Mach1 tracking issue after meridian flip #ASCOM_V2_Driver #APCC


Roland Christen
 

Go to our website, technical support for legacy products. Click on Mach1. There are a number of PDFs for checking and setting backlash. There is also a section that covers other things besides mesh, and is covered in this PDF: https://astro-physics.info/tech_support/mounts/backlash2.pdf

This PDF describes both checking the tightness of the retaining nut on the spur gear, and at the bottom mentions tightening the worm end play nut, otherwise known as the worm bearing preload nut. Since this is a bit specialized depending on your version of the mount, it might be best to contact George here at AP during business hours, and he can walk you thru that.

Roland

-----Original Message-----
From: Leo Shatz <leonid.shatz@...>
To: main@ap-gto.groups.io
Sent: Mon, Jun 27, 2022 12:19 pm
Subject: Re: [ap-gto] Mach1 tracking issue after meridian flip #ASCOM_V2_Driver #APCC

Thank you, Roland and Ray for thoughtful suggestions. I'm preparing for my next imaging session and will be very busy in the next couple of days (and recovering from long travel and lack of sleep), but I'm definitely taking seriously your suggestions. What should I do to test the end play nut? I never done it before. I'd like to check it when things will be back to normal. I'll also check my previous sessions going through logs to check to see if anything similar happened more than once.

Thanks,

Leo

--
Roland Christen
Astro-Physics


Leo Shatz
 

Thank you, Roland and Ray for thoughtful suggestions. I'm preparing for my next imaging session and will be very busy in the next couple of days (and recovering from long travel and lack of sleep), but I'm definitely taking seriously your suggestions. What should I do to test the end play nut? I never done it before. I'd like to check it when things will be back to normal. I'll also check my previous sessions going through logs to check to see if anything similar happened more than once.

Thanks,

Leo


Roland Christen
 

Unless you look thru an eyepiece or set the camera on continuous mode at a fast cadence, you will never know what is happening. One cannot trouble shoot from my location here, one can only suggest what could be happening. And yes, if something mechanically is wrong in the mount, such as the end play nut being loose, you can easily have 15 seconds of no movement.

I write stuff in detail so that I can get people to understand more about their mount mechanically. When it comes to software issues, i suggest stuff only reluctantly and rather leave it up to you gurus.

Roland

-----Original Message-----
From: Ray Gralak <iogroups@...>
To: main@ap-gto.groups.io
Sent: Mon, Jun 27, 2022 11:22 am
Subject: Re: [ap-gto] Mach1 tracking issue after meridian flip #ASCOM_V2_Driver #APCC

Hi Roland,

I discounted the possibility of RA backlash because of the 15-second slew settle time after each slew. Is there still a way for that to occur given this set of events Leo said happened?

1. After meridian flip and settle time of 15 seconds, a plate solve was done and it was successful.
2. As part of precision pointing, a slew was issued followed by another 15 second settle timeout.
3. The second plate solve had failed.

The second plate-solve image is what Leo posted. Shouldn't 15 seconds of settle time be enough to take up any RA backlash?

-Ray








--
Roland Christen
Astro-Physics


Ray Gralak
 

Hi Roland,

I discounted the possibility of RA backlash because of the 15-second slew settle time after each slew. Is there still a way for that to occur given this set of events Leo said happened?

1. After meridian flip and settle time of 15 seconds, a plate solve was done and it was successful.
2. As part of precision pointing, a slew was issued followed by another 15 second settle timeout.
3. The second plate solve had failed.

The second plate-solve image is what Leo posted. Shouldn't 15 seconds of settle time be enough to take up any RA backlash?

-Ray


Roland Christen
 

This can happen if the RA direction is reversed and there is some backlash in the RA axis. It happens when you slew the mount toward the east. What happens then is that it takes a few seconds for the teeth to re-engage in the west direction. Until the worm teeth re-engage the RA axis will not be tracking at the sidereal rate and the stars will appear to drift westward. You may want to set a small backlash compensation setting for the RA axis. This will re-engage the teeth quickly after an eastward slew.

All mounts have backlash except for encoder mounts. Normally RA backlash is not noticed during tracking and guiding because those actions never reverse the axis. Therefore in RA the gear teeth always maintain contact in one direction and during guiding the axis simply slows down or speeds up to move the star back and forth. Not so during slewing where the gear teeth must reverse direction in order to slew the scope eastward. When the mount stops slewing, the teeth must then traverse the backlash distance at the very slow sidereal rate before tracking resumes. During this time the stars will be drifting westward.

Normal backlash time of reversal is usually 1 second or less fort he Mach1, but if the teeth are out of mesh it could be as much as 3 - 4 seconds. So the thing to do is to check the mesh once in a while and re-mesh the gears if they are too far out.

There is one other thing that can cause a large amount of reversal delay, other than mesh. And that would be the end-play nut that holds the worm in place. If that is loose, the worm wont turn the worm wheel in a timely manner when reversal of direction takes place.

Once you understand how the mount works, you won't be chasing red herrings down a blind alley. So it's important to do a bit of digging as to how the mechanics of mounts are constructed and operate.

