Minor changes to autoguider interface.

[tribeiro]

Replaced guiderStopped by guiderStop in order to retain compatibility with guiderStart action. Added "nlost" to configuration, the number of times guider will keep trying to guide before giving up when star is lost (for poor weather, for instance).

[wschoenell]

• What is the difference between nlost and max_tries?

• Do you need to know/use all this instruments/controllers? Can you reduce to a minimum? I think that we don't even have filterwheel instrument yet.

          "telescope": "/Telescope/0",
          "camera": "/Camera/0",
          "filterwheel": "/FilterWheel/0",
          "focuser": "/Focuser/0",
          "autofocus": "/Autofocus/0",
          "point_verify": "/PointVerify/0",

• What are Noffset and Eoffset? Are they the offset of the guider center from the focal plane center? In what unit?

[tribeiro]

nlost is the number of times the star can be lost before giving up on guiding max_tries is the number of times you try to look for a guiding star. Even though they may seem redundant they are actually not.

Well, yes... I think all this will be necessary. filterwheel does exists, but usually it is inherited by the camera as part of a major instrument. It is used like that on autofocus tough.

Yes Noffset and Eoffset are the offset of the auto guider with respect to the focal plane. They are supposed to be given in degrees.

[ankanaan]

in addition to Eoffset and Noffset I may see a need for PA, Position Angle of the center of the guiding CCD wrt to the science CCD.

On Fri, Jul 24, 2015 at 4:06 PM, Tiago notifications@github.com wrote:

nlost is the number of times the star can be lost before giving up on guiding max_tries is the number of times you try to look for a guiding star. Even though they may seem redundant they are actually not.

Well, yes... I think all this will be necessary. filterwheel does exists, but usually it is inherited by the camera as part of a major instrument. It is used like that on autofocus tough.

Yes Noffset and Eoffset are the offset of the auto guider with respect to the focal plane. They are supposed to be given in degrees.

— Reply to this email directly or view it on GitHub https://github.com/astroufsc/chimera/pull/88#issuecomment-124639497.

[wschoenell]

By position angle you mean a rotational angle wrt the north?

On Mon, Jul 27, 2015 at 2:44 PM Antonio Kanaan notifications@github.com wrote:

in addition to Eoffset and Noffset I may see a need for PA, Position Angle of the center of the guiding CCD wrt to the science CCD.

On Fri, Jul 24, 2015 at 4:06 PM, Tiago notifications@github.com wrote:

nlost is the number of times the star can be lost before giving up on guiding max_tries is the number of times you try to look for a guiding star. Even though they may seem redundant they are actually not.

Well, yes... I think all this will be necessary. filterwheel does exists, but usually it is inherited by the camera as part of a major instrument. It is used like that on autofocus tough.

Yes Noffset and Eoffset are the offset of the auto guider with respect to the focal plane. They are supposed to be given in degrees.

— Reply to this email directly or view it on GitHub https://github.com/astroufsc/chimera/pull/88#issuecomment-124639497.

— Reply to this email directly or view it on GitHub https://github.com/astroufsc/chimera/pull/88#issuecomment-125192394.

[tribeiro]

Now that gave me a good idea. Maybe we should move the offsets to the camera interface and it will deal with it automatically, when building the WCS. There is already a rotation I included for that purpose.

Em segunda-feira, 27 de julho de 2015, William Schoenell < notifications@github.com> escreveu:

By position angle you mean a rotational angle wrt the north?

On Mon, Jul 27, 2015 at 2:44 PM Antonio Kanaan <notifications@github.com javascript:_e(%7B%7D,'cvml','notifications@github.com');> wrote:

in addition to Eoffset and Noffset I may see a need for PA, Position Angle of the center of the guiding CCD wrt to the science CCD.

On Fri, Jul 24, 2015 at 4:06 PM, Tiago <notifications@github.com javascript:_e(%7B%7D,'cvml','notifications@github.com');> wrote:

nlost is the number of times the star can be lost before giving up on guiding max_tries is the number of times you try to look for a guiding star. Even though they may seem redundant they are actually not.

Well, yes... I think all this will be necessary. filterwheel does exists, but usually it is inherited by the camera as part of a major instrument. It is used like that on autofocus tough.

Yes Noffset and Eoffset are the offset of the auto guider with respect to the focal plane. They are supposed to be given in degrees.

— Reply to this email directly or view it on GitHub https://github.com/astroufsc/chimera/pull/88#issuecomment-124639497.

— Reply to this email directly or view it on GitHub https://github.com/astroufsc/chimera/pull/88#issuecomment-125192394.

— Reply to this email directly or view it on GitHub https://github.com/astroufsc/chimera/pull/88#issuecomment-125193532.

Sent from Mobile

[ankanaan]

yeap, by position angle, I mean position angle :-)

here is the defintion:

https://en.wikipedia.org/wiki/Position_angle

it is not only used for binary stars as the article seems to imply

On Mon, Jul 27, 2015 at 9:48 AM, William Schoenell <notifications@github.com

wrote:

By position angle you mean a rotational angle wrt the north?

On Mon, Jul 27, 2015 at 2:44 PM Antonio Kanaan notifications@github.com wrote:

in addition to Eoffset and Noffset I may see a need for PA, Position Angle of the center of the guiding CCD wrt to the science CCD.

On Fri, Jul 24, 2015 at 4:06 PM, Tiago notifications@github.com wrote:

nlost is the number of times the star can be lost before giving up on guiding max_tries is the number of times you try to look for a guiding star. Even though they may seem redundant they are actually not.

Well, yes... I think all this will be necessary. filterwheel does exists, but usually it is inherited by the camera as part of a major instrument. It is used like that on autofocus tough.

Yes Noffset and Eoffset are the offset of the auto guider with respect to the focal plane. They are supposed to be given in degrees.

— Reply to this email directly or view it on GitHub https://github.com/astroufsc/chimera/pull/88#issuecomment-124639497.

— Reply to this email directly or view it on GitHub https://github.com/astroufsc/chimera/pull/88#issuecomment-125192394.

— Reply to this email directly or view it on GitHub https://github.com/astroufsc/chimera/pull/88#issuecomment-125193532.

[wschoenell]

@ankanaan: Maybe I didn't understood about the PA (I just checked the figure), but I think that if you have an offset from east and another one from the north having the science camera as reference, you have the PA. @tribeiro: I was thinking on that when doing the question ;-)

On Mon, Jul 27, 2015 at 3:04 PM Antonio Kanaan notifications@github.com wrote:

yeap, by position angle, I mean position angle :-)

here is the defintion:

https://en.wikipedia.org/wiki/Position_angle

it is not only used for binary stars as the article seems to imply

On Mon, Jul 27, 2015 at 9:48 AM, William Schoenell < notifications@github.com

wrote:

By position angle you mean a rotational angle wrt the north?

On Mon, Jul 27, 2015 at 2:44 PM Antonio Kanaan <notifications@github.com

wrote:

in addition to Eoffset and Noffset I may see a need for PA, Position Angle of the center of the guiding CCD wrt to the science CCD.

On Fri, Jul 24, 2015 at 4:06 PM, Tiago notifications@github.com wrote:

nlost is the number of times the star can be lost before giving up on guiding max_tries is the number of times you try to look for a guiding star. Even though they may seem redundant they are actually not.

Well, yes... I think all this will be necessary. filterwheel does exists, but usually it is inherited by the camera as part of a major instrument. It is used like that on autofocus tough.

Yes Noffset and Eoffset are the offset of the auto guider with respect to the focal plane. They are supposed to be given in degrees.

— Reply to this email directly or view it on GitHub <https://github.com/astroufsc/chimera/pull/88#issuecomment-124639497 .

— Reply to this email directly or view it on GitHub https://github.com/astroufsc/chimera/pull/88#issuecomment-125192394.

— Reply to this email directly or view it on GitHub https://github.com/astroufsc/chimera/pull/88#issuecomment-125193532.

— Reply to this email directly or view it on GitHub https://github.com/astroufsc/chimera/pull/88#issuecomment-125198792.

[ankanaan]

You are right. BUT...

This depends on how the guiding camera is mounted. Some move around with X,Y screws. Some move around the central CCD in a circular motion around a ring. In the latter case the ccd orientation changes as you move the ccd. I guess I was thinking of this case, where you need some number to tell you what the ccd orientation is. Still, it is possible to compute what the ccd orientation is based only on Eoffset and Noffset.

Maybe what we need is some parameter describing what kind of mechanical mount the guiding camera has?

On Mon, Jul 27, 2015 at 10:10 AM, William Schoenell < notifications@github.com> wrote:

@ankanaan: Maybe I didn't understood about the PA (I just checked the figure), but I think that if you have an offset from east and another one from the north having the science camera as reference, you have the PA. @tribeiro: I was thinking on that when doing the question ;-)

On Mon, Jul 27, 2015 at 3:04 PM Antonio Kanaan notifications@github.com

wrote:

yeap, by position angle, I mean position angle :-)

here is the defintion:

https://en.wikipedia.org/wiki/Position_angle

it is not only used for binary stars as the article seems to imply

On Mon, Jul 27, 2015 at 9:48 AM, William Schoenell < notifications@github.com

wrote:

By position angle you mean a rotational angle wrt the north?

On Mon, Jul 27, 2015 at 2:44 PM Antonio Kanaan < notifications@github.com

wrote:

in addition to Eoffset and Noffset I may see a need for PA, Position Angle of the center of the guiding CCD wrt to the science CCD.

On Fri, Jul 24, 2015 at 4:06 PM, Tiago notifications@github.com wrote:

nlost is the number of times the star can be lost before giving up on guiding max_tries is the number of times you try to look for a guiding star. Even though they may seem redundant they are actually not.

Well, yes... I think all this will be necessary. filterwheel does exists, but usually it is inherited by the camera as part of a major instrument. It is used like that on autofocus tough.

Yes Noffset and Eoffset are the offset of the auto guider with respect to the focal plane. They are supposed to be given in degrees.

— Reply to this email directly or view it on GitHub < https://github.com/astroufsc/chimera/pull/88#issuecomment-124639497 .

— Reply to this email directly or view it on GitHub <https://github.com/astroufsc/chimera/pull/88#issuecomment-125192394 .

— Reply to this email directly or view it on GitHub https://github.com/astroufsc/chimera/pull/88#issuecomment-125193532.

— Reply to this email directly or view it on GitHub https://github.com/astroufsc/chimera/pull/88#issuecomment-125198792.

— Reply to this email directly or view it on GitHub https://github.com/astroufsc/chimera/pull/88#issuecomment-125200160.

[wschoenell]

Yes. And this can be accomplished by Eoffset, Noffset and rotation_angle of the camera. I actually am more for moving N/Eoffset to the camera interface where rotation_angle is actually.

So, for example, the main camera will be say in (x,y,w) = 0,0,15 (for a camera with 15º of rotation wrt the north) and the piggyback camera can be at (x,y,w) = (10,10,0) (for a camera 10 whatever unit North, 10 wu East and aligned with the north). Everything goes to the WCS and the controllers should know how to deal with it.

Doing this, you can even have a telescope with 4 cameras which no one is actually on (0,0) where the telescope mount is pointing to, like the HATS telescopes:

[wschoenell]

So... My suggestion is to move the offsets to the camera interface together with the WCS fixes (on another PR) and set the __config__ to the autoguider as:

__config__ = {"site": '/Site/0',            # Telescope Site.
              "telescope": "/Telescope/0",  # Telescope instrument. This will be guided by the autoguider.
              "camera": "/Camera/0",        # Guider camera instrument.
              "filterwheel": None,          # Filter wheel instrument, if there is one.
              "focuser": None,              # Guider camera focuser, if there is one.
              "autofocus": None,            # Autofocus controller, if there is one.
              "max_acquire_tries": 3,       # Number of tries to find a guiding star.
              "max_fit_tries": 3}           # Number of tries to fit the guide star offset before being lost.

which will have sane defaults for the optional parts of the guider.

[tribeiro]

Thanks for the suggestions @wschoenell. Those where implemented in this later commit. I vouch for this PR to be accepted... :smile:

[wschoenell]

Ok. Don't forget to add on a future PR the two configuration parameters that we removed here.

              "offset_north": 0.,             # Camera offset in N/S axis in respect to telescope pointing. N is +. (in degrees)
              "offset_east": 0.,              # Camera offset in N/S axis in respect to telescope pointing. E is +. (in degrees)

[ankanaan]

I like it William.

On Mon, Jul 27, 2015 at 11:21 AM, William Schoenell < notifications@github.com> wrote:

Yes. And this can be accomplished by Eoffset, Noffset and rotation_angle of the camera. I actually am more for moving N/Eoffset to the camera interface where rotation_angle is actually.

So, for example, the main camera will be say in (x,y,w) = 0,0,15 (for a camera with 15º of rotation wrt the north) and the piggyback camera can be at (x,y,w) = (10,10,0) (for a camera 10 whatever unit North, 10 wu East and aligned with the north). Everything goes to the WCS and the controllers should know how to deal with it.

Doing this, you can even have a telescope with 4 cameras which no one is actually on (0,0) where the telescope mount is pointing to, like the HATS telescopes:

[image: cimg9943] https://cloud.githubusercontent.com/assets/1873811/8908278/746184a0-347b-11e5-853f-8e4ac8728069.jpg

— Reply to this email directly or view it on GitHub https://github.com/astroufsc/chimera/pull/88#issuecomment-125223538.