Using Pi to auto-align a Goto mounted Telescope

[Scoobie2002]

Hello all,

I own a Telescope on a Goto mount ( Nexstar SE 6). I never got it to align with the sky. I wanted to use this as a project with my two kids. It's sadly collecting dust for some years now.

I read on this forum that some of you managed to use the Pi to direct the mount to a desired star and film/ take photo's with a Goto mounted telescope. In all cases the mount had to be pre aligned.

Since that is my problem, i still couldn't use the Goto mount.

Did any of you ever got a Goto mount to auto-align by use of the Pi?

Ever so grateful for a helping hand!

Scoobs

[Scoobie2002]

Hope someone is out there among you.....

[ame]

Did you try Google?

http://www.raspberrypi.org/forums/viewt ... &start=125

http://www.raspberrypi.org/controlling- ... thematica/
(from there you can get a PDF of the Nexstar protocol)

Finally, this guy is doing cool stuff:
http://blog.raceconditions.net/nexstar- ... pberry-pi/

[gordon77]

http://www.raspberrypi.org/forums/viewt ... &start=125

My work on telescopes mainly uses the ST-4 Autoguide port to Autoguide, although it should be able to guide a Meade LX200 via usb but never tested as l don't have one, so I don't think it will assist you in auto aligning.

Gordon77

[Scoobie2002]

Many thanks for your replies.

In my searcg for auto aligning i first checked the internet and this forum. These posts you mention, i've read a few times and are not answering what i'm after.

The trick is in the difference between auto guiding and auto aligning.

Guiding a telescope means that it will follow a eg a star in the sky whilst the earth is turning. Or when i select the moon, it will slew to the moon.

Before one can guide the telescope, it first wants to know where it is standing and what is pointed at. It's only from that point on, that it can steer to another body in the sky.
This process of establishing where it is at and what the lens is looking at, is called aligning.

The 'where' can be solved by gps coords. I think that won't be the problem. The 'what am i looking at' , is...
Nexstar developed a solution (costs about 400 pounds) that does just this. So it is possible... But is it also possible with a Rasp Pi?

Thank you for any help!

scoobs

[texy]

So basically either your alignment procedure is incorrect, or there is a fault with the telescope?
ie you shouldn't need any form of PC connection to the scope in order to align it, right?
Have you considered talking to a local astronomy club?
Texy

[gordon77]

Is this your scope?

http://starizona.com/acb/basics/using_a ... tarse.aspx

Does this process not work?

[P_Monty]

So basically either your alignment procedure is incorrect, or there is a fault with the telescope?
ie you shouldn't need any form of PC connection to the scope in order to align it, right?
Have you considered talking to a local astronomy club?
Texy

If I've got it right, th o/p wants to use a pi to automate the allignment procedure. If so, I'm curious to see how they get on as that's something I've been wondering about...

[texy]

Hmmm...that's not the way I read it

In all cases the mount had to be pre aligned.

I guess we will need to wait for the OP to reply..............
Texy

[Scoobie2002]

Is this your scope?

http://starizona.com/acb/basics/using_a ... tarse.aspx

Does this process not work?

This is indeed my setup. These steps don't work for me. 1-11 is ok. But step 12 and onwards ends up in an error message that it can't seem to align. The fact that lots of people can get it to work, makes me believe that it's my inexperience that is the the cause of it all.

Ofcourse... I could go to an astronomersclub or ask a specialist... After lottssss of trying to get to work myself and recently reading about guiding with a Rasp Pi as well as seeing that Nexstar developped it's own solution, what better that to try it with a Pi?

I read in a post that in the end it's the trick to automatically take a few pictures of the sky, identify lights as dots and compare those dots with a starmap. But i never read someone who actually pulled this off.

Scoobs

[gordon77]

Understood, if you have any links on how this is proposed to work please post them.

Gordon77

[Burngate]

I found this: http://www.raspberrypi.org/forums/viewt ... 43&t=96338

[EtonMess]

This is a tough ask, as aligning a goto scope is quite a thorough exercise to get it absolutely right.

Normally you are looking for a particular known object in the sky at your location and time. I will find it with a low power eyepiece, align it dead centre, then put in a more powerful eyepiece and align again. Then repeat for the next known sky object. Has to be spot on centre, or there will be error accumulating through scope movements.

The problem is, the camera or CCD will be dealing with a prick of white light in the field of view. Coding a gadget to do that job would be a very fine achievement indeed, in my opinion.

[thagrol]

I've also given this some though but have yet to do more than that.

As I understand it the process would go something like this:

1. Get GPS co-ords, date and time and send to handset.
2. Select alignment method (this may need to be done manually on the handset)
3. Select first aligment star/object
4. Take wide field photo through scope and analyse (google plate solving) to work out where scope is currently pointing.
5. Find co-ords of alignment star
6. Convert that into a more command for the scope and send it.
7. Take wide field photo through scope and analyse to work out where scope is currently pointing.
8. Repeat above steps 5-7 if scope is not pointing at alignment star.
9. Tell handset scope is aligned on target
10. Select second alignment star and repeat steps 4-9

Now the gotchas:

• Co-ords used by the motor controller do not have a fixed zero point. It's reset to where ever the mount is pointing at power on.
  The plate solving software I've looked at return co-ords in RA/Dec format. These would have to be converted into a form suitable for your Alt/Az mount, and possibly for the format needed by the motor controller board.
  The pi may not be powerfull enough to perform the plate solving in a reasonable time.

A Couple of possible work arounds:

• Putting the scope to a know position and zeroing the encoders before staring alignment may simplify some of the calculations. Meade were doing this in some of their scopes with their "level north" system.
  I think there's a thread on the forum somewhere about using an online service for plate solving.

If the nextstar protocol allows it, it may be possible to skip most of the process by calculating the initial poisiton, feeding that to the scope and telling it that it has been aligned. I don't have the docs to hand to check this though.

On the hardware front, on my skywatcher mounts (which also use the nextstar protocol), communication between mount, handset and attched computer is by serial. 5v ttl between mount and hanset but full RS232 levels between handset and computer. Neither of these can be connected directly to a pi.

[gordon77]

"Neither of these can be connected directly to a pi."

Could you use a Serial to USB adaptor/cable ?

Gordon77

[thagrol]

"Neither of these can be connected directly to a pi."

Could you use a Serial to USB adaptor/cable ?

Gordon77

Yep. Or a level shifter. What you can't do is connect them directly to the serial pins on the gpio connecter.

[Scoobie2002]

Thank you for all you’re comments.
I think that some sort of plate solving is the answer.

Using this fact as a startingpoint, it doesn’t matter where you’re scope is pointing, just that it’s pointing to some place in the sky where you can see the sky.

Next is the Plate solving problem. If read youre comments right, the Pi (B or B+) could lack some calculating power here. I read on the internet that some people tried this with a model A and failed (too slow).

I see five options:

1. Add an extra Pi to share the calculations somehow; is this possible?
2. I read about a Banana Pi that would have more possibilities; but is that enough?
3. Add an extra processor for the calculations; cool project also on these forums
4. Do the plate solving online.
5. Only let it calculate the biggest dots (filter all small dots out). But is this possible?

This last option is the most simple one as there is an API option that works quite well. The hold back could be that you always have to be near an internet connection.
The other options are just shots in the dark as i don’t know the first thing about these options.
I know for a fact that the conversion RS232 to USB works. I already got that working in an earlier project of mine (not with a Pi).
I love to hear you’re thoughts on these five options…

Many thanks,

Scoobs

[thagrol]

... it doesn’t matter where you’re scope is pointing, just that it’s pointing to some place in the sky where you can see the sky.

Not necessarily. If you're willing to duplicate the handset's databases and functionality on the pi then, yes. If you want to use the scope's handset for GoTos etc then you'll have to go through its alignment process aided by the pi.

The mount is most likely a two part system: the handset which has all the databases, calculation code etc and the mount which has only a motor controller board (with position feedback). Check the docs for your mount to be sure about this.

[Scoobie2002]

Thagrol, you are completely right.

I see that my goal isn't clear enough. The goal would be that i power the rig and the pi and it starts aligning automatically.
So i would love it ithe pi aligns the telescope; not aiding me in the aligning process.

I read the news today on the pi2. I think i'll try the new pi2. It should have enough calculation power for a reasonable plate solve time.
I'll post my findings when i start the project!

If someone has extra tips or hints on this thread please keep posting them as they could be very helpfull for me!

Thanks a million for all the helpn pointing me in the right direction!

Scoobs