Jump to content


Photo

Pointing accuracy

  • Please log in to reply
54 replies to this topic

#1 inZet

inZet

    Explorer 1

  • -----
  • Posts: 99
  • Joined: 12 Feb 2009
  • Loc: Milan, Italy

Posted 03 September 2013 - 04:26 PM

How do you achieve max pointing precision using a GOTO telescope? Intended as a star in the dead center of the eyepiece, to avoid confusion.
I may be wrong, I'm not so convinced a good polar alignment will suffice.
Having swapped out the old Sphinx StarBook board with a NexSXD my pointing precision has improved a lot, but still within 30 arcsecs or so. If I set the camera to 640x480, I could miss the double. Visually, I can zoom out and in, swap the eyepieces and so on, no problem other than the doubt I observed the right double...

#2 JAT Observatory

JAT Observatory

    NOT a Wimp

  • *****
  • Posts: 9428
  • Joined: 20 Feb 2005
  • Loc: In the Primordial Soup

Posted 04 September 2013 - 07:50 AM

You need a pointing program like T-Point or Max Point .

#3 Eddgie

Eddgie

    Voyager 1

  • *****
  • Posts: 12488
  • Joined: 01 Feb 2006

Posted 04 September 2013 - 10:17 AM

Spend a lot of money on an AP or a Paramont or other high end mount. That is one way.

Pier mount your existing EQ head. I believe that of the pointing accuracy is related to flexure of the tripods, especially with so many people overloading mounts these days. Sure, you can use calibration stars, but done be surprised if stars that center in a 31mm Nagler are not centered in a 6mm Ortho. Just the weight of the eyepeiece can deflect a mount a few arc minutes. Even my CGE will deflect with an eyepiece change.

Putting the mount on a pier will take a lot of that away.

But these may be impractical. I mention them because if you really want perfect centering, that is the bar you need to get over.

Otherwise, what can you do?

Hypertune your mount. Replace the bearings with higher quality bearings and end-float the shafts properly.

Check the worms to ensure they are not bent, and check the end float here.

Lap the worms and worm wheels with a fine lapping compound. Of course this is painfully tedious, and this is the kind of thing that high end mounts get that you won't get from mass market mounts.

The lapping is necessary to ensure that you can get the most precise worm gear mesh to the worm wheel. The lash between these gears is a crucial issue for real high accuracy pointing and you do get some pointing inaccuracy, especially if you have light and heavy eyepeices.

Pre-load. Do not perfectly balance the mount. You want enough weight on the gear faces so that they always settle with the same faces in firm contact.

Use a stiffer dovetail setup. If you are using a light aluminum dovetail, try going to a heavy or solid one.

I also eliminate cone error. It is not supposed to matter because the calibration routines are supposed to correct for this, but I find it helpful.

Do not use vibration suppression pads. Asymetric loading can cause very slight deflections when the scope is in different orientations.

I mean you requirement is for arc-second accuracy, and to get it will be very painful.

Or very expensive. A paramount on a concrete pier would be easier because you are going to do all of these things above and still not get 5 arc second pointing accuracy.

Sorry Charlie.

#4 Spacetravelerx

Spacetravelerx

    Surveyor 1

  • -----
  • Posts: 1756
  • Joined: 23 Dec 2012
  • Loc: New Mexico

Posted 04 September 2013 - 10:37 AM

How do you achieve max pointing precision using a GOTO telescope? Intended as a star in the dead center of the eyepiece, to avoid confusion.
I may be wrong, I'm not so convinced a good polar alignment will suffice.
Having swapped out the old Sphinx StarBook board with a NexSXD my pointing precision has improved a lot, but still within 30 arcsecs or so. If I set the camera to 640x480, I could miss the double. Visually, I can zoom out and in, swap the eyepieces and so on, no problem other than the doubt I observed the right double...


inZet,

Basically Eddgie is spot on. However I would like to add the LX850 to the list. My goto's are spot on. Per the spec it is +/- 1 arcminute for gotos, but all of mine have been dead center. When I get back home I will post a video on this (I will use the MallinCam software with the target window to show this).

Of course this is not a cheap option either.

#5 jrcrilly

jrcrilly

    Refractor wienie no more

  • *****
  • Administrators
  • Posts: 33735
  • Joined: 30 Apr 2003
  • Loc: NE Ohio

Posted 04 September 2013 - 10:41 AM

http://sourceforge.n...illa/home/Home/

#6 rmollise

rmollise

    Hubble

  • *****
  • Posts: 15404
  • Joined: 06 Jul 2007

Posted 04 September 2013 - 11:20 AM

How do you achieve max pointing precision using a GOTO telescope? Intended as a star in the dead center of the eyepiece, to avoid confusion.
I may be wrong, I'm not so convinced a good polar alignment will suffice.
Having swapped out the old Sphinx StarBook board with a NexSXD my pointing precision has improved a lot, but still within 30 arcsecs or so. If I set the camera to 640x480, I could miss the double. Visually, I can zoom out and in, swap the eyepieces and so on, no problem other than the doubt I observed the right double...


The long and the short of it is it depends on which scope/mount you are talking about. Some go-to GEMs require a good polar alignment for good go-tos, and some don't, for example.

I would guess you mean 30 arc-minutes. 30 arc seconds would be pretty dadgum good.

The main secret with the Celestron system? Do all four cal stars and always finish centering with the up and right keys only.

#7 orlyandico

orlyandico

    Fly Me to the Moon

  • *****
  • Posts: 5284
  • Joined: 10 Aug 2009
  • Loc: Singapore

Posted 04 September 2013 - 11:31 AM

well as pointed out by others, backlash in the worm engagement, flexure of the tripod and optical tube, mirror flop if it's an SCT, will all contribute to inaccurate pointing.

the "best" way is to GoTo the desired star, plate-solve, calculate the error, than RCAL the mount. then do another GoTo. even Paramounts are required to do this in my experience (with GRAS/iTelescope) in order to obtain best pointing accuracy. Out of the gate, the Paramount ME "only" gets about 1' to 2' pointing accuracy, and that's with a very, very good polar alignment.

#8 Phil Sherman

Phil Sherman

    Surveyor 1

  • -----
  • Posts: 1505
  • Joined: 07 Dec 2010
  • Loc: Cleveland, Ohio

Posted 04 September 2013 - 11:56 AM

If you need to get absolute pointing accuracy, then you have at least three solutions.

1. Buy an AP or Paramount mount - lots of $$$.
2. Buy an Atlas or EQ6 mount and use EQMOD to point it.
3. Download, install and learn how to use AstroTortilla (a plate solving software suite) to tell you exactly where the scope is pointed. You'll need to take a short test exposure to pass to AT for analysis but it'll tell you exactly where you're pointed. If your scope is being run using the ASCOM platform, AT can move your scope to your intended target.

Phil

#9 WillCarney

WillCarney

    Apollo

  • *****
  • Posts: 1460
  • Joined: 08 Oct 2009
  • Loc: Bloomington, ILL

Posted 04 September 2013 - 12:58 PM

I recently found that once I balance the scope I never seem to set the scope in the exact park location. So on my Atlas I will do a one star alignment first then park and turn off. Then do a three star. This is much more accurate since the mount is parking in it's correct position. Of course proper polar alignment is important. Which I have struggled with.

#10 rmollise

rmollise

    Hubble

  • *****
  • Posts: 15404
  • Joined: 06 Jul 2007

Posted 04 September 2013 - 01:01 PM

well as pointed out by others, backlash in the worm engagement, flexure of the tripod and optical tube, mirror flop if it's an SCT, will all contribute to inaccurate pointing.

the "best" way is to GoTo the desired star, plate-solve, calculate the error, than RCAL the mount. then do another GoTo. even Paramounts are required to do this in my experience (with GRAS/iTelescope) in order to obtain best pointing accuracy. Out of the gate, the Paramount ME "only" gets about 1' to 2' pointing accuracy, and that's with a very, very good polar alignment.


That may help if very tight pointing accuracy is required. A normal a normal 2+4 alignment will put anything from one horizon to the other in the field of a medium power eyepiece or on the chip of my Mallincam Xtreme. "Up and right," BTW, is how Celestron takes backlash into account; fooling with gears is not required.

The ME doesn't have the calibration/cone stars that the Celestron and some other mounts implement. That why the Bisque mounts require a tight polar alignment and other measures if they are to have good go-to performance. ;)

#11 rmollise

rmollise

    Hubble

  • *****
  • Posts: 15404
  • Joined: 06 Jul 2007

Posted 04 September 2013 - 01:04 PM

I recently found that once I balance the scope I never seem to set the scope in the exact park location. So on my Atlas I will do a one star alignment first then park and turn off. Then do a three star. This is much more accurate since the mount is parking in it's correct position. Of course proper polar alignment is important. Which I have struggled with.


Parking/home position will affect how close the mount comes to its initial alignment stars, but will not affect final go-to accuracy.

#12 rmollise

rmollise

    Hubble

  • *****
  • Posts: 15404
  • Joined: 06 Jul 2007

Posted 04 September 2013 - 01:06 PM

If you need to get absolute pointing accuracy, then you have at least three solutions.

1. Buy an AP or Paramount mount - lots of $$$.
2. Buy an Atlas or EQ6 mount and use EQMOD to point it.
3. Download, install and learn how to use AstroTortilla (a plate solving software suite) to tell you exactly where the scope is pointed. You'll need to take a short test exposure to pass to AT for analysis but it'll tell you exactly where you're pointed. If your scope is being run using the ASCOM platform, AT can move your scope to your intended target.

Phil


1. Those are great mounts mechanically, but Celestron's go-to system is considerably more sophisticated.

2. EQMOD works very well, but so does Celestron's NexStar system.

3. For some applications that is the answer, but for me, the standard Celestron alignment is good enough in that it puts objects on my small chip cameras reliably.

;)

#13 Moromete

Moromete

    Viking 1

  • *****
  • Posts: 649
  • Joined: 15 Feb 2012
  • Loc: Romania

Posted 04 September 2013 - 01:14 PM

Rod is right. Unfortunately Celestron still has no competition when it comes to go-to pointing accuracy. Celestron software is still the most complex when using HC only. Skywatcher/Orion are still lagging behind in software go-to precision.

#14 Footbag

Footbag

    Fly Me to the Moon

  • *****
  • Posts: 5761
  • Joined: 13 Apr 2009
  • Loc: Scranton, PA

Posted 04 September 2013 - 01:21 PM

My CPC's pointing was considerably better then my AP Mach 1. The Celestron mounts have pointing models built into the firmware. To get that pointing accuracy out of a AP mount, you will need additional pointing software.

I never had any issue getting perfectly centered targets on my CPC. With my Mach 1, I use additional software to help me point at targets. The targets are close, just not smack in the center.

I do have a non-orthogonal system which impacts pointing with a system without a pointing model.

#15 Spacetravelerx

Spacetravelerx

    Surveyor 1

  • -----
  • Posts: 1756
  • Joined: 23 Dec 2012
  • Loc: New Mexico

Posted 04 September 2013 - 01:25 PM

Rod is right. Unfortunately Celestron still has no competition when it comes to go-to pointing accuracy. Celestron software is still the most complex when using HC only. Skywatcher/Orion are still lagging behind in software go-to precision.


Um, I have to dispute this claim "Celestron still has no competition...".

My LX850 provides excellent gotos (i.e. spot on), and I am pretty certain the same is true for the LX600. Like I said, once I am back home in New Mexico I will use my MallinCam to show how good the Meade setup is.

I cannot speak for all the other Meade units, but I simply have wonderful gotos on my LX200 and LXD75, though qualitatively speaking the LX850/LX600 are better. This is also one of many reasons I chose Celestron over Meade.

Anyways, stay tuned for the video later this month or October (the monsoons have been massive) - you will see what I mean.

#16 Spacetravelerx

Spacetravelerx

    Surveyor 1

  • -----
  • Posts: 1756
  • Joined: 23 Dec 2012
  • Loc: New Mexico

Posted 04 September 2013 - 01:28 PM

Rod is right. Unfortunately Celestron still has no competition when it comes to go-to pointing accuracy. Celestron software is still the most complex when using HC only. Skywatcher/Orion are still lagging behind in software go-to precision.


Um, I have to dispute this claim "Celestron still has no competition...".

My LX850 provides excellent gotos (i.e. spot on), and I am pretty certain the same is true for the LX600. Like I said, once I am back home in New Mexico I will use my MallinCam to show how good the Meade setup is.

I cannot speak for all the other Meade units, but I simply have wonderful gotos on my LX200 and LXD75, though qualitatively speaking the LX850/LX600 are better. This is also one of many reasons I chose Celestron over Meade.

Anyways, stay tuned for the video later this month or October (the monsoons have been massive) - you will see what I mean.



Oh and I forgot to add - the accurate gotos (spec is +/- 1 arcminute) are with the stock LX850/LX600. No need for a separate computer or software. "It just works"...

#17 jrcrilly

jrcrilly

    Refractor wienie no more

  • *****
  • Administrators
  • Posts: 33735
  • Joined: 30 Apr 2003
  • Loc: NE Ohio

Posted 04 September 2013 - 01:38 PM

Oh and I forgot to add - the accurate gotos (spec is +/- 1 arcminute) are with the stock LX850/LX600. No need for a separate computer or software. "It just works"...


It does work well, and it's so convenient. I'm sure that C will add this capability to their very affordable Starsense system soon - and I've been saying for a while that such capability will become an expected feature in midrange gear in the near future.

#18 Whichwayisnorth

Whichwayisnorth

    Apollo

  • *****
  • Posts: 1440
  • Joined: 04 Jul 2011
  • Loc: Southern California

Posted 04 September 2013 - 01:53 PM

I concur that the Starlock system puts objects dead center (or close enough). I think the theme of this thread is how can you take an existing mount that doesn't have Starlock and get accurate pointing. The answer has already been given. Make sure your hardware is tight, eliminate flexure, make sure your Cogswell Cogs and Spacely Space Sprockets are working the best they can and if all else fails get Astro Tortola. One more thing that was already mentioned and is very important. Something center at a short focal length will not necessarily be centered at a long one. For instance, my little 80mm refractor F/4.8 typically has targets in the center(or close enough) but when I use my F/10 SCT they not.

#19 rmollise

rmollise

    Hubble

  • *****
  • Posts: 15404
  • Joined: 06 Jul 2007

Posted 04 September 2013 - 02:22 PM


Um, I have to dispute this claim "Celestron still has no competition...".

My LX850 provides excellent gotos


He means "by telescope companies who are still in business."

JUST KIDDING. Maybe. :lol:

#20 Eddgie

Eddgie

    Voyager 1

  • *****
  • Posts: 12488
  • Joined: 01 Feb 2006

Posted 04 September 2013 - 02:59 PM

The OP said he is getting 30 arc seconds now, but does not satisfy him. I would be thrilled with 30 arc second pointing, but apparently he is not.

I think he is aiming at "dead center" meaning maybe +/- 5 arc seconds or so?

I don't know. 30 arc seconds is pretty good to me though.

I am lucky to get 10 arc minutes out of the CGE.

#21 jrcrilly

jrcrilly

    Refractor wienie no more

  • *****
  • Administrators
  • Posts: 33735
  • Joined: 30 Apr 2003
  • Loc: NE Ohio

Posted 04 September 2013 - 03:09 PM

The OP said he is getting 30 arc seconds now, but does not satisfy him. I would be thrilled with 30 arc second pointing, but apparently he is not.


He's apparently shooting at only 550mm so he'd never notice a 30 arcsecond error; that'd be only 15-20 pixels. I just presumed he got the unit of measure wrong.

#22 gdd

gdd

    Apollo

  • -----
  • Posts: 1398
  • Joined: 23 Nov 2005
  • Loc: N Seattle suburb, WA

Posted 04 September 2013 - 05:07 PM

Other than using pointing software, how about doing a GOTO an easily identifiable star near the intended target, sync to it, then do one more GOTO - would that get you close enough?

Gale

#23 inZet

inZet

    Explorer 1

  • -----
  • Posts: 99
  • Joined: 12 Feb 2009
  • Loc: Milan, Italy

Posted 04 September 2013 - 05:57 PM

Thank you all for the answers! Mods moved my post, I almost missed it... :-)
I mainly observe and measure double stars. To be sure I'm measuring THAT double a good pointing precision is important, otherwise errors can be performed.
My setup is very lightweight. I have a short tube (WO Megrez 90), rings, dovetail, ASI120 webcam, cables. 5 Kilograms payload at most, the Sphinx SXD specs are 15 Kg max.
I observe from my backyard, on grass. Tripod is HAL130, aluminium, it seems sturdy but who knows? I never thought about flextures. It may be a land subsidence, do you believe pieces of wood under the tips may help?
Balancing is a nightmare, axes are frictioned, they come very tight from Vixen. However, if possible I unbalance east to let the gears to always be in traction.
The SXD is not an HEQ5, it is supposed to be well-build (???), in fact I'm not aware of an hypertune: the weak link in the chain is the electronic, but replacing the StarBook with the NexSXD solves all the problems and transforms the mount in a Celestron AdvGT/CGE Pro. I've not bought the HC, I'm using NexRemote software to control the mount.
I don't have too much money to invest, AP is not a solution.
Even with the best mechanic I learned from you it's hard to achieve necessary precision; a plate solve is required. The question is if a solve can be found even if small fields are imaged.
The camera has 3.75 um pixels, imaging scale is 1.3 arcsec/pixels, near the scope's resolution limit.

#24 inZet

inZet

    Explorer 1

  • -----
  • Posts: 99
  • Joined: 12 Feb 2009
  • Loc: Milan, Italy

Posted 04 September 2013 - 06:14 PM

Good idea to sync on a near star. Because I usually observe nearby doubles, I should need to sync only one time per zone.

#25 jrcrilly

jrcrilly

    Refractor wienie no more

  • *****
  • Administrators
  • Posts: 33735
  • Joined: 30 Apr 2003
  • Loc: NE Ohio

Posted 04 September 2013 - 06:21 PM

Good idea to sync on a near star.


Just turn on the "Precise goto" function in the controller. That won't get you within 30 arcseconds but it will get you closer than you are now if you aren't using it - probably within an arcminute or so.






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics