursion in RA, something we sometimes refer to as cross-talk although that's probably not a great label.  It can happen when there is enough inertial resistance on the Dec axis to cause the RA axis to "rock" by a small amount.  Both parts of this behavior are usually mechanical - static resistance in Dec and a bearing weakness in RA.  The time delay between the preceding dither and the following Dec excursion is often over a minute so I don't see a way the dither could be actually causing the problem unless you have a cable routing problem of some kind.  The pointing position here was probably unusual, Dec was 69 degrees.  This may put the system in an orientation that causes cabling to affect movement of the scope and certainly puts more emphasis on the balance of the OTA in Dec.

The earlier GA run shows something similar but there we have the extra information that no guiding was being done, the Dec motor was idle, and the RA drive was just running at the sidereal rate:

This suggests it may not be this mechanical "cross-talk" triggered by a large Dec correction but may simply be something external to the mount that is tugging on the guiding/imaging assembly. 

Sorry I can't be more specific, it's a bit of a head-scratcher for me.

Regards,

Bruce

[Bill Richards]

Bruce,

Thanks for the detailed analysis.  But I'm curious - is there a way to know for sure that the impulse is on the Dec axis and affecting the RA axis vs. the other way around?  Since the excursions occur simultaneously, how can one tell which is being affected first - or if it's both?

I had some cable "snagging" problems back in December that ruined a lot of images one night, so I spent a lot of time rerouting cables to ensure that there is no contact with anything else regardless of the mount position.  I power the ASI2600MC and EAF off the DC jack on the CEM40 saddle.  I use the USB jack at the base of the CEM40 that connects to a USB port in the saddle, which I then route to the ASI2600MC imaging camera.  I use the 2 USB ports on the ASI2600MC to connect to the ASI290MM guide camera and the EAF.  I use the cable channel that runs through the mount for the dew heater cable.  So there are no cables hanging off the payload whatsoever.  

This is what concerns me - I never had this problem before when I did have cables hanging off the payload.  But after one bad night when the cables were dangling a little too loosely, I improved the cable management and things improved.  Then after I sent the CEM40 in for servicing (because PHD2 calibrations were failing regularly), this behavior appeared.

In the past few weeks, I added a PowerBox Advance and cleaned up the cable management even more.  So the rig now looks like this:

After I took this photo, I realized that I'm no longer using the USB connection on the saddle to I can remove one of the USB cables attached to the base of the mount.  The other is iPolar, which could also be disconnected after polar alignment.  That would leave only the PBA power and USB running through the mount cable channel and I've made sur both have plenty of slack to cover 90 degree rotation in both directions.  

But honestly, I inspected the equipment closely several times during the previous session and none of the cables were contacting anything, so I'm skeptical that the issue is relate to cable snags.  My biggest fear is that there is something wrong with the mount and that iOptron messed it up when they "repaired" it last month.

 

- Bill

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/2cce672e-c389-4374-8026-b38066629e02n%40googlegroups.com.

[Bruce Waddington]

You make a good point, this is an unusual situation where the two deflections occur on the same guide camera frame so there's no way to know the order of events.  That also means that any guide correction isn't causing the other axis to move.  If you returned the mount for service, you must have removed the mount from the tripod, removed the plate and ring assembly from the mount, etc.  Is there any possibility that re-assembly left anything loose?  What about the fasteners for polar alignment?  But you're right, there's nothing here to rule out a problem in the mount brought about either by the repair or by shipping.   Isn't this hobby a joy?  It might be useful to do a really long GA run with a giant search region so you don't lose the star.  You caught one of these events in the earlier GA so it would be interesting to see if you can capture several of them.  That might make it easier to debate the issue with iOptron because there would be no guiding involved.

Good luck,

Bruce

[Bill Richards]

I was looking at the guide data a bit closer and noticed something that must be relevant.  

• 20:27:46  - 1st long guide run (a little over an hour): There are no random excursions on either axis - none.  During this period, I was imaging M81/M82 on the west side of the pier.
• 21:32:23  - 2nd guide run (~20 min): There is one huge anomaly at 21:36 which is most likely an exception.  During this time, I was still imaging M81/M82 on the west side of the pier.
• 21:55:32 - 3rd  guide run (~38 min): The excursions are present on pseudo-regular intervals.  Looking at my images and the NINA log, this guide run was immediately after the meridian flip.
• 22:36:21 - 4th guide run: We see the same behavior.
• 22:23:27 - 5th and 6th guide runs (very short): I was probably trying to solve this issue.
• 23:39:53 - 7th guide run (~52 min): We see the same behavior but it appears to be getting worse (more frequent).
• 00:35 :00 - I switched targets to M51, placing the telescope on the west side of the pier again.  
• 00:52:55 - 8th guide run (~31 min): There are still some random excursions, but they are less frequent.
• 01:25:00: Meridian flip took place.
• 01:26:04 - 9th guide run (~10 min): There are a couple anomalies.
• 01:39:07 - 10th guide run (~74 min): Littered with anomalous excursions, but some periods are clean and some are pretty bad.  
• 02:56:07 - 11th guide run (~127 min): Littered with anomalous excursions, but some periods are clean and some are pretty bad.  

There are periods of 15 minutes or more that are clean, and other times when the excursions occur regularly at ~4 minute intervals.  It's not perfectly periodic, but close enough that they tend to align well with my 2 minute exposures and dithering every 2 frames.  I'm not a big believer in coincidences so this pseudo-periodic behavior is suspicious to me.

And it seems that the situation is worse when the telescope is on the east side of the pier, regardless of the RA, which is inconsistent with cable snags.

I'm just baffled.

Regards,

Bill

[a_pet...@yahoo.com]

The problem may be the Dec axis is now free to 'float'  through its backlash .

Since " I never had this problem before when I did have cables hanging off the payload.",

the Cables may have been useful by loading and keeping the Dec gears engaged.

Try simulating cables with bungee or something = preload the Dec axis.

Alex

[Bill Richards]

One other thing I changed recently was fine-tuning my balance.  The counterweight bar on my CEM40 used to be very slightly off-center so the payload was always slightly east-heavy regardless of which side of the pier it was on.  I adjusted that so the payload was very nearly perfectly balanced.  So now I'm wondering if that may have done more harm than good.  I've read that it's best to have a slight east-heavy configuration to ensure there is always a little bit of drag on the RA axis, but other posts have said that's not really an issue with the belt-driven systems.

Tomorrow night, I'll try setting the counterweights such that there is always an east-heavy load and see if that makes a difference,

[a_pet...@yahoo.com]

Bill,

You Always want the RA set East Heavy .... Else it will drift uncontrollably within the backlash zone with the slightest tug of a cable or breeze.

( and drive the guide system crazy trying to compensate )

The gears need to be engaged and Doing Work ...

Alex

[Bill Richards]

Bruce & Alex,

We had pretty good conditions last night, allowing me the opportunity to run more tests and validate the theory that this was caused by my balancing being "too good".  I imaged M51 and M13 from 9PM to 4:30 AM and achived consistently good tracking accuracy, ranging from 0.6 to 0.9 arc-sec throughout the night.  There were a few instances of unexplainable excursions, but nothing like what I experienced last month when they are happening about every 4 minutes.  And when those excursions did happen they were almost always on just one axis, not both as we saw before.

Finally, regarding the original topic of this thread, I didn't experience any Dither Settling failures.

So thanks to both of you for your help and advice.  This matter appears to be resolved.

Bill

[Bill Richards]

I did have one question, though - how does one achieve an "east-heavy" payload before and after the meridian flip without adjusting the counterweights at the time of the meridian flip?  A lot of astrophotography rigs are remotely controlled (or the operators are comfortably snoozing) so there is no one around to adjust the balance at the meridian flip.  So how is that done?

Bill

[a_pet...@yahoo.com]

That is a Problem.. only way I know of is to use a hanging weight on the East side of the RA to keep a constant force bias .

.

What's REALLY Bad is to have the system West Heavy ..You get into the situation where the drive moves the mount 'westerly' ,

then the RA just drifts further West on its own due to the weight and the control system then needs to move it East !

Alex

[Bryan]

Bill

See https://groups.google.com/g/open-phd-guiding/c/8c1S27gnqdc/m/v3MZ11VKAAAJ

for techniques to being east heavy on both sides of the meridian.

Bryan