Roland

-----Original Message-----
From: Leo Shatz <leonid.shatz@...>
To: main@ap-gto.groups.io
Sent: Mon, Jun 27, 2022 9:30 am
Subject: Re: [ap-gto] Mach1 tracking issue after meridian flip #ASCOM_V2_Driver #APCC

A screen shot of failed plate solve image 

--
Roland Christen
Astro-Physics


Leo Shatz
 

I hope you are right. I'm afraid to overtighten the scope rings so to avoid pinching its optics. It's either the scope was slightly adapting to new physical position, or there was a cable tug, as you said. I will keep watching if this issue repeats itself again.

Thank you!


Ray Gralak
 

The plate solve exposure was 6 second long.
Leo, this is probably not anything in the mount or software. The likely cause is something in the image train shifting.

-Ray


Leo Shatz
 

The plate solve exposure was 6 second long. I'm not sure what it the cause, the cables are all routed internally trough the Mach-1 mount. The sequence was restarted automatically in about 2 minutes time, and no more issues happened in the next hour and a half until end of the session. There could be some safety limit triggered by APCC, but I don't see any errors in APCC log. The target was at high altitude. I was shooting a time lapse with DSLR and caught the pier flip in long time exposure. I don't see double flips in that time lapse sequence.


Ray Gralak
 

BTW, nor is that stopped tracking. The trails would be longer. Look to something in your imaging train as the cause, including maybe a tugging cable

-Ray

-----Original Message-----
From: main@ap-gto.groups.io [mailto:main@ap-gto.groups.io] On Behalf Of Leo Shatz
Sent: Monday, June 27, 2022 7:30 AM
To: main@ap-gto.groups.io
Subject: Re: [ap-gto] Mach1 tracking issue after meridian flip #ASCOM_V2_Driver #APCC

A screen shot of failed plate solve image


Ray Gralak
 

That's definitely not a slew. Star trails would be much longer when slewing. It looks like possible settling in the image train.

-Ray

-----Original Message-----
From: main@ap-gto.groups.io [mailto:main@ap-gto.groups.io] On Behalf Of Leo Shatz
Sent: Monday, June 27, 2022 7:30 AM
To: main@ap-gto.groups.io
Subject: Re: [ap-gto] Mach1 tracking issue after meridian flip #ASCOM_V2_Driver #APCC

A screen shot of failed plate solve image


Leo Shatz
 

A screen shot of failed plate solve image 


Leo Shatz
 

On Mon, Jun 27, 2022 at 07:25 AM, Leo Shatz wrote:

"Error out of range" - this error is just a warning that first slew wasn't precise enough within specified limits. 

Also, I forgot to mention that I don't use pointing model, it's disabled in APCC.


Leo Shatz
 

Hello Ray,

I was also thinking that it could be a pier flip back, as the target was a 2-panel mosaic, but I have seen a message that mount finished slewing, just before failed plate solve. The clutches were fastened for good, as I usually do.

Attached (as a log file) is a very short section of the Voyager log showing partial sequence of events that I'm talking about.

There could be some hint in these specific lines of Voyager log, but I'm not sure what does it mean:

2022/06/25 03:34:40 388 - INFO      - [PrecisePointing              ] - [CHECK_POINTING_ERROR_Code                    ] - Pointing Error Spherical => 00 09 14.92 
2022/06/25 03:34:40 399 - INFO      - [PrecisePointing              ] - [CHECK_POINTING_ERROR_Code                    ] - Pointing Error Position Angle => 251 17 26.42 
2022/06/25 03:34:40 408 - WARNING   - [PrecisePointing              ] - [CHECK_POINTING_ERROR_Code                    ] - Error out of range [Max 00 00 18.00 ], retry another time [1]

Thanks!
 


Ray Gralak
 

Hi Leo,

I experienced a strange problem during the last session with Mach-1 and Voyager. After meridian flip and settle
time of 15 seconds, a plate solve was done and it was successful. As part of precision pointing, a slew was issued
followed by another 15 second settle timeout. The second plate solve had failed. I've inspected the image which
caused the plate solve to fail - it was showing that the mount was either slewing or stopped tracking.
It could be each correction was across the other side of the meridian limit, and each slew caused a pier-flip back to the other pier-side.

It could also be one of the clutches was slipping, or even a bad pointing model, if you were using one.

-Ray


Leo Shatz
 
Edited

Hello,
 
I experienced a strange problem during the last session with Mach-1 and Voyager. After meridian flip and settle time of 15 seconds, a plate solve was done and it was successful. As part of precision pointing, a slew was issued followed by another 15 second settle timeout. The second plate solve had failed. I've inspected the image which caused the plate solve to fail - it was showing that the mount was either slewing or stopped tracking. 
 
After a minute (60 seconds) the third plate solve was attempted and it failed again. The new plate solve image also showed that the mount still hasn't begun tracking.
 
That was followed by the Voyager starting a recovery sequence, including unparking, restarting guiding, etc, which was successful. 
 
I still wonder what went wrong. I can send APCC and Voyager logs for relevant time window, failed plate solve FITs, APCC settings, etc.

I'm using APCC Pro 1.9.4.3 and one of the latest firmware versions of CP4.
 
Thanks,
 
Leo