Jump to content


Photo

Open Source Mount Control Suite

  • Please log in to reply
96 replies to this topic

#1 Starhawk

Starhawk

    Space Ranger

  • *****
  • Posts: 5614
  • Joined: 16 Sep 2008
  • Loc: Tucson, Arizona

Posted 13 July 2013 - 12:26 PM

This came up on a thread in Cats & Casses. I am starting a thread here since it is more on topic, here.

I'm interested in this idea, myself. There isn't really anything especially magical coming up in the drives to make it impossible to figure this out if it were made to be able to talk to the right elements. So, here's a sketch off the top of my head of what it seems we'd want it to do:

Control capabilities:

(1) RA positioning drive
(2) RA Clock drive (there are architectures possible where 1 and 2 are separate)
(3) DEC positioning drive
(4) Telescope focus motor
(5) Dew heater
(6) Other motor 1 (e.g. if a guide scope were on a powered stage or camera rotator)
(7) Other motor 2 (e.g. if a guide scope were on a powered stage or camera rotator)
(8) Other motor 3 (Second focusing element)
(9) Dew heater 2
(10) Data output 1 (for pad or computer)
(11) Dome data output GEM (adjusted for as-built mount geometry, and dome data)
(12) Dome data output FORK (adjusted for as-built mount geometry and dome data)
(13) Roll off Roof coordination (e.g. ROR park function).

Accepted inputs:

(1) Bluetooth control- all data.
(2) WiFi hosting and control for all data.
(3) RA motor encoder.
(4) RA position encoder (several possible arrangements)
(5) DEC motor encoder.
(6) DEC position encoder (several possible arrangements)
(7) Focuser motor encoder
(8) Focuser absolute position encoder
(9) Dome position encoder
(10) Dome shutter open/close indicator
(11) System status serial commands
(12) System start control
(13) System normal shut down control
(14) Event-driven system safe/ emergency shut down control

There is going to be a much longer list to get everything. Something I see as an elephant in the room here is "What hardware does this run on?" The need to get the right control system together appears to have a great need for what a mount/ observatory system hub has to be. I personally see great benefit in this being one device. Ideally, it has these parameters:

(A) rock-solid stability- they just don't crash.
(B) remote-friendly capability- it resets itself to the right state if the power goes out.
© Modern data compatibility- I really want to be able to attack the rats' nest of cables.
(D) Mobility capability- it should be able to be really easy to use an built pointing models on the fly- all the good stuff.
(E) Low power requirement- even to the point of being active on a semi-old-school mount with just a clock drive and position encoders.

This doesn't sound at all like a Windows based system to me. I would really like to think something reasonable exists off the shelf which matches this description, though I don't know what that is off the top of my head. I don't know if any of the tablets we have seen up to now could become that device. That's really one for more knowledgeable folks.

Anyway, here's to the start of a thread. My understanding is ASCOM kind of started this way, so it may be possible for a 21st century universal mount and observatory controller to come out of this.

Thoughts?

-Rich



We need an open-source telescope. Probably most of the bits already exist.



#2 orlyandico

orlyandico

    Fly Me to the Moon

  • *****
  • Posts: 5724
  • Joined: 10 Aug 2009
  • Loc: Singapore

Posted 13 July 2013 - 02:25 PM

hi Rich,

There is something of this sort - RTS2 - http://www.rts2.org/

This was someone's doctoral dissertation and is now used at several professional telescopes.

On the Linux side there is INDI, which already has drivers for your favorite mount. :D

Or do you mean an open-source telescope mount firmware?

#3 cn register 5

cn register 5

    Viking 1

  • -----
  • Posts: 760
  • Joined: 26 Dec 2012

Posted 13 July 2013 - 02:47 PM

There's an argument that we have at least the basis of this in the form of Mel Bartel's scope control system. Open source and using easily available hardware.

Nowadays it may be better replacing the DOS based PC hardware with something like an Arduino and a couple of stepper control boards.

The challenge I see is the mount hardware. This is much more difficult for the average amateur to make - I know, I've tried! The only real source could be defunct commercial mounts.

I'd avoid adding too many bells and whistles at the hardware level, better to have separate modules handling things like focusers and dew heaters and only combine them at the application level.

Chris

#4 orlyandico

orlyandico

    Fly Me to the Moon

  • *****
  • Posts: 5724
  • Joined: 10 Aug 2009
  • Loc: Singapore

Posted 13 July 2013 - 03:13 PM

I didn't know that SCOPE.EXE was open-source...

#5 cn register 5

cn register 5

    Viking 1

  • -----
  • Posts: 760
  • Joined: 26 Dec 2012

Posted 13 July 2013 - 05:27 PM

Yes, GPL according to this:
http://www.bbastrode...m.html#DOWNLOAD

Chris

#6 Starhawk

Starhawk

    Space Ranger

  • *****
  • Posts: 5614
  • Joined: 16 Sep 2008
  • Loc: Tucson, Arizona

Posted 14 July 2013 - 11:47 AM

I specifically mean firmware and software. The idea I am thinking about is if someone can get basic mechanisms together such as bearings, motors, or something like an aeroquest gearing kit, it seems like it should be possible to make a first rate mount out of it. It should work in mobile mode out in the field, or in a remote observatory. It should be easy to set up and have the right reflexes to look after itself, such as power outage or loss of communications recovery.

It needs to be PC free. It needs to be completely wireless capable for all functions. It needs to be internet aware and secure. We're talking at least as smart as a $50 HP printer.

So, quite literally I'm talking about something you'd put together like Legos. Tell all the blocks what their job is and the obvious way to do it is the default, so it just works. They are looking to arrange themselves, so all you need to figure out is power. If you can arrange for a commercial swivel in your power cable, you're golden. If you power it down, drive 12 hours, and set up again, the system gets its act together on its own and tells you how to correct the polar alignment.

-Rich

#7 cn register 5

cn register 5

    Viking 1

  • -----
  • Posts: 760
  • Joined: 26 Dec 2012

Posted 14 July 2013 - 12:11 PM

Rich, have you any experience in designing or developing this sort of system?

Chris

#8 Starhawk

Starhawk

    Space Ranger

  • *****
  • Posts: 5614
  • Joined: 16 Sep 2008
  • Loc: Tucson, Arizona

Posted 14 July 2013 - 12:17 PM

Sort of- I've set up things like automated optical benches and things like that. Other people were writing the code, but I had to figure out what it had to do and if it was working.

-Rich

#9 cn register 5

cn register 5

    Viking 1

  • -----
  • Posts: 760
  • Joined: 26 Dec 2012

Posted 14 July 2013 - 01:52 PM

It's my impression that you are seriously underestimating the difficulty of what you want. All that auto configuration software is difficult and adding wifi won't help.

I know that HP can do it for a $50 printer but they are making printers in the millions and making their profit on the ink.

The thing to bear in mind is that the complexity cannot be reduced, all that can be done is move it around. You want to move it to the developer rather than the user so the development task is much more complex.

I've seen this at work, where we make microanalysis systems. Managers assume that a simple product that's easy to use will be simple to develop. The converse is true.

Chris

#10 orlyandico

orlyandico

    Fly Me to the Moon

  • *****
  • Posts: 5724
  • Joined: 10 Aug 2009
  • Loc: Singapore

Posted 14 July 2013 - 08:06 PM

Apparently it is doable - the closest commercial analog I can see is 10Micron - but is it doable by a hobbyist in their spare time...

Reality is "open source" is a "scratch my itch" kind of thing. If a commercial product scratches your itch at a reasonable cost (and in a non-"evil" way) then the motivation to build it yourself diminishes. I can see the desire to DIY as amusement.. but once you try to productize it the amusement rapidly vanishes.

Some similar projects - PICgoto and Soundstepper - they're doing interesting stuff.

#11 Starhawk

Starhawk

    Space Ranger

  • *****
  • Posts: 5614
  • Joined: 16 Sep 2008
  • Loc: Tucson, Arizona

Posted 15 July 2013 - 12:15 AM

I don't know what I said to imply the project was trivial. Rather, I see a lot of what we use in this hobby becoming inscrutable to the next generation of astronomers. I mean, we literally are at the point of having to use an emulator to run code to cause a USB plug to pretend to be a serial port to run and old DOS program trying to limp along in the 21st century. Getting a new mount to work is asking people to get familiar with COM when this will literally be the only COM device they will ever use.

So, once crossing the bridge to "OK, so I need to figure out a way to get this to be more like a bluetooth keyboard", it's pretty apparent everything else is going to have to be the same.

In a way, moving away from a PC is an attempt to simplify this- I don't know that it's realistic to think linking scopes to a given computer platform is going to help. So, the idea is to modernize the platform. Having it act a little more like a modern device would be nice for us. But for kids coming up, it's like explaining rotary dialing.

It's possible something like Sky Safari might try to grow into this if they had someone to talk to on the other side.

I feel like this is a real necessity. Of course, the fact COM support equipment is disappearing fast may force the mainstream here soon enough.

-Rich

It's my impression that you are seriously underestimating the difficulty of what you want. All that auto configuration software is difficult and adding wifi won't help.

I know that HP can do it for a $50 printer but they are making printers in the millions and making their profit on the ink.

The thing to bear in mind is that the complexity cannot be reduced, all that can be done is move it around. You want to move it to the developer rather than the user so the development task is much more complex.

I've seen this at work, where we make microanalysis systems. Managers assume that a simple product that's easy to use will be simple to develop. The converse is true.

Chris



#12 petely

petely

    Lift Off

  • -----
  • Posts: 15
  • Joined: 06 Jan 2011
  • Loc: England

Posted 15 July 2013 - 01:55 AM

It needs to be PC free. It needs to be completely wireless capable for all functions. It needs to be internet aware and secure. We're talking at least as smart as a $50 HP printer.

So, quite literally I'm talking about something you'd put together like Legos.

It all sounds eminently do-able as there's nothing in the list that is particularly hardware intensive. With all the $100 Android boards around: with WiFi, loads of I/O pins, serial connections and USB the Lego brick approach should be possible.

So for a visual observer, what you're proposing seems feasible. It would just be a case of someone with the skills wanting one for themselves and getting down to do it all.

#13 orlyandico

orlyandico

    Fly Me to the Moon

  • *****
  • Posts: 5724
  • Joined: 10 Aug 2009
  • Loc: Singapore

Posted 15 July 2013 - 03:21 AM

Anat has written some Linux code for guiding, etc. already

http://www.cloudynig...Number/5369851/

and also... http://www.raspberry...=15723&p=246412

he seems to have gotten quite a ways already.. he can guide via a QHY5 from the RasPi. The nice part - he already does plate solves and can sync the mount with the RA/DEC that's been plate solved. Pretty good... and consider a RasPi is $35.

project web site:
http://sourceforge.n...ects/iastrohub/

#14 Hilmi

Hilmi

    Soyuz

  • *****
  • Posts: 3841
  • Joined: 07 Mar 2010
  • Loc: Muscat, Sultanate of Oman

Posted 15 July 2013 - 03:33 AM

This is very doable, especially if there is a team of developers rather than one person. The Gemini I and Gemini II mount control system was written by one person and the Gemini II is based on an ARM processor if I am not mistaken. I believe that for such an initiative to truly take off it needs to be backed by a commercial venture, such as for example somebody selling the hardware in a ready to install pack that allows you to install the mount controller on various mounts without having machining skills. Another possibility is if it was adopted officially by one of the mount manufacturers who have good mounts but less than stellar controllers. This would give the initiative the momentum and credibility it needs to really pick up.

Don't forget that open source is fine as long as it doesn't inconvenience the user. Most users are not able to make their own hardware to allow such as system to work.

#15 orlyandico

orlyandico

    Fly Me to the Moon

  • *****
  • Posts: 5724
  • Joined: 10 Aug 2009
  • Loc: Singapore

Posted 15 July 2013 - 03:40 AM

There are a bunch of people already who sell 3rd-party mount controllers. Too many, in my opinion.

FS2
Boxdoerfer
Pulsar (on the Gemini G41/G53 mounts)
SiTech
Ursa Minor

and someone has already written Arduino code to emulate a Synscan motor board, so you can use an existing Synscan hand controller to drive your franken-mount.

plus, the NexSXW/NexSXD folks (Maite) have reverse-engineered the Celestron protocol to the point that they can fool a Celestron hand controller into controlling Vixen servo motors. It's just one more step to making a completely generic mount control system, i.e. by providing the motors.

I believe it is also possible albeit expensive to pay Software Bisque to put an MKS-5000 in whatever mount you want, on their web page they have some info on a Byers mount that was converted to MKS-3000 (?) so basically it's a Paramount.

IMHO with such a tiny market and with so many vendors/hobbyists there, I can't see the value of yet another project unless it is head-and-shoulders above the rest.

Everybody can make their own franken-mount controller, but to do it cheaply... and with a good feature set, is hard. Probably impossible.

Where I can see value:

- provide a good motor set and gears; everybody can buy Pittman motors, I'm talking about a Maxon motor, gearhead, and optical encoder. Something of the same class as an AP motor/gearhead.

- because the people who want cheap GoTo will buy a Celestron or a Synscan.

- a very robust controller to manage the motor gearbox.

Basically, something like the AP handset and GTO CP3, but totally generic. SiTech comes closest, but requires a PC.

#16 Geo.

Geo.

    Gemini

  • *****
  • Posts: 3059
  • Joined: 01 Oct 2008
  • Loc: Upstate NY

Posted 15 July 2013 - 04:59 PM

the NexSXW/NexSXD folks (Maite) have reverse-engineered the Celestron protocol to the point that they can fool a Celestron hand controller into controlling Vixen servo motors. It's just one more step to making a completely generic mount control system, i.e. by providing the motors.


Well, Vixen should sign them up and do a deal with Celestron (Synta seems OK selling the system as OEM). I hear the the reason the new StarBook is stepper based is the guy who developed the servo system walked and they felt they couldn't replace him.

As the average smart phone is more powerful than a 5 year old desktop and Celestron and Meade have moved to phone processors in their hand controllers, I'd say the PC is here to stay. Just in a lot smaller package.

#17 orlyandico

orlyandico

    Fly Me to the Moon

  • *****
  • Posts: 5724
  • Joined: 10 Aug 2009
  • Loc: Singapore

Posted 16 July 2013 - 01:17 AM

Post deleted by orlyandico

#18 orlyandico

orlyandico

    Fly Me to the Moon

  • *****
  • Posts: 5724
  • Joined: 10 Aug 2009
  • Loc: Singapore

Posted 16 July 2013 - 08:06 AM

Anat's iAstroHub software is a pretty complete and open source solution.

http://sourceforge.n...ects/iastrohub/

It runs on the RikoMagic Mk802/Mk808 android set top boxes. These are extremely cheap ($40 for the Mk808, $75 for the Mk802) "computers on a stick" that normally run Android, but iAstroHub requires that they run Linux.

The key weakness of any Linux based solution is very limited drivers for astro gear. INDI library supports LX200 (including Losmandy Gemini), AP GTO, Celestron, and some other controllers.

cccd image capture supports QHY cameras. Apogee and SBIG cameras are supported via XmCCD, which AFAIK doesn't work with iAstrohub at present.

lin_guider for guiding currently works best on the QHY5 with the onboard ST-4 port. GPUSB support is not available at present. I am thinking it should be straightforward to hack lin_guider to support LX200 pulseguide commands.

#19 Starhawk

Starhawk

    Space Ranger

  • *****
  • Posts: 5614
  • Joined: 16 Sep 2008
  • Loc: Tucson, Arizona

Posted 16 July 2013 - 11:44 AM

I do not disagree with any of this. My main vision is to enable something like your last line- but no PC.

I think Hilmi has a good point about needing a sort of sponsor. If someone were building a truly modular mount system, maybe that's what this gets launched as part of. I know iOptron has built controller/ motor sets to go with old mounts, like old CG-5s and such to turn them into GOTO mounts.

But here I would want it to be something good. Otherwise, why not just get a CGEM and live with whatever it does?

On the other hand, if it were good enough, it could, for example, be a major upgrade above the handpad on the Mach 1 GTO and its siblings with the ability to do a lot of what the AP command suite does, but in a mobile format. Hmmm, I wonder if I ran this concept by them if they'd tell me to buzz off or not. They didn't seem to understand what I was getting at with comments I made before about PC-free options.

The different little processor boxes floating around are definitely interesting to me, especially since so many are designed to be very low power.

-Rich

There are a bunch of people already who sell 3rd-party mount controllers. Too many, in my opinion.

FS2
Boxdoerfer
Pulsar (on the Gemini G41/G53 mounts)
SiTech
Ursa Minor

and someone has already written Arduino code to emulate a Synscan motor board, so you can use an existing Synscan hand controller to drive your franken-mount.

plus, the NexSXW/NexSXD folks (Maite) have reverse-engineered the Celestron protocol to the point that they can fool a Celestron hand controller into controlling Vixen servo motors. It's just one more step to making a completely generic mount control system, i.e. by providing the motors.

I believe it is also possible albeit expensive to pay Software Bisque to put an MKS-5000 in whatever mount you want, on their web page they have some info on a Byers mount that was converted to MKS-3000 (?) so basically it's a Paramount.

IMHO with such a tiny market and with so many vendors/hobbyists there, I can't see the value of yet another project unless it is head-and-shoulders above the rest.

Everybody can make their own franken-mount controller, but to do it cheaply... and with a good feature set, is hard. Probably impossible.

Where I can see value:

- provide a good motor set and gears; everybody can buy Pittman motors, I'm talking about a Maxon motor, gearhead, and optical encoder. Something of the same class as an AP motor/gearhead.

- because the people who want cheap GoTo will buy a Celestron or a Synscan.

- a very robust controller to manage the motor gearbox.

Basically, something like the AP handset and GTO CP3, but totally generic. SiTech comes closest, but requires a PC.



#20 orlyandico

orlyandico

    Fly Me to the Moon

  • *****
  • Posts: 5724
  • Joined: 10 Aug 2009
  • Loc: Singapore

Posted 16 July 2013 - 01:47 PM

Rich, as far as I understand, the smarts of the AP GTO controller is in the CP3 control box.

The handset is really just a fancy device for allowing input into the CP3 control box. In theory it should be possible to do everything APCC does, via a device that connects directly to the CP3. I know that the handset jack is really a serial port.

So it should be possible, via a bluetooth dongle attached to the first serial port, to have some sort of super controller using an Android tablet (iOS doesn't allow serial-over-bluetooth, the same reason Sky Safari for iOS cannot connect to a mount over bluetooth).

I mean let's get real.. a CP3 control box is $1000. If an end-user isn't willing to spend that much they're best off with a CGEM. The only issue is... AP won't sell you a CP3 or motors, unless you can prove you have an AP mount. So it's not a viable path for 3rd party mechanicals.

#21 orlyandico

orlyandico

    Fly Me to the Moon

  • *****
  • Posts: 5724
  • Joined: 10 Aug 2009
  • Loc: Singapore

Posted 17 July 2013 - 02:31 AM

I went through the last dozen pages of the SCT thread about Meade that spawned this thread...

The reality is it's possible to have an open-source mount control suite today. And for the mechanicals, any will do.

According to Dan Grey, the SiTech can be controlled without a PC. As I mentioned, Sky Safari can do it. None of this is open-source, however.

Incidentally, the Planewave Ascension is now bundled with SiTech. I could have sworn they used AP GTO before. That's a resounding plus vote for SiTech, I reckon..

How about this: the Arduino-based Synscan motor controller that I talked about, is compatible with EQMOD. With these two things, you have a full open-source solution. Requires a PC, however.

#22 Starhawk

Starhawk

    Space Ranger

  • *****
  • Posts: 5614
  • Joined: 16 Sep 2008
  • Loc: Tucson, Arizona

Posted 17 July 2013 - 09:53 AM

So, what is really driving my thinking is it is possible to independently do the following:

(1) add a drive to a given mount
(2) build a pointing model on-board a mount
(3) Control a mount with a phone or tablet
(4) automatically align a mount
(5) autoguide without a computer
(6) control a camera with a computer
(7) control a few cameras without a computer.


Leaving the PC out is huge in terms of complexity of setup for mobile operation. However, it's also becoming clear every update to windows is a threat to the whole shebang.

So, if a high end mount could be controlled to build a pointing model on the fly in the field without a PC, that's new. If the technique used would allow you to motorize a custom mount and do the same, then that's huge for ATM or for observatories, but even more so, would allow a far easier entry for new mount makers.

What these things don't do at all is present themselves to your mobile device as effortlessly as a Bluetooth keyboard. I feel if we can get mounts to that point, the hobby will be vastly more accessible to modern users. If it were an Arduino project, that definitely sounds like a neat idea to me.

Like you say, there is a lot of stuff around. And you can get a low end mount which makes its own pointing model. Or you can get a high end mount to do it- but only with a PC. and we're still married to Com, and I really am starting to wonder if even the rs-232 converters will be available in 5 years (like the 5.25" drive).

-Rich

#23 Hilmi

Hilmi

    Soyuz

  • *****
  • Posts: 3841
  • Joined: 07 Mar 2010
  • Loc: Muscat, Sultanate of Oman

Posted 17 July 2013 - 10:17 AM

Wll above features are actually available with various setups. Its just that nobody does it all. Meade got close and if they had added shutter release control with support for dithering they would have been spot on. But that's not open source

#24 Raginar

Raginar

    Fly Me to the Moon

  • *****
  • Posts: 6138
  • Joined: 19 Oct 2010
  • Loc: Rapid CIty, SD

Posted 17 July 2013 - 12:03 PM

So... you're basically making the Gemini-2? Maybe Rene would just let you help him out with that instead of reinventing the wheel. Much of what you're requesting is already done and working adequately.

#25 orlyandico

orlyandico

    Fly Me to the Moon

  • *****
  • Posts: 5724
  • Joined: 10 Aug 2009
  • Loc: Singapore

Posted 17 July 2013 - 03:02 PM

Thing is Rene has spent a good part of his life working on the hardware and software for Gemini 2 so it's extremely unlikely - impossible even - that he would open-source it.






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics