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

TPoint Paramount vs. AP

equipment mount
  • Please log in to reply
55 replies to this topic

#1 hyperion0001

hyperion0001

    Lift Off

  • -----
  • topic starter
  • Posts: 14
  • Joined: 26 Sep 2007
  • Loc: Concord, MA

Posted 17 April 2019 - 02:59 PM

This is a newbie question so I apologize ahead of time if  I may be making bad assumptions about things I don't quite understand. 

 

I hear TPoint mentioned often in relation to Paramount mounts, but more rarely in relation to Astrophysics mounts. I'm piecing together that TPoint is an algorithm that can be used with either type of mount, but it seems to be standard with only Paramount. Does the AP use a different algorithm for modeling the sky? I've read at least one instance where someone has used the SkyX program and TPoint to control an AP scope...is that necessary to improve on the pointing and/or tracking accuracy of an AP scope? 

 

terry

 



#2 leemr

leemr

    Ranger 4

  • -----
  • Posts: 348
  • Joined: 21 Mar 2014

Posted 17 April 2019 - 03:22 PM

I can't speak to the algorithms used in either of them, but one thing to note is that you can use TPoint with the SkyX to improve the pointing of any mount, but protrack, which improves the *tracking* of a mount based off a TPoint model only works with a Paramount mount. TPoint also provides a fantastic polar alignment solution that works with any mount.

 

APCC will do both pointing and tracking correction for Astro-Physics mounts.


  • hyperion0001 and psandelle like this

#3 Eric H

Eric H

    Messenger

  • *****
  • Posts: 400
  • Joined: 12 Feb 2010
  • Loc: Washougal, WA

Posted 17 April 2019 - 03:23 PM

T Point modeling is an add-on for the Sky X by Software Bisque. The Sky X comes standard with their mounts. You can control most other mounts with the Sky X and t point modeling.


Edited by Eric H, 17 April 2019 - 03:25 PM.

  • hyperion0001 likes this

#4 MJB87

MJB87

    Apollo

  • *****
  • Posts: 1306
  • Joined: 17 Feb 2014
  • Loc: Talbot County, MD

Posted 17 April 2019 - 03:41 PM

I operate a Astro-Physics mount using TheSkyX with the T-Point add-on.  I use T-Point to refine my pointing model with about 30 or so points in the model. Works just fine with the AP mount.



#5 pyrasanth

pyrasanth

    Surveyor 1

  • *****
  • Posts: 1537
  • Joined: 08 Jan 2016

Posted 17 April 2019 - 03:52 PM

I operate a Astro-Physics mount using TheSkyX with the T-Point add-on.  I use T-Point to refine my pointing model with about 30 or so points in the model. Works just fine with the AP mount.

Take it to 200 samples plus then use a supermodel- your accuracy will be vastly improved. I use a 600 point model (and I know 300 will suffice but since its fast-why not!). 30 points is barely adequate even though you will get good results by virtue of the AP being an exceptionally good mount.


  • hyperion0001, Ishtim and Gene3 like this

#6 Stelios

Stelios

    Fly Me to the Moon

  • *****
  • Moderators
  • Posts: 6884
  • Joined: 04 Oct 2003
  • Loc: West Hills, CA

Posted 17 April 2019 - 07:07 PM

Moved to Mounts for a better fit.


  • hyperion0001 likes this

#7 hyperion0001

hyperion0001

    Lift Off

  • -----
  • topic starter
  • Posts: 14
  • Joined: 26 Sep 2007
  • Loc: Concord, MA

Posted 18 April 2019 - 08:20 AM

Thank you for your answers, but I was more curious about why AP needs TPoint at all. Paramount seems to rely on the TPoint approach, whereas AP does not (by default) use it, and yet, both mounts are equally well regarded.

terry



#8 dhaval

dhaval

    Vendor

  • *****
  • Vendors
  • Posts: 1488
  • Joined: 21 Jul 2008
  • Loc: Round Rock, TX

Posted 18 April 2019 - 09:48 AM

Thank you for your answers, but I was more curious about why AP needs TPoint at all. Paramount seems to rely on the TPoint approach, whereas AP does not (by default) use it, and yet, both mounts are equally well regarded.

terry

Terry -

AP mounts don't need TPoint or any other pointing model for them to work. If I remember right from my time of owning the MyT, Paramounts don't need a pointing model as well to work. Having said that, Paramounts use the point model to correct tracking. AP mounts have their own version of TPoint in APCC Pro (APPM). APPM essentially does the same thing as TPoint, helps correct tracking errors resulting from a few different things. 

 

CS! 


  • hyperion0001 and psandelle like this

#9 CharlesW

CharlesW

    Long time member

  • ***--
  • Posts: 3471
  • Joined: 01 Nov 2012
  • Loc: Chula Vista, CA

Posted 18 April 2019 - 09:56 AM

You are losing sight of the fact that every mount needs an initial alignment. Software Bisque happens to call theirs, “TPoints.” Even if you polar aligned an AP mount, don’t think for a second that it would accurately point at anything. As mentioned above, TPoints will give you extremely accurate pointing, often with a one/two arcsecond error. And this is accomplished while you are in the house mixing a whiskey and Coke. It is also compatible with many mounts. Protrack on the other hand, is made from hundreds of plate solves all over the sky that tell a Paramount how to adjust tracking, instant by instant, which lessens the load on guiding. This is exclusive to Paramounts. This is done while you are drinking your whiskey and Coke. 


  • hyperion0001, psandelle, WesC and 2 others like this

#10 George N

George N

    Fly Me to the Moon

  • *****
  • Posts: 5209
  • Joined: 19 May 2006
  • Loc: Binghamton & Indian Lake NY

Posted 18 April 2019 - 10:46 AM

You are losing sight of the fact that every mount needs an initial alignment. Software Bisque happens to call theirs, “TPoints.” ….

"Tpoint" is a specific mount modeling system invented over 20 years ago by a professional astronomer in the UK - and first used on large professional telescopes - like the Hale 200". Bisque purchased a license to use the code and the name in TheSky - and have always offered it as an 'add-on'. As implemented by Bisque, "Tpoint" will greatly improve the pointing accuracy of any mount controlled with TheSky, but will improve tracking only on their mounts.

 

At least 10 years ago the UK guy was still selling a stand-alone DOS and LINUX version of Tpoint - that with certain mounts at least offered even more modeling features (but you had to manually create the model based on the pointing error data) than implemented by Bisque, including tracking improvements. I have no idea if it still is for sale separately.

 

Others ran with the idea - and developed their own error modeling scheme. Today it is hard to find mount control software, separate or built-in, that does not contain an effective pointing error model method.

 

Even the NexusDSC unit has error-modeling built in for Equatorial PushTo mounts, alt/az mounts, and Dobs -- which works best for a permanently set up scope of course. Who wants to spend time at a dark sky site to point to 200 stars to build a model? However, even a dozen pointings will help a lot.


Edited by George N, 18 April 2019 - 10:57 AM.

  • hyperion0001 likes this

#11 555aaa

555aaa

    Vendor (Xerxes Scientific)

  • *****
  • Vendors
  • Posts: 1264
  • Joined: 09 Aug 2016
  • Loc: Lynnwood, WA, USA

Posted 18 April 2019 - 02:14 PM

Prism also has a pointing improvement feature that looks like Tpoint - it uses model terms that have the same name, so I assume it is the same thing. It also will work on any telescope that Prism is controlling. Tpoint isn't modeling the sky, it is modeling all the defects in the mount and the optics. It has the terms for polar misalignment, for non-orthogonality of the RA and Dec axes, for droop in the OTA, for long-term encoder error (swash), and so forth. Atmospheric corrections for refraction aren't part of the model but are treated separately.


  • psandelle likes this

#12 SteveInNZ

SteveInNZ

    Viking 1

  • *****
  • Posts: 634
  • Joined: 07 Oct 2007
  • Loc: Auckland, New Zealand

Posted 18 April 2019 - 05:33 PM

 Tpoint isn't modeling the sky, it is modeling all the defects in the mount and the optics. It has the terms for polar misalignment, for non-orthogonality of the RA and Dec axes, for droop in the OTA, for long-term encoder error (swash), and so forth.

That would be interesting/useful information to have. I'm not sure how much I'm willing to pay to satisfy that curiosity though. Are there other tools that output that information ?

 

Steve



#13 rockstarbill

rockstarbill

    Fly Me to the Moon

  • *****
  • Posts: 5660
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 18 April 2019 - 06:24 PM

Don't confuse T Point and Pro Track. They are coupled, but Pro Track only works with Paramounts.
  • George N, hyperion0001 and psandelle like this

#14 DeanS

DeanS

    Skylab

  • *****
  • Posts: 4263
  • Joined: 12 Jul 2005
  • Loc: Central Kentucky

Posted 18 April 2019 - 08:18 PM

I have never used any pointing model with my AP mounts and it works just fine.  I do use PemPro for polar alignment which takes about as much time as a good T-point run.  I use the SkyX for plate solving when imaging.  

 

My point is don't think you MUST have a pointing model.   It helps but can live without it when using good mounts.


  • George N, hyperion0001, Peter in Reno and 2 others like this

#15 DS INC

DS INC

    Vostok 1

  • -----
  • Posts: 154
  • Joined: 17 Jun 2013

Posted 18 April 2019 - 08:27 PM

 Software Bisque happens to call theirs, “TPoints.”

Its "TPoint" and it is not needed to initialize a Paramount, that is what homing does  (MEII with OAE doesn't even need to be homed).

 

The latest iteration does not require a sync just a model but if you're polar aligned and just home the mount pointing will be good, TPoint will make it great.



#16 Peter in Reno

Peter in Reno

    Voyager 1

  • *****
  • Posts: 10258
  • Joined: 15 Jul 2008
  • Loc: Reno, NV

Posted 18 April 2019 - 09:10 PM

My only reason for creating pointing model is to use dual axes tracking for unguided imaging especially if the mount has absolute encoders. But to use pointing models just for pointing, all you need is plate solving. 

 

Peter


  • schmeah, hyperion0001 and psandelle like this

#17 Gene3

Gene3

    Viking 1

  • -----
  • Posts: 513
  • Joined: 02 Feb 2016
  • Loc: Del Mar, CA

Posted 19 April 2019 - 08:51 AM

You are losing sight of the fact that every mount needs an initial alignment. Software Bisque happens to call theirs, “TPoints.” Even if you polar aligned an AP mount, don’t think for a second that it would accurately point at anything. As mentioned above, TPoints will give you extremely accurate pointing, often with a one/two arcsecond error. And this is accomplished while you are in the house mixing a whiskey and Coke. It is also compatible with many mounts. Protrack on the other hand, is made from hundreds of plate solves all over the sky that tell a Paramount how to adjust tracking, instant by instant, which lessens the load on guiding. This is exclusive to Paramounts. This is done while you are drinking your whiskey and Coke. 

Hi Charles,

What is the minimum number of samples that Tpoint would need to collect in order to have Protrack work effectively? Just wondering if I need to nurse my W&C or guzzle it down.

Also does it make a difference if you bin at 2x2 vs 3x3 to collect the Tpoint samples? With my QHY16200a I get about 3 per minute at 2x2.



#18 MJB87

MJB87

    Apollo

  • *****
  • Posts: 1306
  • Joined: 17 Feb 2014
  • Loc: Talbot County, MD

Posted 19 April 2019 - 09:32 AM

Take it to 200 samples plus then use a supermodel- your accuracy will be vastly improved. I use a 600 point model (and I know 300 will suffice but since its fast-why not!). 30 points is barely adequate even though you will get good results by virtue of the AP being an exceptionally good mount.

Agreed. I will be doing a major reconfiguration of my setup in the coming weeks and will update the T-Point model at that point. I've never done the automatic modeling with the camera, just the manual one. Maybe it is time to do the automatic one.



#19 psandelle

psandelle

    Vanguard

  • *****
  • Posts: 2416
  • Joined: 18 Jun 2008
  • Loc: Los Angeles

Posted 19 April 2019 - 09:38 AM

Agreed. I will be doing a major reconfiguration of my setup in the coming weeks and will update the T-Point model at that point. I've never done the automatic modeling with the camera, just the manual one. Maybe it is time to do the automatic one.

I've used modelling on Paramounts, APs  and 10Microns, and I gotta tell ya: AUTOMATE! You just sit back and watch your scope buzz around (don't let anyone near, or they might get knocked unconscious) snapping images and platesolving and when it's done...voila! If I had to do it manually, I'd probably have ripped my own head off years ago.

 

Paul


  • schmeah, hnau, Peter in Reno and 1 other like this

#20 WesC

WesC

    Fly Me to the Moon

  • *****
  • Posts: 5320
  • Joined: 06 Feb 2013
  • Loc: La Crescenta, CA

Posted 19 April 2019 - 12:30 PM

Hi Charles,

What is the minimum number of samples that Tpoint would need to collect in order to have Protrack work effectively? Just wondering if I need to nurse my W&C or guzzle it down.

Also does it make a difference if you bin at 2x2 vs 3x3 to collect the Tpoint samples? With my QHY16200a I get about 3 per minute at 2x2.

Patrick Wallace recommends 50 sample points for ProTrack to be effective. I do 75-80 points each session (I'm fully portable) and Protrack works great. However the more samples you give your model the better ProTrack will work.


  • Gene3 likes this

#21 555aaa

555aaa

    Vendor (Xerxes Scientific)

  • *****
  • Vendors
  • Posts: 1264
  • Joined: 09 Aug 2016
  • Loc: Lynnwood, WA, USA

Posted 19 April 2019 - 01:20 PM

If you want precision tracking but not pointing, with the right SW you don't need to make a model with lots of points. You only take samples along the trajectory that you are going to image on, which is the optimal sample set. So suppose you are going to image a target at RA = 13 hrs and dec = 30 degrees. What you do is do test images and plate solves at RA=13hrs, RA=12h30m, RA=12h, RA=11h30min, etc which are just along the trajectory that the mount is going to follow, and then the SW implements precision tracking along that path and side of pier because it knows the mount errors for that one trajectory and it doesn't care about some other part of the sky where you are not imaging. I know that I've seen SW that does this but for the life of me I can't recall what it is.

 

I have Tpoint for my observatory mount and I've played with it but so far all I do is a 1 star align and then I plate solve near my target and re-sync. I actually do a "zero star" zenith initialization by pointing the OTA at zenith and measuring that with a level or a tilt sensor, I do the initial sync to that, then I go to something in the neighborhood of my target, focus, plate solve, and then I go to the target. 

 

I have to say that one of my whinges is that we have variable axis rates in ASCOM but we don't have any way in ASCOM to move the model into the mount and there is no flag AFAIK to tell the planetarium SW if the mount is modeling or not. Normally variable rates are used to track comets and asteroids and shouldn't be used for modeling corrections because with modeling turned on, you want to report the on-sky coordinates and not the mechanical coordinates, but to make the model in the first place, you need to report mechanical coordinates. There's a "needs refraction" flag but no atmospheric data. If you want to do this via ASCOM you're better off making fake guide pulses to implement model-based precision tracking, and then of course you have to make that play nice with PEC if PEC is outside or inside the mount, and it could be in either place (PPEC for example is in PHD, not in the mount). That's one reason that you tend to have proprietary tracking correction solutions. OK rant over.



#22 CharlesW

CharlesW

    Long time member

  • ***--
  • Posts: 3471
  • Joined: 01 Nov 2012
  • Loc: Chula Vista, CA

Posted 19 April 2019 - 03:14 PM

Software Bisque recommends 45ish TPoints for Accurate Polar Alignment. This is easily done and takes about 20 minutes, before Astronomical Twilight. If I was setting up every night I would not bother with Protrack. I would just get my 45 in, APA, and start imaging. When I was portable I could leave my gear up for several weeks at a time and I did do Protrack. You can reasonably expect to do 180 targets an hour so you only lose a few hours on the first night building a large model. 



#23 lucam

lucam

    Messenger

  • -----
  • Posts: 414
  • Joined: 09 Oct 2017
  • Loc: Upstate NY, USA

Posted 19 April 2019 - 03:28 PM

Software Bisque recommends 45ish TPoints for Accurate Polar Alignment. This is easily done and takes about 20 minutes, before Astronomical Twilight. If I was setting up every night I would not bother with Protrack. I would just get my 45 in, APA, and start imaging. When I was portable I could leave my gear up for several weeks at a time and I did do Protrack. You can reasonably expect to do 180 targets an hour so you only lose a few hours on the first night building a large model. 

If you are not going to use ProTrack and you are imaging, not visual observing jumping from target to target, why would you spend time building a model at all? When I image, plate solving won't take more than a couple of iterations to get perfect pointing to the target of interest. 

 

I have APCC Pro with APPM and one of these nights I'll actually build a tracking model and see how much better guiding gets on the Mach1 but the utility of a pointing model has always been a bit of a mystery to me. I guess it would be nice at a outreach event so I can quickly zoom from target to target with the certainty that it will be centered in the eyepiece.


Edited by lucam, 19 April 2019 - 03:29 PM.

  • psandelle likes this

#24 pyrasanth

pyrasanth

    Surveyor 1

  • *****
  • Posts: 1537
  • Joined: 08 Jan 2016

Posted 19 April 2019 - 04:10 PM

If you are not going to use ProTrack and you are imaging, not visual observing jumping from target to target, why would you spend time building a model at all? When I image, plate solving won't take more than a couple of iterations to get perfect pointing to the target of interest. 

 

I have APCC Pro with APPM and one of these nights I'll actually build a tracking model and see how much better guiding gets on the Mach1 but the utility of a pointing model has always been a bit of a mystery to me. I guess it would be nice at a outreach event so I can quickly zoom from target to target with the certainty that it will be centered in the eyepiece.

You would bother because everything becomes better and easier.- especially guiding on a Paramount with Protrack enabled.



#25 lucam

lucam

    Messenger

  • -----
  • Posts: 414
  • Joined: 09 Oct 2017
  • Loc: Upstate NY, USA

Posted 19 April 2019 - 04:23 PM

You would bother because everything becomes better and easier.- especially guiding on a Paramount with Protrack enabled.

CharlesW in a post above said: "If I was setting up every night I would not bother with Protrack. I would just get my 45 in, APA, and start imaging." 

 

My question was, in his experience, what would the value of a limited pointing model be for imaging, if one doesn't use it for tracking correction. 

 

I get what you are saying that ProTrack might make guiding easier and more accurate. That's what I plan to test on my AP mount as well. It just seems that if one does not turn ProTrack on, there are better ways to use that twilight time than building a pointing model when plate solving is perfectly accurate. 




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





Also tagged with one or more of these keywords: equipment, mount



Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics