Multicamera automation

[Sp@ce_d]

I'm putting together a multishooter, 3 if all works out on a single mount. As clear skies seem to be getting rarer and rarer, the intention is to capture (L)RGB, HaSIIOIII either in one go or combinations and automate the whole process.

Now when I say automation, I mean the whole lot controlled under one program for unattended use. I know I can run several instances of Artemis capture, coupled with EqMod, Astrotortilla, PHD, CDC etc. That works, but I need to be there to kick it off monitor and shut it down.

So, whilst investigating the necessary software required I've discovered that there appears to be a lack of support for multiple camera's within the available software programs. More and more people seem to be interested in going down this route, so I'm starting a thread where we can investigate what can be done to achieve this.

There are several aspects to automation, I've stumbled at the first hurdle when I attempted to set up my camera's. It may also be that multi filter wheels, focussers, rotators etc also share the same fate.

My first foray was finding out what supports my focusing hardware (and telescope control, camera, guider, filter wheel etc.) Now, FocusMax's Aquirestar seems to be the weapon of choice so this led me down the path to MaximDL. It seems that this is now the ONLY way to interface a camera with FocusMax. Looking further I discovered that Automation software ie. CCD commander, ACP needs to use Maxim too. So I started out trialing Maxim. Unfortunately, Maxim will only talk to a maxim 'um (see what I did there.. ) of 2 camera's including the guide cam! It does appear to let me run another instance & hook up another image cam but I don't see that can be sync'd to the first instance. I suspect this will be the same issue for focussers, filterwheels & rotators too.

Is anyone else managing to do this? from what I understand the available options are:

Automation:

ACP

http://acp.dc3.com/index2.html

Requires MaxIm DL/CCD 5.0.

If you want auto focusing, you need FocusMax & only Maxim seems to talk to it now that CCDsoft is deprecated anyway.

CCD Commander

http://ccdcommander.com (needs Maxim for FocusMax)

Capture & Control (That works with automation software):

MaximDL

http://www.cyanogen.com/maxim_main.php

CCD Autopilot

http://www.ccdware.c...roducts/ccdap5/ (needs Maxim for FocusMax)

I have found an interesting approach that uses multiple computers running multiple copies of MaximDL http://astro.neutral...camera-docs.pdf but the cost of going down that route is spiralling way out of control!

In fact that's a very good point, would that also mean multiple copies of say ACP too!

OK, so what are you using and what works. How "automated" is it. Should I forget FocusMax? What other angles should I be approaching this from?

[ian_bird]

Hi

If your obsy computer has enough grunt - try vmware.

http://www.vmware.com/download/player/download.html

Cheers

Ian

[Sp@ce_d]

Hi

If your obsy computer has enough grunt - try vmware.

http://www.vmware.co...r/download.html

Cheers

Ian

Yes, most astro non processing (I run PI on a MacPro, now that flies) software isn't too resource hungry, so VM's wouldn't be a problem I don't think as long as the hardware / ports emulation copes ok. I've run stuff under Win VM's on the Macs in the past. I run the obsy on a netbook at the moment which copes fine using teamviewer back to the Mac. The hardware's not the issue for me really, the problem is syncing everything together under one piece of automation software for unattended use.

[Sp@ce_d]

I'm going to start adding info as I find it, hopefully it will make the thread a useful reference at least.. I don't expect to find a "One Ring To Rule Them All" so hope may be in finding the best compromise.

Sequence Generator Pro

http://www.mainsequencesoftware.com/

This looks interesting. Whilst it looks as if it too only supports one imaging camera, the license model is such that it covers installation on 3 personally owned systems. Also, you can manage your own licenses. Certainly a plus point in the seedy world of software licensing.. believe me!

Again, I haven't found any clue as to whether you can sequence dependencies between instances. Tthis is the key to the synchronous automation of multiple imagers. Also, focusing routines do not seem to be as powerful as Focussmax & maybe somewhat fiddly?

I'm just concentrating on what I see as the key downfalls of multiple device support (Cams, Filterwheels, focussers, rotators) when I'm initially checking things out at the moment. I'm taking it for granted that they will control a scope, a guider, plate solve etc.

Be good to hear feedback from users.

[Gina]

Good thread I'm using 3 instances of Artemis, PHD, EQMOD/ASCOM, AstroTortilla and CdC.. I have not yet sorted out auto-focussing. It's getting pretty complicated with a triple imaging system. It would be great if the whole thing could be controlled with one script. For this to work the software has to be modular with the possibility of control from other software. This principle is used by ASCOM but little astro software is fully ASCOM compliant. I might have tackled this as a software project in the past but unfortunately I seem to have lost too many "little grey cells"

[Sp@ce_d]

I know Gina. I had such an abundance of those "little grey cells" once too! Can't remember where I left them now! I wish I'd taken more interest in coding, so tempting to have a go but I fear time's probably against me on this one. Closest I get on the day job these days is a bit of unix scripting.

Sent from my iPhone using Tapatalk

[Sp@ce_d]

Oh just realised those bold headers in white don't show up in Tapatalk. I'll stop that then.

Sent from my iPhone using Tapatalk

[steppenwolf]

Although

requires more than one PC, Tom How does offer a fully synchronised system running in MaxIm DL so it perhaps has a place in this thread.

I am not convinced that synchronisation is that critical though - I successfully run a two imaging camera plus autoguider system from time to time and have never worried about synchronisation of the imaging cameras - but then, I don't insist on dithering my frames. Having recently upgraded from V4.58 to version 5.24 of MaxIm DL, I may run both imaging cameras from this software and guide via PHD which will give me a single program solution for the imaging part.

[JamesF]

I'd hazard a guess that the software writers in the main don't consider it a very high priority to support multiple cameras. I imagine that people using multi-camera systems aren't that large a proportion of their customers yet, so potentially it's a load of work just to support a few users who they might make more money from if they bought two licences anyhow. I'm struggling to find a compelling reason there

James

[Sp@ce_d]

Although

requires more than one PC, Tom How does offer a fully synchronised system running in MaxIm DL so it perhaps has a place in this thread.

I am not convinced that synchronisation is that critical though - I successfully run a two imaging camera plus autoguider system from time to time and have never worried about synchronisation of the imaging cameras - but then, I don't insist on dithering my frames. Having recently upgraded from V4.58 to version 5.24 of MaxIm DL, I may run both imaging cameras from this software and guide via PHD which will give me a single program solution for the imaging part.

Thanks Steve, so that's the system I found the documentation for in my web trawls. It does seem to offer the only way to do this I've found so far. However, it requires at least 2 copies of Maxim to run 3 imaging camera's from what I can see. An additional $499 just to use the capture function seems a bit steep to me.

Even if you don't dither, how would you sync automation control of each camera during auto focus, plate solving, guiding etc. The capture control still needs to be aware of whats going on?

[Sp@ce_d]

I'd hazard a guess that the software writers in the main don't consider it a very high priority to support multiple cameras. I imagine that people using multi-camera systems aren't that large a proportion of their customers yet, so potentially it's a load of work just to support a few users who they might make more money from if they bought two licences anyhow. I'm struggling to find a compelling reason there

James

Yes, it does seem that a lot of astro software development has its roots firmly back in "old" days of the 90's. I don't expect there is a massive demand that would cover the cost of redevelopment I agree, but with something like an additional module that already exists.. I wonder just how much more effort that really is?

I don't have a problem paying for additional bolt-ons. I do find it hard to swallow having to pay full wack for the use of one module though.

There are more people using more than one camera. Hmm.. I wonder what they use for the WASP settups?

[chrisshillito]

Yes, it does seem that a lot of astro software development has its roots firmly back in "old" days of the 90's. I don't expect there is a massive demand that would cover the cost of redevelopment I agree, but with something like an additional module that already exists.. I wonder just how much more effort that really is?

Changing code that is structured to support a single instance to support multiple instances is a pretty fundamental change that would affect every part of the code. when it comes to software folks seem to forget that even a seemingly minor code change (which this isn't) can require significant effort in testing, and documentation changes (manuals, wiki, websites, video tutorial etc.) before it is suitable for public release. Clearly for a commercial product that effort has to justified against potential sales lost or gained.

I don't have a problem paying for additional bolt-ons. I do find it hard to swallow having to pay full wack for the use of one module though.

Perhaps you can negotiate a multi license discount if you contact the developer and explain what you're up to and why the existing licensing model doesn't work.

Alternatively, if you go the ASCOM route all the camera, focuser, mont functions are scriptable - you can always write your own automation software to glue it together and that meets your specific purpose. Being able to write your own software may require a significant learning curve but it is very empowering and cost effective in the long run

Chris

[Sp@ce_d]

Yes.. I'm investigating further with ACP & Maxim.

I may have to take a peek at some scripting at least but I've been trying to avoid taking on something I just know would eat up my life! I'd probably enjoy it but you know how these things lead on.

Any good pointers on starting out might be enough to temp me tho..

[ian_bird]

Come - let's be honest. Maxim can't even guide whilst it downloads. Multi-Threading anyone?

Sigh!

Cheers

Ian

[Gina]

Maxim sounds highly overpriced to me!

[Sp@ce_d]

Maxim sounds highly overpriced to me!

Well, it does appear so at this stage of my planning compared to alternatives for capture. The problem is it's a required component of Focusmax and ACP if I want to go down that route. I was a bit surprised when I first fired it up and looked at it, took me back to a long forgotten time in computing history..

However, it's the code that counts. I'm not interested in some glossy front end to a bug ridden program. I can put up with the looks if the code works. To be honest I'm very tempted with the ACP/Maxim/Focusmax route, (they are somewhat tied in together & appear to have a monopoly). It's just that I hadn't considered a lack of automated multi imager support when I planned the triple setup. I still need to check out alternatives but interoperability for an automated/unattended setup looks a bit thin on the ground.

I've made contact with Bob Denny at DC3 who tells me that ACP doesn't support multiple simultaneous imaging trains. However you can quickly switch configurations. I need to understand a bit more on that but doesn't sound too hopeful. I've also asked on the Maxim group if anyone has tried this. I'm sure there's some demand for it. I can't believe nobody else does this. Being able to grab say 3 hours worth of RGB an hour at a time, especially under our skies, makes sense doesn't it? Being able to automate it this way just seems obvious to me.

There's a possibility of multi instances of Maxim doing this but I'm not sure if this handles auto focus, filterwheel etc too. Tom How has sent me some info.

http://astro.neutral...sync-maxim.html

http://astro.neutral...camera-docs.pdf

I need to check this out with ACP and I've asked Bob if he's come across anyone using this.

[perfrej]

I have performed some tests. It is easy to write software that connects to MaximDL and "subscribes" to event notifications. My test included a small application that got notification of the "start exposure" event from Maxim and then sent a signal to Nebulosity to shoot an image. It worked OK and did the job.

Now, as for full automation... ACP doesn't really have any support for multi-rig setup. It would, however, not be difficult to modify a few ACP scripts to fire off five other cameras at the same time as the one it is controlling. The problem is focusing. I can envision this as such:

• A PC runs ACP and controls mount, scope, dome, weather and one camera/focuser
  
• A virtual machine in the same physical PC runs MaximDL and FocusMax and handles one scope/camera/focuser
  
• Add one virtual machine per additional scope/camera/focuser
  
• In each of the virtual machines, a small application that receives commands over the virtual network controls MaximDL and Focusmax in that VM
  

Should work...

/p

Edited August 13, 2013 by perfrej

[Sp@ce_d]

I have performed some tests. It is easy to write software that connects to MaximDL and "subscribes" to event notifications. My test included a small application that got notification of the "start exposure" event from Maxim and then sent a signal to Nebulosity to shoot an image. It worked OK and did the job.

Ok, so the possibility to trigger other events (or at least capture) running on the same host looks viable. This sounds good from a licensing point of view and fairly simple (if one knows what one is doing.. ) in terms of running multiple captures.

Now, as for full automation... ACP doesn't really have any support for multi-rig setup. It would, however, not be difficult to modify a few ACP scripts to fire off five other cameras at the same time as the one it is controlling. The problem is focusing. I can envision this as such:

• A PC runs ACP and controls mount, scope, dome, weather and one camera/focuser
  
• A virtual machine in the same physical PC runs MaximDL and FocusMax and handles one scope/camera/focuser
  
• Add one virtual machine per additional scope/camera/focuser
  
• In each of the virtual machines, a small application that receives commands over the virtual network controls MaximDL and Focusmax in that VM
  

Should work...

/p

Hmm... automation would be nice.. I appreciate your input on this, so there is hope then. I had mentioned to Bob Denny earlier on about using Tom Hows similar method to this but he told me ACP's design didn't allow for simultaneous control.

You realise you've now opened yourself up to a whole load of questions.. I can handle the hardware, VM's & I've done some Unix scripting. Would you point me in the right direction for getting up to speed on what I'll need to know & use to have a go at this?

It looks like it'll need the network stack & has to run as a separate host as such then. Thats a shame, I had found a post by Diffraction saying whilst they don't support multiple instances, license wise they didn't have a problem running them on the same host.

So, I take it this would need to use bi-directional comms/events rather than just sending trigger events? I would imagine feedback from focussing for example would be needed before an event trigger for exposure.

Right I'd better stop there.. if I could get some pointers to get up to speed so I can have a go that would be great..

Thanks..

Graham

[FraserClarke]

I'm not at all surprised that existing software doesn't support this. Afterall, the only common point in your system is the mount. it's not unreasonable that you'd need to get multiple copies of software to control the multiple copies of telescopes / focusers / filters / cameras that you have... However, it's a good problem and I agree though that it would be nice to have a single piece of controlling software, to help synchronize all the different systems.

Maxim allows you to connect to an "ASCOM" camera (and an "ASCOM" filter wheel, but not as far as I can see an ASCOM focuser). So I'd probably approach this by using a single copy of Maxim as the controlling programme, and developing an ASCOM driver which could handle multiple cameras. Then Maxim just deals with a single psuedo-camera, and the driver deals with all the complexity of farming out the commands/responses to the multiple real cameras. Filter wheel(s) could be similarly handled, though you need a link to know which camera has which filter wheel...

I don't know if you've written any ASCOM stuff before? It's pretty well structured and documented, and I found it fairly easy to develop the few modules I've written; though I imagine a programme like this would require extensive debugging.

Some assumptions/restrictions would could make it easier;

• You're observing the same target with each telescope
  
• The cameras are multiple copies of the same hardware
  
• You're using the same exposure time in each filter
  

I certainly have a use to piggyback a widefield camera on our main telescope (which would break at least two of the above assumptions), so I'd be happy to contribute (if I have time!)

[Tom OD]

This a very interesting thread. I'm trying to set up along with a friend a dual rig, which will have the ability to be automated. Each scope and camera are the same, but we have different filter wheels. Per you have seen the set uo on the EM200 that we were trying.

I'm looking at purchasing the 10 Micron 1000HPS as it can handle the payload. I was very impressed by the demo you gave on it. So I'm guessing then that I would need to jump on board with writing this ASCOM script to be able to get the 2nd camera to focus and capture. The different filter wheels may be another problem altogether. I have not written s/w scripts or code since college, and I was no good at it. This could be fun. If any one has hints on where to start, please send me on some links.

Thanks Tom.

[Sp@ce_d]

I've only just returned from holiday, so I haven't had chance to look any further since my last post but I'm definitely up for having a crack at this. However, I've not done any ASCOM stuff before. My scripting has been limited to Unix but I'm prepared for the learning curve. Some guidance on selecting the right tools & direction from others more experienced would be good. It would save false starts & wasted time. I think this would be a worthy project & there seems to be enough interest for contribution, it just needs a cunning plan

[Magnum]

I’m reading this 6 years later, the Tom How Maxim plugin is still the only option I can find.
Lee 

[Starflyer]

I thought I'd heard APT had multi camera support these days.

[bottletopburly]

5 hours ago, Starflyer said:

I thought I'd heard APT had multi camera support these days.

Yes it does https://ideiki.com/astro/usersguide/multi_camera_operation.htm

[Fellside]

Hi

There is an Custom Array version of Voyager, but it is not generally available. Maybe contact the developer Leo to discuss availability.

It is mentioned in the Voyager Wiki  https://voyager.tourstar.net/index.php?title=Array

There is a discussion in this thread  https://forum.starkeeper.it/t/support-for-second-imaging-camera/589

Regards

Graham