Jump to content


Photo

How long 'til....?

  • Please log in to reply
14 replies to this topic

#1 buckeye_hunter

buckeye_hunter

    Mariner 2

  • *****
  • Posts: 203
  • Joined: 07 Dec 2008
  • Loc: Ohio

Posted 05 March 2009 - 01:00 PM

encoders will be obsolete and GPS will allow you to simply attach a transmitter/receiver to your scope tube to find your way around the night sky?

Just a thought...

Clear Skies and Peace Out!

Bob

#2 waassaabee

waassaabee

    Fly Me to the Moon

  • *****
  • Posts: 5116
  • Joined: 26 Nov 2007
  • Loc: Central California Coast

Posted 05 March 2009 - 01:20 PM

To be economically feasible? For the hobbyist? Probably not any time soon...

#3 Luigi

Luigi

    Fly Me to the Moon

  • *****
  • Posts: 5320
  • Joined: 03 Jul 2007
  • Loc: MA

Posted 05 March 2009 - 02:17 PM

Never? GPS can't tell the direction something is pointing, only its position. Things like the Skyscout uses an eletronic magnetic compass and level to figure out where it's pointing.

#4 divers

divers

    Viking 1

  • *****
  • Posts: 577
  • Joined: 21 Jan 2008
  • Loc: Apopka, FL

Posted 05 March 2009 - 02:45 PM

Luigi - I would not be overly suprised to see a company combine just what you are saying into a telescope. Just turn the unit on and the mount calibrates itself. Of course I think to work the mount and telescope would have to be one single unit like an ETX. Always nice to speculate....I am still mind-boggled over my CGEM All-Star alignment.

#5 waassaabee

waassaabee

    Fly Me to the Moon

  • *****
  • Posts: 5116
  • Joined: 26 Nov 2007
  • Loc: Central California Coast

Posted 05 March 2009 - 03:02 PM

Luigi... Never say never. Accurate positioning is only the first part of the equation. What the mount could do with that information is quite simple. But the mount would need to 'see' the sky to accurately calculate star locations. Maybe a CCD/guidescope sort of a setup that communicates to the mount firmware what it's seeing vs. what it should be seeing sending correcting pulses to the motors?
Maybe it's not that far off...

#6 Kolenka

Kolenka

    Apollo

  • *****
  • Posts: 1181
  • Joined: 01 Jun 2008
  • Loc: Seattle Area, WA, USA

Posted 05 March 2009 - 04:03 PM

Luigi - I would not be overly suprised to see a company combine just what you are saying into a telescope. Just turn the unit on and the mount calibrates itself. Of course I think to work the mount and telescope would have to be one single unit like an ETX. Always nice to speculate....I am still mind-boggled over my CGEM All-Star alignment.


Uhm, ETX-LS?

#7 InterStellarGuy

InterStellarGuy

    Viking 1

  • *****
  • Posts: 561
  • Joined: 25 Jun 2008
  • Loc: Overland Park, KS

Posted 05 March 2009 - 04:06 PM

hahah Kolenka I was about to say the same thing. Thats exactly the description of the ETX-LS

#8 Luigi

Luigi

    Fly Me to the Moon

  • *****
  • Posts: 5320
  • Joined: 03 Jul 2007
  • Loc: MA

Posted 05 March 2009 - 04:24 PM

>>>I would not be overly suprised to see a company combine just what you are saying into a telescope<<<

Answer is it's already been done.

>>>Uhm, ETX-LS?<<<

The integral camera is what is used for final alignment.

What the original poster asked was if it can be done with GPS, and I said it can't. Cameras, compasses, and levels are not GPS. Perhaps a technical point, but it shouldn't be beyond the ken of most forum members.

the fact is, it could be done without GPS, cameras, or compasses. All that's needed is a camera and software. Think about it, anybody who knows the sky or is reasonably familiar with a planisphere can look at the night sky and figure where things are without needing to know where they are or the time(GPS), the direction (compass), or which way is up (level).

#9 buckeye_hunter

buckeye_hunter

    Mariner 2

  • *****
  • Posts: 203
  • Joined: 07 Dec 2008
  • Loc: Ohio

Posted 05 March 2009 - 05:18 PM

I understand what you are saying about position rather than direction, but what about two transmitter, one on the front of the scope and one near the focuser. Wouldn't the use of two of them give the ability to calculate direction?

Clear Skies and Peace Out!

Bob

#10 Kolenka

Kolenka

    Apollo

  • *****
  • Posts: 1181
  • Joined: 01 Jun 2008
  • Loc: Seattle Area, WA, USA

Posted 05 March 2009 - 06:19 PM

The integral camera is what is used for final alignment.

What the original poster asked was if it can be done with GPS, and I said it can't. Cameras, compasses, and levels are not GPS. Perhaps a technical point, but it shouldn't be beyond the ken of most forum members.

the fact is, it could be done without GPS, cameras, or compasses. All that's needed is a camera and software. Think about it, anybody who knows the sky or is reasonably familiar with a planisphere can look at the night sky and figure where things are without needing to know where they are or the time(GPS), the direction (compass), or which way is up (level).


I'd argue (as a programmer myself here) that the GPS, compass and level makes setting up much easier from a software perspective. We have a level of computing power in our brains that lets us map the sky, and intuitively find north, level and a rough time without actively thinking about it. In particular human brains have an astounding ability to abstractly identify things.

Using GPS, compass data, and level data, we can build that rough pointing model much in the same way a human brain does, by setting the expectations, and /then/ running plate solving to compare our expectations to reality in order to get a more precise pointing model (although the CCD needs to be good, and the skies/exposures somewhat long in order for full plate-solving to work). What I suspect the LS is doing is short exposures and centering the 'bright star' much like we already do when aligning a telescope. Otherwise the exposures would take awhile to get a good plate solve, especially in more light polluted skies. Centering the bright star is faster because of the short exposures you can use, but only works if you already have a rough pointing model.

None of this of course will really replace encoders.

#11 Luigi

Luigi

    Fly Me to the Moon

  • *****
  • Posts: 5320
  • Joined: 03 Jul 2007
  • Loc: MA

Posted 06 March 2009 - 08:32 AM

It can be done with only a camera. The software may be more of a project, but the question is what will the overall ROI be if you do it with hardware or do it with software.

>>>but what about two transmitter<<<

Only the satellites have transmitters. The things we use are receivers. Yes, two receivers place a distance apart could be used to determine direction, and that is an interesting proposition. I don't know how long the baseline would be to get the required accuracy using differential GPS. In this case the two recievers could be linked by wire, or linked by RF, in which case one is a transmitting, tho not GPS signals in the normal sense. Investigating what's being done with DGPS in surveying might provide the answer. Not cheap yet, but it could be in the future. BTW, the software for these devices isn't a walk in the park either.

If you could put one receiver on each end of the OTA, encoders wouldn't be necessary. Otherwise, they'd be necessary to keep track of where the OTA is pointing WRT to some arbitrary reference.


#12 rmollise

rmollise

    Hubble

  • *****
  • Posts: 15396
  • Joined: 06 Jul 2007

Posted 06 March 2009 - 10:25 AM

encoders will be obsolete and GPS will allow you to simply attach a transmitter/receiver to your scope tube to find your way around the night sky?

Just a thought...

Clear Skies and Peace Out!

Bob


Answer? Never. GPS cannot do anything but provide location and altitude and time without help. It can't even give speed unless the device in question is moving enough to develop decent velocity north and east from which speed over ground can be derived. To be much help aligning a scope you need, in addition to GPS, accelerometers and a compass. Best, simplest, and cheapest tool for the job, and that is and will continue to be optical (or magnetic) encoders... ;)

#13 rmollise

rmollise

    Hubble

  • *****
  • Posts: 15396
  • Joined: 06 Jul 2007

Posted 06 March 2009 - 10:26 AM

Uhm, ETX-LS?


Nope, still used them dadgummed encoders. GPS gives it initial position and time info.

#14 Kolenka

Kolenka

    Apollo

  • *****
  • Posts: 1181
  • Joined: 01 Jun 2008
  • Loc: Seattle Area, WA, USA

Posted 06 March 2009 - 01:04 PM

It can be done with only a camera. The software may be more of a project, but the question is what will the overall ROI be if you do it with hardware or do it with software.


Yet, software isn't cheap because it is software. It is cheap because of how well it scales with the number of units you sell. That goes out the window the moment you have to go the route of upgrading the hardware to support your more complex software. I couldn't imagine trying to do a plate-solve on the old 12-20Mhz CPUs that tend to live in these mounts without it taking a toll on the user's wait time. And if you upgrade the electronics to cope with the more complex software, you are still adding per-unit costs.

It never is just about the unit cost, but also about the functionality being worth it to the user. Again, the big problem with plate-solving I see is the exposures required. I can pick up bright stars in exposures at a second or less. Plate solving will take much more time. So centering a bright star lets me do quick 1-second adjustments (with an even half-way decent algorithm for centering using a CCD). The minor per-unit costs of the GPS and accelerometers required to do the rough pointing model are actually pretty low. Last time I used them in prototyping, the accelerometers were on the order of a couple bucks for 1 set. GPS is getting pretty cheap these days as well (especially if you already have a CPU in your system to talk to).

I'm not saying it /can't/ be done. I am saying that right now, it doesn't make sense to do it that way. And I don't think the likes of Meade or Synta have the money to hire the quality of programmer who can make it feasible in this market in terms of per-unit cost. Programmers on that level are in demand, hard to get, even in a recession and cost 6 figures a year. You need to have good scales of economy going to scrape out any win there.

Nope, still used them dadgummed encoders. GPS gives it initial position and time info.


I wasn't saying they didn't, I was responding to the guy who was saying "I wouldn't be surprised if someone creates a telescope that you turn on and it aligns itself." That is kinda the goal of the ETX-LS right down to the ad copy. If you read my other post in this thread, I'm well aware of what GPS does. ;)

#15 Jared

Jared

    Fly Me to the Moon

  • *****
  • Posts: 5036
  • Joined: 11 Oct 2005
  • Loc: Piedmont, California, U.S.

Posted 06 March 2009 - 10:52 PM

It can be done with only a camera. The software may be more of a project, but the question is what will the overall ROI be if you do it with hardware or do it with software.

>>>but what about two transmitter<<<

Only the satellites have transmitters. The things we use are receivers. Yes, two receivers place a distance apart could be used to determine direction, and that is an interesting proposition. I don't know how long the baseline would be to get the required accuracy using differential GPS. In this case the two recievers could be linked by wire, or linked by RF, in which case one is a transmitting, tho not GPS signals in the normal sense. Investigating what's being done with DGPS in surveying might provide the answer. Not cheap yet, but it could be in the future. BTW, the software for these devices isn't a walk in the park either.

If you could put one receiver on each end of the OTA, encoders wouldn't be necessary. Otherwise, they'd be necessary to keep track of where the OTA is pointing WRT to some arbitrary reference.


You wouldn't be able to get the two GPS receivers far enough apart with a typical telescope. GPS simply can't provide any degree of accuracy when the distance separating the two receivers is a few tens of inches. Even the military grade GPS signal that has been available for commercial use since 2000 is only accurate +/- 5 meters. To get much more precise GPS locations would require all new GPS satellites as well as the use of much better oscillators in the GPS units. Not exactly commercially viable.

That being said, if you are just looking for a way that the telescope could completely align itself, there is no need to use GPS to do the pointing. Meade's announced LS technology is a perfect example.






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics