Perfect alignment, how?

[lonepal]

How to be sure that our camera, tube system, lens and carrier rail is absolutely straight and moving absolutely straight?
Bubble level? Laser? or just make a stack and see?

How to detect it without taking many photos?

I need a simple but effective advice.



( Bytheway what do you think about my paint skills )

[Ichthyophthirius]

Hi Omer,

For microscope alignment, pinholes are used that screw into the various mount threads.

So pinholes and lasers, maybe?

[rjlittlefield]

How to detect it without taking many photos?

This is a common question.

As far as I know, nobody has ever come up with a good answer.

In the end, there is always some "gotcha", like "And how do you know that your laser is exactly lined up along the optical axis?"

The best method I know relies on taking pictures and seeing if the center stays centered.

--Rik

[Smokedaddy]

Not that it matters, I know how to laser collimate a telescope but haven't a clue what the process would be elsewhere. Couldn't you draw a couple of diagonals on the output test image to find the sensor intersection then see if the laser is at that intersection.

On a similar note, I would like to know how to verify that your camera sensor is parallel with a flat specimen, such as a wafer, or semiconductor die.

-JW:

[Saul]

...On a similar note, I would like to know how to verify that your camera sensor is parallel with a flat specimen, such as a wafer, or semiconductor die.

I'm using Plumb Bob 5 or SurfaceLevel 4 apps, works great for me.
Phone on the microscope table (calibrate to 0) & camera LCD screen aftewards - adjustment.

[Smokedaddy]

Thanks Saul, I should of been clearer. I was talking about a horizontal camera and specimen setup (wafer, semiconductor die etc.).

-JW:

[Saul]

Thanks Saul, I should of been clearer. I was talking about a horizontal camera and specimen setup (wafer, semiconductor die etc.).

-JW:

For adjustment hang it on the wall
If seriously - it is better to see your horizontal setup. App allows calibration at the any position - so use rubber bands to attach to the camera LCD screen and something similar to your wafer holder

[Smokedaddy]

No offence, you don't understand the question. <g> A level would do no good. Getting the specimen parallel (or very close) could save taking several dozen (depending on the steps) of unnecessary images. I know HOW to make the necessary adjustments via a specimen rotating stage but not how to accurately achieve parallelism.





-JW:

[Saul]

You are right, I didn't understand the question . Can you keep your horizontal setup temporarily vertically to accurately achieve parallelism with the app ?

[Smokedaddy]

That wouldn't make a difference (if I understand what you're saying). It's a big deal 'for me'. If I'm trying to make a panorama of a semiconductor die and divide the die it up into 12 sections, parallelism becomes a big deal in many ways (horizontal and vertical parallelism too). Not only does it become a PITA focus wise but all the additional images that need to be stacked.



-JW:

[rjlittlefield]

use rubber bands to attach to the camera LCD screen

I hope I'm missing something, but why do you believe your LCD screen is parallel to your sensor?

No, I'm not deliberately trying to be cheeky, although sometimes I achieve it by accident. But if the screen tilts, then I don't know any reason to think that the rest position will be parallel to the sensor. If the screen is fixed, I still don't know any reason to think that it's parallel to the sensor to the same level necessary to achieve 4-corner focus. Likewise for the frame of the eyepiece.

For centering, at least, it might be very handy if somebody sold a laser in a camera lens mount, precisely manufactured so that the laser beam lies along the optical axis. But I have not seen such a beast.

how to accurately achieve parallelism

I use Live View and adjust alignment until all sides of the image are in focus at the same time. I assume that this fails somehow with your panorama setup, but I would like to hear the details of how that works.

--Rik

[Saul]

...
I hope I'm missing something, but why do you believe your LCD screen is parallel to your sensor?...--Rik

I checked parallelism between camera mount and LCD. How it is accurate - this is another question (app, Iphone's geometry, how mount is parallel to the sensor etc ). For my purpose it worked OK (original idea to use this type of app came from Morfa).

Original discussion:
http://www.photomacrography.net/forum/v ... sc&start=0

[Smokedaddy]

It might be very handy if somebody sold a laser in a camera lens mount, precisely manufactured so that the laser beam lies along the optical axis. But I have not seen such a beast.

Telescope wise they make very precise (and calibrated) laser collimators, some expensive and some not.

https://www.collimator.com/category/laser-collimators

http://syoshi-i.s287.xrea.com/kizai_tyo ... ujiku.html

and some use a star (and even evaluate the optics).

http://garyseronik.com/no-tools-telescope-collimation/

http://www.astrophoto.fr/collim.html

how to accurately achieve parallelism

I use Live View and adjust alignment until all sides of the image are in focus at the same time. I assume that this fails somehow with your panorama setup, but I would like to hear the details of how that works.
--Rik

Live View doesn't work for me at all. Maybe if I had some sort of test test target with vertical and horizontal divisions on it and if I was able to visually distinguish when the left and right most, upper most and lower most, lines were in focus that would help ... dunno. Now that I have the EOS Utility setup on my computer, that might be helpful but I haven't tried it with a panorama like I described. I suppose the most logical setup would be to use my MM-11 and epi illumination. I just thought someone here might have already solved that problem.

-JW:

[elf]

How are you stitching the panorama: orthogonal, spherical, or cylindrical?

[Smokedaddy]

I'm not sure how to answer that sorry, in PTGui, I've used many different projections (rectilinear, cylindrical, transverse equirectangular, etc.), or ICE or Photoshop photomerge similar.