Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Uh-oh, another PHD2 help request

  • Please log in to reply
11 replies to this topic

#1 GoFish

GoFish

    Apollo

  • *****
  • topic starter
  • Posts: 1123
  • Joined: 30 Nov 2016
  • Loc: Kentucky/Colorado

Posted 11 January 2019 - 09:26 AM

I just finished a breakdown, clean, lube, adjust of my Atlas (same as NEQ-6) mount.  Outwardly, the breakdown project seems to be a success.  Stiction is gone from both DEC and RA, it balances much easier, and slews smoothly.  (This is a used mount that needed some serious altitude adjustment mechanism repair when I got it, but before the breakdown it guided reasonably well, 1.2" - 1.5" RMS without too much effort).

 

I've adjusted the worm meshes to the best of my ability by going right up to motor stall, then backing off a smidge to elminate stalls.  Each worm float was adjusted carefully so that the worm turns freely, but there is zero axial movement.  I loaded the mount, then unloaded it and readjusted the worm mesh grub screws again as described above.  Not sure what else to do on that front.

 

PHD2 guiding last night was a disaster.  Rather than describe all of the symptoms, I have attached 2 log files.  In between the first and second logs, I balanced the RA axis a little bit more east heavy and slewed back a little further east of the meridian.

 

A very notable result from the calibration performed in the first log session was how far out of square, i.e. non-orthogonal, the RA and DEC movements were.  The calibration in the second log session was much better, and I had not experienced that non-orthogonality in calibrations before the recent service I did.

 

Each log consists of a calibration, guiding assistant run, then some guiding.

 

I'm not sure where to start looking for the cause of this bad performance, and recognize it could possibly be related to my guidescope and camera as well (50mm x 162mm Orion mini-guide, ASI178MM camera).  The guidescope is mounted on the standard Orion single stalk mounting.  The stalk mounting is fastened to an 8" dovetail which is mounted across the tops of the OTA rings.  The guide camera is pulled as close as possible to the rear of the 6" Newt to help with DEC balance.

 

Suggestions?

 

Attached File  PHD2_GuideLog_2019-01-10_201356.txt   35.37KB   5 downloads

 

Attached File  PHD2_GuideLog_2019-01-10_203818.txt   37.01KB   4 downloads

 

Calibration from the first log session:

Capture1.JPG

 

Tracking duing GA from the first log session

Capture2.JPG


Edited by GoFish, 11 January 2019 - 10:11 AM.


#2 zxx

zxx

    Viking 1

  • -----
  • Posts: 653
  • Joined: 27 Jan 2010

Posted 11 January 2019 - 11:06 AM

Looks like a lot of backlash in DEC  ,I have the CGEM DX that also has a lot of backlash in DEC . I don't bother getting a perfect PA ,I only use the polar scope that way i'm 

off a tad witch allows me to guide DEC in one direction witch gives me much better guiding 


  • GoFish likes this

#3 GoFish

GoFish

    Apollo

  • *****
  • topic starter
  • Posts: 1123
  • Joined: 30 Nov 2016
  • Loc: Kentucky/Colorado

Posted 11 January 2019 - 11:22 AM

Looks like a lot of backlash in DEC  ,I have the CGEM DX that also has a lot of backlash in DEC . I don't bother getting a perfect PA ,I only use the polar scope that way i'm 

off a tad witch allows me to guide DEC in one direction witch gives me much better guiding 

Yes, there is a lot of DEC backlash.  I saw much better DEC performance when I just turned off DEC guiding.

 

I am probably more concerned about the RA performance.



#4 zxx

zxx

    Viking 1

  • -----
  • Posts: 653
  • Joined: 27 Jan 2010

Posted 11 January 2019 - 11:43 AM

When guiding DEC in one direction with my DX it helps the RA smooth out on my mount , maybe you got your RA worm too tight ,a little slop in RA won't 

hurt as long as you are east heavy .


Edited by zxx, 11 January 2019 - 12:09 PM.


#5 GoFish

GoFish

    Apollo

  • *****
  • topic starter
  • Posts: 1123
  • Joined: 30 Nov 2016
  • Loc: Kentucky/Colorado

Posted 11 January 2019 - 01:20 PM

The attached image, from the second logging session, shows something I suspect is a clue.  During guiding assistant analysis, the RA axis moves to a huge tracking error of 30" during the first 2 minutes of unguided operation.  After that, it stabilizes somewhat.

 

Capture.JPG



#6 555aaa

555aaa

    Vendor (Xerxes Scientific)

  • *****
  • Vendors
  • Posts: 1144
  • Joined: 09 Aug 2016
  • Loc: Lynnwood, WA, USA

Posted 11 January 2019 - 03:17 PM

The Atlas has stepper motors which are used in microstepping mode when guiding (any time when tracking at siderial rate). The motor control is basically guessing where the motor is during microstepping and the higher the load, the worse the guess is. So having them run just below stall is probably not so good. In RA you can have more free play. I suggest to free up that RA worm mesh so there actually is a little free play. That will greatly lower the motor torque which should help a lot. In high torque there is also more belt tooth deformation and belt stretch.
  • zxx and GoFish like this

#7 KenS

KenS

    Viking 1

  • *****
  • Posts: 846
  • Joined: 10 Jan 2015
  • Loc: Melbourne, Australia

Posted 11 January 2019 - 04:35 PM

The attached image, from the second logging session, shows something I suspect is a clue.  During guiding assistant analysis, the RA axis moves to a huge tracking error of 30" during the first 2 minutes of unguided operation.  After that, it stabilizes somewhat.

That's normal. GA turns off guiding so it can do its analysis. During that time the underlying PE of the mount becomes visible and 30" p-p is typical for this class of mount.

The main issue I can see is that there is a strong (2") periodic response at around 10seconds. This is the tooth mesh frequency for the EQ6. That is the time for each tooth on the spur gears between the motor and worm  to engage and disengage. It indicates that the gears are not meshing well.


  • GoFish likes this

#8 GoFish

GoFish

    Apollo

  • *****
  • topic starter
  • Posts: 1123
  • Joined: 30 Nov 2016
  • Loc: Kentucky/Colorado

Posted 11 January 2019 - 04:57 PM

That's normal. GA turns off guiding so it can do its analysis. During that time the underlying PE of the mount becomes visible and 30" p-p is typical for this class of mount.

The main issue I can see is that there is a strong (2") periodic response at around 10seconds. This is the tooth mesh frequency for the EQ6. That is the time for each tooth on the spur gears between the motor and worm  to engage and disengage. It indicates that the gears are not meshing well.

There are 2 possible locations on each axis for poor spur gear mesh:

  1. On the stepper motor assembly, from the small gear to the larger idler
  2. Between the idler on the motor assembly and the spur gear inside the worm housing

Care to speculate which of these locations is responsible?

 

I will add that I purposely left a little lash between the worm spurs and the idlers because I had experienced motor stalling and I wasn't sure what was causing it.  I did not put much pressure on the motors when I adjusted their positions to mesh with the worm spur.  Maybe I left too much lash there?  I really don't know how tight that mesh should be.

 

That is a very tough adjustment to make, BTW, having only 2 hands and needing to peek through that little screw hole while adjusting the motor position!

 

(Can't thank you enough for replying and giving me some leads!.  This has been quite frustrating.)


Edited by GoFish, 11 January 2019 - 05:04 PM.


#9 GoFish

GoFish

    Apollo

  • *****
  • topic starter
  • Posts: 1123
  • Joined: 30 Nov 2016
  • Loc: Kentucky/Colorado

Posted 11 January 2019 - 05:01 PM

The Atlas has stepper motors which are used in microstepping mode when guiding (any time when tracking at siderial rate). The motor control is basically guessing where the motor is during microstepping and the higher the load, the worse the guess is. So having them run just below stall is probably not so good. In RA you can have more free play. I suggest to free up that RA worm mesh so there actually is a little free play. That will greatly lower the motor torque which should help a lot. In high torque there is also more belt tooth deformation and belt stretch.

If I understand correctly, you are saying the motors might be working against too much load due to tightness of the worm mesh in RA.

 

My vision of how this system works, though, is that loosening this mesh will add backlash to the RA drive.  This backlash wouldn't seem to be much of an issue when tracking unguided in only one direction, but it seems like negative guide pulses would get swallowed up by the backlash?  Or maybe PHD2 compensates for this during calibration?

 

Many thanks for taking the time to reply.



#10 KenS

KenS

    Viking 1

  • *****
  • Posts: 846
  • Joined: 10 Jan 2015
  • Loc: Melbourne, Australia

Posted 11 January 2019 - 05:11 PM

There are 2 possible locations on each axis for poor spur gear mesh:

  1. On the stepper motor assembly, from the small gear to the larger idler
  2. Between the idler on the motor assembly and the spur gear inside the worm housing

Care to speculate which of these locations is responsible?

 

I will add that I purposely left a little lash between the worm spurs and the idlers because I had experienced motor stalling and I wasn't sure what was causing it.  I did not put much pressure on the motors when I adjusted their positions to mesh with the worm spur.  Maybe I left too much lash there?  I really don't know how tight that mesh should be.

 

That is a very tough adjustment to make, BTW, having only 2 hands and needing to peak through that little screw hole while adjusting the motor position!

 

(Can't thank you enough for replying and giving me some leads!.  This has been quite frustrating.)

It could be either or both but the more likely is #2. That is adjusted by moving the motor housing. For #1 I think you need to remove the motor as the idler is attached to the motor housing. I've since installed a belt mod and adjusting the belt tension is equivalent to #2.

For these spur gears I don't think you need to allow any lash. Any binding would happen at the worm/ring gear interface. Happy to be corrected on that.  But if you loosen the clutch and spin the motor you would soon find out if there's an issue



#11 zxx

zxx

    Viking 1

  • -----
  • Posts: 653
  • Joined: 27 Jan 2010

Posted 11 January 2019 - 05:37 PM

If I understand correctly, you are saying the motors might be working against too much load due to tightness of the worm mesh in RA.

 

My vision of how this system works, though, is that loosening this mesh will add backlash to the RA drive.  This backlash wouldn't seem to be much of an issue when tracking unguided in only one direction, but it seems like negative guide pulses would get swallowed up by the backlash?  Or maybe PHD2 compensates for this during calibration?

 

Many thanks for taking the time to reply.

The RA motor always has a load on it  [ if your east heavy ] It  slows down or stops to let the star catch up .I don't believe it runs in reverse while guiding 

So a little play in RA won't hurt 


Edited by zxx, 11 January 2019 - 10:58 PM.

  • GoFish likes this

#12 GoFish

GoFish

    Apollo

  • *****
  • topic starter
  • Posts: 1123
  • Joined: 30 Nov 2016
  • Loc: Kentucky/Colorado

Posted 11 January 2019 - 07:05 PM

The RA motor always has a load on it  [ if your east heavy ] It  slows down or stops to let the star catch up .I don't believe it runs in reverse will guiding 

So a little play in RA won't hurt 

 

Makes sense. The guide rate is 0.5x, so worst case it would be running at half speed forward. 


  • 555aaa likes this


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics