Which way is up?

[SteveBz]

Hi All,
I just spent a lot of effort trying to get North up for a slightly dodgy picture of M57 (see below, cropped), with the viewfinder aligned with the front of the scope and the short edge aligned with the side.    Like  next picture but with the viewfinder up:

And when I uploaded it to nova.astrometry.net, it said "up is 88 degrees East".  So I understand that in the dark, I might have been 2 degrees out, hence the 88 bit, but really, it wants the viewfinder to the side? That seems to defy logic.  Can someone explain please?
Regards

Steve.

[Ant]

Will rotating the whole scope in the cradles make the difference that you need? Or try rotating the camera in the focuser by 88 degrees clockwise. 

Just out of interest, why are you bothered about which way is up? Especially for something like M57? I understand why orientation is preferable for larger targets that have an unofficial "which way is up", but I never worried about smaller targets.

Ant

[vlaiv]

I think that problem with newtonian in this setup is that motion in DEC axes rotates sensor relative to RA axes.

With setup as in image if you look at relation between RA axes and shorter side of sensor (or whole camera orientation) - it is in same direction, so up on camera is in direction of RA.

Now imagine if you rotate ota in DEC - sensor "up" will now no longer be pointing in RA axes direction but at an angle to it. So changing DEC of where scope is pointing with newtonian changes sensor orientation (or rotates it). Not so with refractor or cas.

Now if you place your focuser on a side instead on top of ota (I know, bad for balance, but just explaining). Then motion in DEC will not induce rotation of sensor relative to RA (at least I think so).

In configuration that you showed on image sensor lies in plane that is perpendicular to DEC axes, so any rotation in DEC axes will rotate sensor. If sensor is on right it is no longer perpendicular to DEC axes but parallel to it.

I hope I'm correct and that I've explained properly, it gets a bit dizzy in my head when trying to rotate imaginary view around to check what is rotating and why

[SteveBz]

35 minutes ago, Ant said:

Will rotating the whole scope in the cradles make the difference that you need? Or try rotating the camera in the focuser by 88 degrees clockwise. 

Just out of interest, why are you bothered about which way is up? Especially for something like M57? I understand why orientation is preferable for larger targets that have an unofficial "which way is up", but I never worried about smaller targets.

Ant

It feels like holding the map upside down.  I just think it's a bit more professional.

[SteveBz]

12 minutes ago, vlaiv said:

I think that problem with newtonian in this setup is that motion in DEC axes rotates sensor relative to RA axes.

With setup as in image if you look at relation between RA axes and shorter side of sensor (or whole camera orientation) - it is in same direction, so up on camera is in direction of RA.

Now imagine if you rotate ota in DEC - sensor "up" will now no longer be pointing in RA axes direction but at an angle to it. So changing DEC of where scope is pointing with newtonian changes sensor orientation (or rotates it). Not so with refractor or cas.

Now if you place your focuser on a side instead on top of ota (I know, bad for balance, but just explaining). Then motion in DEC will not induce rotation of sensor relative to RA (at least I think so).

In configuration that you showed on image sensor lies in plane that is perpendicular to DEC axes, so any rotation in DEC axes will rotate sensor. If sensor is on right it is no longer perpendicular to DEC axes but parallel to it.

I hope I'm correct and that I've explained properly, it gets a bit dizzy in my head when trying to rotate imaginary view around to check what is rotating and why

Hi Vlaiv,

I think you're exactly right.  As you mount the camera, if you rotate the DEC axis, the camera will move UP and DOWN, showing which is the correct camera orientation.

Perfect.

That's why I always come to stagazerslounge.com

Regards

Steve.

[Ant]

Pleased I'm not a professional then lol. I just try and get something pretty

Looks like you might be sorted

Cheers

Ant 

 

[ollypenrice]

To orientate you camera NSEW, 1) 'eyeball' it as you did to be parallel with the dovetail or CW bar or whatever. 2) switch off the tracking. The stars will exit the field as they drift west so the side they leave on is west. 3) shine a flashlight just over the N side of the tube, or put your fingers into the light path on the N side and take a sub. Either the glow of the light or the shadow of your fingers will show the north side. 4) If you want perfect alignment, slew slowly on one axis while taking a four second sub. Fine tune the camera rotation till the trails are vertical or horizontal.

Remember you can rotate the image afterwards in software.

Olly

Edit, Sorry, crossed with the previous two posters.

[SteveBz]

3 minutes ago, Ant said:

Pleased I'm not a professional then lol. I just try and get something pretty

Looks like you might be sorted

Cheers

Ant 

 

Sorry, Ant, I'm just a bit anal about these things.  Plus I'm a rubbish photographer.  You saw the state of my m57, poor coma, light pollution, vignetting, poor collimation and poor focus, but at least I thought it was the right way up!!

[Ant]

Just now, SteveBz said:

Sorry, Ant, I'm just a bit anal about these things.  Plus I'm a rubbish photographer.  You saw the state of my m57, poor coma, light pollution, vignetting, poor collimation and poor focus, but at least I thought it was the right way up!!

We're all in this game for something different. I'd just be happy to get some photon on a chip... Haven't done that for soooooo long. 

[SteveBz]

Just now, Ant said:

We're all in this game for something different. I'd just be happy to get some photon on a chip... Haven't done that for soooooo long. 

I know.  That was my first night out since spring and just everything went wrong.  I'm outside now trying to disentangle the wires and work out what's not working.

But you know, everyone was asleep indoors and I was out until 3am, and I just loved it.

[Ant]

I went out for half an hour Saturday night and just sat back and watched meteors, it was really refreshing. 

Imaging... you got to love it! 

[SteveBz]

21 minutes ago, ollypenrice said:

 

To orientate you camera NSEW, 1) 'eyeball' it as you did to be parallel with the dovetail or CW bar or whatever. 2) switch off the tracking. The stars will exit the field as they drift west so the side they leave on is west. 3) shine a flashlight just over the N side of the tube, or put your fingers into the light path on the N side and take a sub. Either the glow of the light or the shadow of your fingers will show the north side. 4) If you want perfect alignment, slew slowly on one axis while taking a four second sub. Fine tune the camera rotation till the trails are vertical or horizontal.

Remember you can rotate the image afterwards in software.

 

Nice, I like step 2), it's quite intuitive.

I guess that N is going to flip during the meridian flip. And as I probably don't want to move it during a night, I can then flip afterwards with software.  But really it's quite hard to rotate a few degrees with s/w, easier to rotate in right angles.

[Ant]

The image orientation will change by 180 degrees during a flip. But the stacking software will take that into account and auto rotate to match. 

Ant

[SteveBz]

1 minute ago, Ant said:

The image orientation will change by 180 degrees during a flip. But the stacking software will take that into account and auto rotate to match. 

Ant

ALL stacking software?  Wow, actually you may be right.  In spite of the fact that my stacker is hand-rolled, I use Python's OpenCV image editing suite, and it has an align function to account for slight movements between images, so I guess that will do it.  I didn't even think of that.

Very cool.

Thanks for that.

[ollypenrice]

14 minutes ago, SteveBz said:

ALL stacking software?  Wow, actually you may be right.  In spite of the fact that my stacker is hand-rolled, I use Python's OpenCV image editing suite, and it has an align function to account for slight movements between images, so I guess that will do it.  I didn't even think of that.

Very cool.

Thanks for that.

Not necessarily all. You have to ask the one I use, AstroArt, to do rotations up to 180 then it will do them. No big deal though.

Olly