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

Ball scope DSC

  • Please log in to reply
12 replies to this topic

#1 David Stevenson

David Stevenson

    Explorer 1

  • -----
  • topic starter
  • Posts: 68
  • Joined: 28 Oct 2019
  • Loc: Italy

Posted 14 August 2020 - 05:00 AM

Hello
I am thinking of building a ballscope, but I can't think of a way of having digital DSC, does anybody have a solution for this? Ideally it would be using accelerometer and magnetometers but I have no experience of this.
Is this doable anyway? Maybe something similar to a drone sensor usage.

Thanks

#2 Pierre Lemay

Pierre Lemay

    Surveyor 1

  • *****
  • Posts: 1,509
  • Joined: 30 Jan 2008
  • Loc: MontrĂ©al, Canada

Posted 14 August 2020 - 06:54 AM

David,

We've had many discussions over the years on this forum on that subject. Many theoretical solutions were proposed (computer mice regularly comes up) but for reasons too long to explain, none of them work.

 

So up until the end of last year there was no easy way to attach a coordinate location system to a ball scope. I know because I've been building ball scopes for thirty years and try as I might, I could not find a solution. But last year Celestron introduced the Starsense plate solving system which uses one's iPhone or Android camera and accelerometers to locate where in the sky the telescope is pointing. And IT works on ball scopes.

 

You can read more about this system in the CN thread: https://www.cloudyni...sense-explorer/.


  • markb likes this

#3 m. allan noah

m. allan noah

    Apollo

  • *****
  • Posts: 1,183
  • Joined: 14 Aug 2009
  • Loc: Virginia, USA

Posted 14 August 2020 - 07:03 AM

I actually have a couple of crude prototypes of ball scope DSCs. They use a camera inside the base, which looks up at the bottom of the ball. A computer then attempts to interpret the image, looking for a pattern on the bottom of the ball. I found that the precision was not great, and the process was pretty slow. But, it was absolute, so you did not have to worry about slewing speed. I bet Starsense would be easier :)



#4 petertinkerer

petertinkerer

    Explorer 1

  • -----
  • Posts: 73
  • Joined: 17 Mar 2018

Posted 14 August 2020 - 10:37 AM

Hello
I am thinking of building a ballscope, but I can't think of a way of having digital DSC, does anybody have a solution for this? Ideally it would be using accelerometer and magnetometers but I have no experience of this.
Is this doable anyway? Maybe something similar to a drone sensor usage.

Thanks

I developed the Tri-Finder system about 30 years ago and in many ways it is a manual version of the Starsense system that has recently been released.  It would certainly work on your ball scope.  A full description, with all the star maps, can be found on my website petertinkerer.com.  Below is a brief description which maybe a bit confusing but essentially it uses your brain and vision to "plate" solve the location of any DSO in the sky through triangulation using just two bright stars of 3rd magnetude or brighter.

 

Time moves on and although finding objects with the TRi-Finder is very satisfying, the fully automated, digital  version that exists today would be the easier way to proceed.

 

 

Brief description of the Tri-Finder:

 

 

This is the image you see which is superimposed on the sky. The central circle is where the binoculars are actually pointed and represents about a 1.5 degree field of view.  The 16 in line dots are separated by 1 degree increments and it is on this scale you place guide star # 1. The 3 vertical dots mark the 5, 10, and 15 degree points on the scale. Guide star #2 is placed on the direction line which in this image is about at 1 o'clock.  The direction line can be set anywhere from +70 to -70 degrees  and this whole pattern can be rotated by rotating the Tri-Finder disc.  By moving the binoculars in altitude and azimuth whilst simultaneously rotating the disc it can be made to "fit" the two guide stars, (it is a lot easier to do this than  to explain it). The location of any celestial object can be uniquely defined as an angular  distance  from a particular star and at a particular angular direction to another guide star.  Both eyes are kept open when using the Tri-Finder so it is easy to see the red "triangle" superimposed on the star field.  It is like a super red dot finder which enables you to see not only where the binoculars  are pointing but to actually point them to within 1 degree of any specific celestial object.   The advantage of Tri-Finder coordinates is that they are locked to the stars of the sky thus  they are independent of whether  you are in the northern or southern hemisphere, you do not need to know your location or the local time and the mount does not need any particular alignment to the horizontal or vertical.  You just open the tripod and start locating the desired celestial object.  To demonstrate the  format of the Tri-Finder coordinates consider M37 in Auriga, they would be:

M37, El Nath (6.9), Capella (-99),

which means M37 is located 6.9 degrees from El Nath  with an anti-clockwise  rotation angle of 99 degrees.  The negative sign in the direction angle means an anticlockwise rotation angle where as a positive direction angle means a clockwise rotation angle.  To use the Tri-Finder  to locate all the NGC objects you do have to learn where the basic constellations are and be able to identify about 40 of the brightest stars in the sky but that is what  makes it a great learning tool which so often the modern GOTO scopes fail to do.  However, to get started you just need to locate a few bright stars in the current evening sky and by using the appropriate star map you can start locating at least the main Messier objects within a few minutes.   

Attached Thumbnails

  • IMG_2335.JPG
  • Screen Shot 2020-08-14 at 11.11.54 AM.png

  • hamishbarker and Jakke like this

#5 hcf

hcf

    Viking 1

  • -----
  • Posts: 507
  • Joined: 01 Dec 2017

Posted 14 August 2020 - 02:26 PM

Hello
I am thinking of building a ballscope, but I can't think of a way of having digital DSC, does anybody have a solution for this? Ideally it would be using accelerometer and magnetometers but I have no experience of this.
Is this doable anyway? Maybe something similar to a drone sensor usage.

Thanks

Not a full blown traditional DSC, but this DIY project could help you  see on SkySafari where your scope is pointed to at any time. Once built it can be used on most scopes.

 

The code is on github.

 

https://www.cloudyni...sual-astronomy/



#6 David Stevenson

David Stevenson

    Explorer 1

  • -----
  • topic starter
  • Posts: 68
  • Joined: 28 Oct 2019
  • Loc: Italy

Posted 15 August 2020 - 03:02 PM

Googling a bit I found this:

https://lucidar.me/e...ino-9-axis-imu/

I am not expert but I guess this might be useful.

Watch the video also.
What you think?


Edited by David Stevenson, 15 August 2020 - 03:08 PM.


#7 hamishbarker

hamishbarker

    Messenger

  • -----
  • Posts: 496
  • Joined: 26 Apr 2017

Posted 15 August 2020 - 05:02 PM

Hello
I am thinking of building a ballscope, but I can't think of a way of having digital DSC, does anybody have a solution for this? Ideally it would be using accelerometer and magnetometers but I have no experience of this.
Is this doable anyway? Maybe something similar to a drone sensor usage.

Thanks

here's an idea for a universal one: a small guidescope and camera, fed to raspberry pi running kstars with a small touchscreen, with a dedicated button platesolving and outputting the required alt-az correction numbers? a bit slow perhaps but maybe not if it just uses a wide angle lens and bright stars. Celestron's new smartphone-guided starsense explorer beginner scopes use this, with the phone's own camera, a holder and diagonal mirror to redirect the camera view in the direction of the telescope while laying flat parallel to the optical axis.

 

I guess the simplest method could be to build a bracket and mirror same as the starsense explorer, and run the celestron software on your phone since it has already been written.



#8 hcf

hcf

    Viking 1

  • -----
  • Posts: 507
  • Joined: 01 Dec 2017

Posted 15 August 2020 - 06:18 PM

here's an idea for a universal one: a small guidescope and camera, fed to raspberry pi running kstars with a small touchscreen, with a dedicated button platesolving and outputting the required alt-az correction numbers? a bit slow perhaps but maybe not if it just uses a wide angle lens and bright stars.


That is sort of what the PSWAI from my earlier post is. It uses a copy of the astrometry.net software locally on the raspberry pi/linux laptop, an inexpensive action cam with a wide lens, and displays the platesolved location on Sky Safari Pro on a phone/tablet. Earlier versions used a wiimote to trigger the capture/platesolve, in the most recent version, you can start the capture/platesolve from Sky Safari Pro itself.


Edited by hcf, 15 August 2020 - 11:12 PM.

  • hamishbarker likes this

#9 hcf

hcf

    Viking 1

  • -----
  • Posts: 507
  • Joined: 01 Dec 2017

Posted 16 August 2020 - 10:34 AM

Googling a bit I found this:

https://lucidar.me/e...ino-9-axis-imu/

I am not expert but I guess this might be useful.

Watch the video also.
What you think?

Although I have not used the mpu-9250, people have tried to use cell phones , and other similar sensors with mixed results. Some swear by  these as finders, others (inlcuding me) have found them to be not  accurate enough consistently to be usable. Especially the magnetometer in the presence of iron tubes/parts.

 

Your best bet to try in this area, is the app SkEye on a phone. It can connect to some external sensors, so check if the mpu-9250 can be used.



#10 hamishbarker

hamishbarker

    Messenger

  • -----
  • Posts: 496
  • Joined: 26 Apr 2017

Posted 16 August 2020 - 03:44 PM

I developed the Tri-Finder system about 30 years ago and in many ways it is a manual version of the Starsense system that has recently been released.  It would certainly work on your ball scope.  A full description, with all the star maps, can be found on my website petertinkerer.com.  Below is a brief description which maybe a bit confusing but essentially it uses your brain and vision to "plate" solve the location of any DSO in the sky through triangulation using just two bright stars of 3rd magnetude or brighter.

 

Time moves on and although finding objects with the TRi-Finder is very satisfying, the fully automated, digital  version that exists today would be the easier way to proceed.

 

 

Brief description of the Tri-Finder:

 

 

This is the image you see which is superimposed on the sky. The central circle is where the binoculars are actually pointed and represents about a 1.5 degree field of view.  The 16 in line dots are separated by 1 degree increments and it is on this scale you place guide star # 1. The 3 vertical dots mark the 5, 10, and 15 degree points on the scale. Guide star #2 is placed on the direction line which in this image is about at 1 o'clock.  The direction line can be set anywhere from +70 to -70 degrees  and this whole pattern can be rotated by rotating the Tri-Finder disc.  By moving the binoculars in altitude and azimuth whilst simultaneously rotating the disc it can be made to "fit" the two guide stars, (it is a lot easier to do this than  to explain it). The location of any celestial object can be uniquely defined as an angular  distance  from a particular star and at a particular angular direction to another guide star.  Both eyes are kept open when using the Tri-Finder so it is easy to see the red "triangle" superimposed on the star field.  It is like a super red dot finder which enables you to see not only where the binoculars  are pointing but to actually point them to within 1 degree of any specific celestial object.   The advantage of Tri-Finder coordinates is that they are locked to the stars of the sky thus  they are independent of whether  you are in the northern or southern hemisphere, you do not need to know your location or the local time and the mount does not need any particular alignment to the horizontal or vertical.  You just open the tripod and start locating the desired celestial object.  To demonstrate the  format of the Tri-Finder coordinates consider M37 in Auriga, they would be:

M37, El Nath (6.9), Capella (-99),

which means M37 is located 6.9 degrees from El Nath  with an anti-clockwise  rotation angle of 99 degrees.  The negative sign in the direction angle means an anticlockwise rotation angle where as a positive direction angle means a clockwise rotation angle.  To use the Tri-Finder  to locate all the NGC objects you do have to learn where the basic constellations are and be able to identify about 40 of the brightest stars in the sky but that is what  makes it a great learning tool which so often the modern GOTO scopes fail to do.  However, to get started you just need to locate a few bright stars in the current evening sky and by using the appropriate star map you can start locating at least the main Messier objects within a few minutes.   

I really like this design.

 

Peter, is the lens arranged so that the leds are at the focal plane, so that the reflected image of them appears at infinity to the observer?



#11 petertinkerer

petertinkerer

    Explorer 1

  • -----
  • Posts: 73
  • Joined: 17 Mar 2018

Posted 16 August 2020 - 05:15 PM

I really like this design.

 

Peter, is the lens arranged so that the leds are at the focal plane, so that the reflected image of them appears at infinity to the observer?

Yes, the red image shown in the photo appears at infinity just like in a red dot finder or Telrad.  The Tri-Finder uses a light table illuminated by 4 LEDs  and the red dots are formed by small holes in the rotatable disc on the light table.  The  head section  of the Tri-Finder has a 125mm focal length plano-convex lens and a simple plexi-glass beam splitter and has a very wide field of view (40 degrees) if the eye is placed next to it, as if its an eyepiece.  There are always two bright stars that can be used to triangulate any desired DSO.  The star maps that can be printed out from petertinkerer.com show all the Tri-Finder coordinates of the Messier objects, and if anyone is interested in building one I have calculated the Tri-Finder coordinates of the entire NGC and IC catalogues.

 

Over the years I have always been surprised how easily the brain and eyes can solve what is quite a complex mathematical problem of moving the telescope in alt-azm whilst simultaneously rotating the Tri-Finder disc until the apparent red triangle in the sky fits the two bright guide stars and the DSO is always within 1 degree of the eyepiece FOV.

 

Peter


  • hamishbarker likes this

#12 chantepierre

chantepierre

    Vostok 1

  • -----
  • Posts: 149
  • Joined: 27 Mar 2020
  • Loc: France

Posted 17 August 2020 - 12:06 AM

That’s amazing ! I like the analog approach and will further read on this tri-finder method. Is there a guide or build log on your website ?

Edit : never mind, I found the PDF. Thanks for taking the time to document this.

Edited by chantepierre, 17 August 2020 - 12:11 AM.


#13 hamishbarker

hamishbarker

    Messenger

  • -----
  • Posts: 496
  • Joined: 26 Apr 2017

Posted 17 August 2020 - 07:14 AM

Yes, the red image shown in the photo appears at infinity just like in a red dot finder or Telrad.  The Tri-Finder uses a light table illuminated by 4 LEDs  and the red dots are formed by small holes in the rotatable disc on the light table.  The  head section  of the Tri-Finder has a 125mm focal length plano-convex lens and a simple plexi-glass beam splitter and has a very wide field of view (40 degrees) if the eye is placed next to it, as if its an eyepiece.  There are always two bright stars that can be used to triangulate any desired DSO.  The star maps that can be printed out from petertinkerer.com show all the Tri-Finder coordinates of the Messier objects, and if anyone is interested in building one I have calculated the Tri-Finder coordinates of the entire NGC and IC catalogues.

 

Over the years I have always been surprised how easily the brain and eyes can solve what is quite a complex mathematical problem of moving the telescope in alt-azm whilst simultaneously rotating the Tri-Finder disc until the apparent red triangle in the sky fits the two bright guide stars and the DSO is always within 1 degree of the eyepiece FOV.

 

Peter

perhaps worth adding a sketch of the optical layout to your excellent pdf. Thanks! I'm thinking of using plastic optical fibre from a toy light for illuminating the points of the trifinder light table. This will be a big help to find stuff with my big dob.




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