Array or Multi-instance?

I’m new to Voyager as I’m looking to automate my spectroscopy. My question relates to how multi-instance and array operates so I can determine whether either or both meet my needs.

My system is common for simple spectroscopy. I have a science camera to take spectra and a guide camera that guides off a slit mirror - similar to an off axis guider. The slightly unusual element is that my guider manages autofocus and pointing as well as guiding.

Like most software programs, Voyager assumes that the main imaging camera manages autofocus and pointing. Using the trial version, I can’t see how I can reallocate roles between the cameras.

The precision pointing system used by Voyager is excellent! I have a margin of error of 2 pixels in RA and my tests last night indicate that Voyager can handle this which is brilliant.

My thinking is to use 2 instances of Voyager - or the array version - so that I can focus and point with the first instance and the guide and image with the second. I’d use TSX Camera Add On in the first instance and Maxim in the second. Because pointing and guiding don’t occur at the same time and use different programs, I can connect the guide camera twice.

Key to this model is the ability to run both instances concurrently with the first focusing and pointing and then the second taking the science images.

I’ve looked at Master/slave systems previously. The slaved system waits for an image to be started on the master before starting to image on the slave. I need to be able to start imaging when pointing is done but without starting an image in the master.

Do either multi instance or array allow this?

Pete

1 Like

I’ve done a little more research on this - but have yet to reach a conclusion.

If I use the array option, I must set control guiding from the Master. So in the Master my guide camera would be connected as both imaging camera (using TSX Camera Add On) and guider (probably with TSX also or PHD2). I would then set the science camera as the imaging camera in the Slave - using Maxim DL.

I would like to establish a sequence whereby the Master does a precision pointing run and starts the guider and then instructs the Slave to take a science image - importantly, without the Master taking an image.

Is this possible?

Can I achieve the same objective with multi-instance? I suspect not as the sequences for each instance will run independently.

If neither work, then I am out of ideas.

Hope this clarifies my query

Pete

Hi Pete,

with array the master have the mount and the guiding system in charge, you cannot transfer in slave.
You can create a sequence where the master do not have any shot configured, the sequence is one to configure but the shot can be configured at node level. Focusing is for each node, and you can decide for each single node if disable it. So this will work for you (ask to support for an Array trial license). This is the sequence configuration form with evidenced what I wrote before:

About Multi Instance, there isn’t synchonization between.

If you are programmer you can use the Application server inside Voyager and the API to synchonize the operation between multi instance, this is another chance but not easy.

I hope this helps.

All the best
Leonardo

Leonardo

that is excellent news!

Sadly, I am not a programmer - far from it actually - so multi-instance is not really an option. But the Array set up will work for me.

So to confirm - I set my guide camera as both imager and guider in the Master set up. I point with the master and then start guiding without am image being taken with the Master. I then start imaging with the Slave. Both are run from the same Sequence.

If this works, this is the Holy Grail for spectroscopists like me - the ability to automate with off the shelf software.

I will need to integrate the ArNe calibration process and weather/observatory control but am confident that this can be done with your remarkable software.

I will give this a trial and report back.

Pete

BTW - here is my demonstration of the precision pointing feature. I set the target RA and Dec based on the offset required to place the target on the optimum position on the slit. I had set the Max Error at 3 arc seconds which is close enough for me. Ideally I’ll drop that again to 2 arc seconds which represents about 2.5 pixels - my slit is 4 pixels wide.

Pete

Hi Pete,

your image was not attached to the message, message edited.

All the best
Leonardo

Hi, I’d like to know how this got on.
I have now got a similar setup, where I have a 12". RC with spectroscope , fitted with reflective slit guide camera and main spectroscopy imaging camera and a 2nd scope with another camera aligned to the spectroscope guide and slit field. The telescopes are 12" f/9 and 6" f/7 respectively.
I’d really like to use robotarget to run continuous spectroscopy targeting reliably.

I could run the 2nd scope as master for pointing, solving and guiding but my current view is that the alignment between scopes is not sufficiently rigid to stay within 2-4 pixels across the sky as needed by the spectroscope slit.
I could run the 2nd scope for scope control , pointing and solving and use that voyager instance to also run the first scope guide camera. I then have the synchronisation issue of imaging between nodes and advanced doesn’t support array mode yet.
So I think the issue remains how to allocate and if needed, switch camera modes during a robotarget session. I’m thinking acquisition and pointing on the 2nd scope for rough pointing before fine pointing,guiding and acquring on the first scope .The reason for the fine pointing on the guide camera is that the slit camera field of view is 7’ square, making finding enough stars and being able to solve position quite a task.
All thoughts and input welcome
Cheers
Mike

Hi Mike

yes, I have been using Voyager for spectroscopy for about 18 months now.

This thread provides you with an outline of what I do.

Once you’ve read that, if you have any further questions or would like to see the system in action, drop me a message or email me Peter at oblaw dot net dot au

Pete

Cheers Pete, I’ll do that.