Determining Camera Tilt

[Aidan]

Thought it best to start a clean thread regarding the tilt problem i am trying to diagnose.

Taking things one step at a time to determine exactly where the problem is.     So, the problem i am seeing is elongated/out of focus stars in the bottom left corner of my images as can be seen in the attached image.

If it is camera tilt, if i rotate the camera through 180 degrees, should i then see the elongated/out of focus stars in the top right corner?

Aidan

 

[carastro]

Sounds about right Aidan, though if you think that is bad you should have seen mine before I tried to sort it out.  

I have added a 3rd bolt to the focusser so the nosepiece is being gripped on 3 sides and not just two (one of my scopes won't take a click lock as there is no thread) and ignored all the misleading info on spacer distance as none of them seemed to work and took pot luck on a spacer I had in the box which together seems to have done the trick.  (watch this space Lol).  (Lol pun not intended)

Carole 

[cjdawson]

Rotating the camera will only move the position of the tilt, if indeed something in the camera is causing the tilt.    If it's something else that causing the tilt, the problem will move relative to the image.   i.e. stay in the same place optically, but will land on a different part of the sensor.

[Aidan]

I want to test one thing at a time..

1st rotation of the camera and determine if the camera is or isn’t the problem.

So, as per my original post,  if distortion is bottom left and camera is rotated 180 degrees, if the camera is at fault the distortion should then be top right??

 

[michael8554]

Do you mean the sensor is tilted in the camera body ?

Then it won't matter where you rotate to, the same corner of the image will still be tilted, unless it's loose in the body.

But if you rotate and the tilt moves, then the tilt is forward of the camera, like lack of third screw as Carole suggests.

Michael 

[Aidan]

I am just not getting my head around this...  You know those times when someone tells you something and you dont doubt them, but just cant compute what is being said... Thats where i am with this!!

What i dont seem to be able to fathom is:    If the image is showing distortion in the bottom left and its due to the sensor being physically tilited, if i rotate the camera, the tilt will follow the orientation of the camera.  So, if i move the camera 90 degrees clockwise, the distortion would end up top left of the image as thats where the affected cornor of the camera is now located?

Am i just being totally dumb in my way of thinking?

[geordie85]

12 minutes ago, Aidan said:

I am just not getting my head around this...  You know those times when someone tells you something and you dont doubt them, but just cant compute what is being said... Thats where i am with this!!

What i dont seem to be able to fathom is:    If the image is showing distortion in the bottom left and its due to the sensor being physically tilited, if i rotate the camera, the tilt will follow the orientation of the camera.  So, if i move the camera 90 degrees clockwise, the distortion would end up top left of the image as thats where the affected cornor of the camera is now located?

Am i just being totally dumb in my way of thinking?

I believe you've got this the wrong way round. What works in my mind is if the tilt is in your sensor, then the affected stars will always be in the same location in your images as that's the area that the fault lies. If the affected area moves when rotating the camera, then your tilt lies elsewhere. 

[Aidan]

Sorry, still dont get it i'm afraid.  But, please bear with me as it will click into place soon enough ? (i hop)

I'm running out of ways to explain how i am thinking this through.   Ok, how about this.   If i have a camera and put a mark on the lens in the bottom left corner - a physical mark which is equivilant to a physical defect (tilt). then if i move the camera through 90 degrees, the mark will move 90 degrees too would it not?   So in my mind the physical defect (tilt) on the sensor moves through 90 degrees too... So, when an image is taken, the physical defect of the sensor records that information and the problem is then in the new position.

I really am getting a bit embarrassed about this, as everyone else seems to get it, but i dont.... Laughable isnt it!!!!!!

[Marci]

The camera rotates 90 degrees in physical space. The lens rotates 90 degrees in physical space. The mark therefore moves 90 degrees in physical space, BUT...

The bottom of the camera is the bottom of the camera. When you take a picture from that camera, the bottom edge of the picture will ALWAYS be the bottom edge of the camera, regardless of whether the camera is rotated 90 degrees, pointing straight up, pointing straight down... the defect (if it’s sensor tilt) will always be in the same place on the picture. 

If the tilt is NOT caused by the camera, and instead is caused by (for instance) focuser slop or something else external to the camera, then the issue WILL move around the picture as you rotate the camera.

 

 

 

[geordie85]

1 hour ago, Aidan said:

Sorry, still dont get it i'm afraid.  But, please bear with me as it will click into place soon enough ? (i hop)

I'm running out of ways to explain how i am thinking this through.   Ok, how about this.   If i have a camera and put a mark on the lens in the bottom left corner - a physical mark which is equivilant to a physical defect (tilt). then if i move the camera through 90 degrees, the mark will move 90 degrees too would it not?   So in my mind the physical defect (tilt) on the sensor moves through 90 degrees too... So, when an image is taken, the physical defect of the sensor records that information and the problem is then in the new position.

I really am getting a bit embarrassed about this, as everyone else seems to get it, but i dont.... Laughable isnt it!!!!!!

No, it will be in a different place in relation to the stars as the stars will rotate, but the defect is still in the same place in relation to the sensor. 

Imagine you place a black spot on the sensor with a marker pen. No matter how you rotate the camera, the mark will still show up in the same place in your images. 

[Aidan]

44 minutes ago, geordie85 said:

No, it will be in a different place in relation to the stars as the stars will rotate, but the defect is still in the same place in relation to the sensor. 

Imagine you place a black spot on the sensor with a marker pen. No matter how you rotate the camera, the mark will still show up in the same place in your images. 

I think we’re getting somewhere... i think  it’s my description that’s hindering my understanding.    So, going by what you are saying regardless of the orientation of the camera, the image taken will always remain the same. 

BUT, the defect caused by the sensor would move around on the IMAGE and appear in relation to the sensors position.

Is that making sense, or am I still dumb!

[geordie85]

3 minutes ago, Aidan said:

I think we’re getting somewhere... i think  it’s my description that’s hindering my understanding.    So, going by what you are saying regardless of the orientation of the camera, the image taken will always remain the same. 

BUT, the defect caused by the sensor would move around on the IMAGE and appear in relation to the sensors position.

Is that making sense, or am I still dumb!

If the issue is with the sensor, then the distortion will always be in the same corner (bottom left for example) but if the issue is with the scope/ focuser/ other, then the distortion will change from bottom left to top left after rotating the camera 90 degrees. 

[Aidan]

Aaahhhhhh.....  I honestly hear what you are saying, and I can not in a million years doubt what your saying... but I STILL can’t get my head around the explanation!

I don’t hink I’m ever going to fully understand this unless someone actually shows me face to face and draws pictures for me!!  

Have you ever met anyone who can’t grasp something which in all probability is actually very simple!!

I don’t live too far from you going by your signature...  I’ll buy you a couple of pints!

[michael8554]

Here's my explanation.

Imagine a daytime shot of a card marked TL, TR, BL, BR in the corners.

You take an image Landscape orientation.

The letters BL appear to be tilted.

Rotate camera 90 degrees clockwise and take image .

PC displays image Landscape, now the letters TL are in the bottom left corner of the image.

If only the sensor is tilted at bottom left, the bottom left corner of the sensor has now imaged the letters TL, and now the letters TL will appear tilted .

However, if the sensor is square-on and the Reducer or focuser is tilted, the letters BL will still appear tilted - the Reducer or focuser haven't moved in relation to the card.

Hope this helps.........

Michael

[whipdry]

Just remember it's the camera that's dumb it doesn't know you've rotated it 180 deg....  in reality the star field the camera sees has rotated, but the tilt problem stays in the same place on the sensor and therefore also on the image.

Peter

[sharkmelley]

Here's a different way to explain it, using an (improbable) thought experiment.  Suppose someone had somehow bent the one of the corners of your sensor - impossible, but bear with me!  This causes the stars appearing in the bottom left of the image to appear distorted and out of focus.  Now rotate the camera.  Different stars will now occupy the bottom left corner of the image and they will be the ones that are distorted and out of focus because it is the bent corner of the sensor that is causing the problem.

Sensor tilt is simply a less severe version of a bent sensor!

Mark

[Aidan]

I’m going to take these suggestions and put pen to paper and see if that works for... In the meantime, I went out early this morning and took some images with just the ca,era attached.  I’ll post them up shortly and we’ll see what the opinion is on wha the problem is...  Going by what’s been said here, it looks like the scope could be the problem 

[Aidan]

Here are the images i took earlier.   4 images - 0, 90, 180 and 270 Degrees rotation of the camera.   

 

[sharkmelley]

As you rotate the camera, the sharpest stars are moving from corner to corner.  I agree with your diagnosis: the problem is not camera sensor tilt.

Mark

[Aidan]

Thanks Mark,

Although i am struggling with the mechanics of the understanding...   The pictures do not lie.   So, at least i know that the main area of concentration should be the image train.  I'm hoping that its either the filter wheel or OAG and not the optics.

[carastro]

It's more likely to be the focusser than the FW and I can;t see how an off axis guider would cause this, unless it was causing uneven weight.  

How do you fix your camera to the focusser? 

If you are clamping it with two fixing bolts then that's probably the culprit.  If you are using some sort of centring clamp less likely unless there is slop in the focusser.

Carole 

 

[Aidan]

Hi Carole,

The focuser has a male M48 thread on the end of it.  So, the whole of my image train is screw connections.   ZWO 1600 --> ZWO FW --> ZWO OAG --> 10mm M48 Spacer --> OTA.   I checked the focuser earlier, all the screws are tight and i cant feel any movement when i gently move the camera etc. 

The focuser has a rotator built in where the image train screws to it.  There is a nylon tipped locking knob to stop it moving.  I'm wondering if i am tighting it too much and this is causing the problem?    I have a few things to try now, so just a case of wating for a clear night!