Parallel sensor alignment / digital spirit levels

Have questions about the equipment used for macro- or micro- photography? Post those questions in this forum.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

chris_ma
Posts: 570
Joined: Fri Mar 22, 2019 2:23 pm
Location: Germany

Parallel sensor alignment / digital spirit levels

Post by chris_ma »

Hi everybody,

I'm thinking about easier alignments of the sensor parallel to flat objects.
at lower magnifications I can use a mirror, but at higher mag that doesn't prove to be very helpful.

so I was wondering if any of those smaller digital spirit levels are any good?
there seem to be no-name ones for around 30EUR and brand companies selling them for 60EUR, and if they are as accurate as claimed, I figured I could measure if the lens tube is pendicular to the object surface, or the camera screen is parallel to the object surface.

or maybe there are better tricks some of you discovered?
Robert, how do you align your wafers if I might ask?

all the best
chris
chris

RobertOToole
Posts: 2627
Joined: Thu Jan 17, 2013 9:34 pm
Location: United States
Contact:

Re: Parallel sensor alignment / digital spirit levels

Post by RobertOToole »

Hi Chris,

For wafer photography I place modeling clay between two aluminum plates and use the Z movement to compress the clay and align the lens and the top plate. This works up to about 10x, I've gotten things lined up to 20x.

Update: I should add that this will never work with anything other than a rigid setup. Any flex at all in the lens/camera/mount and won't align. The SM2 system I use has zero flex so once its aligned it stays aligned for days and I can change lenses freely without changing the alignment. This was the main reason I went to SM2. Using a rig with a single mount, camera or lens, the alignment would constantly change if I with the slightest nudge or touch.

Credit for this trick goes to Chris R!

It's interesting to note that without getting sensor and subject aligned I would be spending 5x the amount of time I do now for a test. With the wafer - sensor aligned I still run a focus stack and check for best corners since the lens field can curve. But the stack can be only 5-10 images instead of something like 25-50 without alignment.

Another interesting thing is that in the past few months I've been using movements, tilt, swing, with all my general photography and its been so much fun and productive to the point that I don't see myself going back to "parallel" unless its for a sharpness test.


Best,

Robert

Scarodactyl
Posts: 1617
Joined: Sat Apr 14, 2018 10:26 am

Re: Parallel sensor alignment / digital spirit levels

Post by Scarodactyl »

A coaxial illuminator helps for visual alignment using my magnetic positioning stand, DIC helps a bit more. I also have a couple interferometry objectives if I'm feeling particularly special, though they're really too sensitive so they've mostly just been showing me I need to get some sorbothane pads for my illuminator.

elf
Posts: 1416
Joined: Sun Nov 18, 2007 12:10 pm

Re: Parallel sensor alignment / digital spirit levels

Post by elf »

I have one I use in woodworking, but I only need to get things aligned to about a degree :D It works fine for that. The display resolution is .01 degree but the accuracy is definitely .1 degree or worse. What are your required tolerances?

seta666
Posts: 1071
Joined: Fri Mar 19, 2010 8:50 am
Location: Castellon, Spain

Re: Parallel sensor alignment / digital spirit levels

Post by seta666 »

I just use a 5 axis positioning stand ( X, Y, 360° and 2x goniometers), for something like wafers I think is more than enough ; Z axis is the motorized rail for stacking.

For other subjects ( even for butterfly wings) the flatter angle is not always the best one.

Taking the pictures is just a fraction of the time I spend with images; retouching and editing acounts for 90% of the time so it does not really bother me to take 50-100% more images if needed.

RobertOToole
Posts: 2627
Joined: Thu Jan 17, 2013 9:34 pm
Location: United States
Contact:

Re: Parallel sensor alignment / digital spirit levels

Post by RobertOToole »

seta666 wrote:
Mon May 03, 2021 12:45 am
I just use a 5 axis positioning stand ( X, Y, 360° and 2x goniometers), for something like wafers I think is more than enough ; Z axis is the motorized rail for stacking.
Sounds good.

Having the capability to align something and actually being able to get it aligned is a whole different story in my experience. :-k

Goniometers can take tons of time but at least once the stage and camera are aligned it should stay aligned so you can swap wafers.
For other subjects ( even for butterfly wings) the flatter angle is not always the best one.
There is no flat with a butterfly wing in my experience :D
Taking the pictures is just a fraction of the time I spend with images; retouching and editing acounts for 90% of the time so it does not really bother me to take 50-100% more images if needed.
Yes. True. Casually shooting stacks with one lens is easy for sure.

With testing there is almost no retouching, just processing which can be done in an action so its just killing time.

The biggest challenge? Keeping the stacks labeled and organized properly! I've messed up and had to start over more than once over the years. My tube lens test in 2018, 20 tube lenses, multiple copies of a few, so 25 lenses, 3 f-stops each, 3 distances each and 2 lens to objective distances, what is that 25x3x3x2=225 stacks. Took all day and a half. Lost my note pad with all the test notes. Had to start over. That one almost killed me. #-o Never did find the legal notepad, I think my wife tossed it in the trash bin. :-k

{-( Will never running a test like that big again.

Best,

Robert

chris_ma
Posts: 570
Joined: Fri Mar 22, 2019 2:23 pm
Location: Germany

Re: Parallel sensor alignment / digital spirit levels

Post by chris_ma »

interesting thoughts, thanks for all the replies!

stacking would be easiest of course (and maybe I'm going to resort to that in the end anyway), but it would be nice to get things in one shot.
RobertOToole wrote:
Sun May 02, 2021 10:23 am
For wafer photography I place modeling clay between two aluminum plates and use the Z movement to compress the clay and align the lens and the top plate. [...]
Credit for this trick goes to Chris R!
very clever trick indeed!
probably not going to work for my setup, but I could use a tube extension (I also use SM2 parts, mainly thanks to your website) and use the front of the last tube to align the mounting plate.

I'm thinking about building a custom bellow system using two THK20 rails (one for the front and one for the back ring) for quicker adjustment of tube length, but I first need to figure out how to design that so the precision parts don't cost a fortune.
Last edited by chris_ma on Mon May 03, 2021 11:12 am, edited 1 time in total.
chris

chris_ma
Posts: 570
Joined: Fri Mar 22, 2019 2:23 pm
Location: Germany

Re: Parallel sensor alignment / digital spirit levels

Post by chris_ma »

elf wrote:
Mon May 03, 2021 12:45 am
I have one I use in woodworking, but I only need to get things aligned to about a degree :D It works fine for that. The display resolution is .01 degree but the accuracy is definitely .1 degree or worse. What are your required tolerances?
yeah, from the data sheets of the brand models, the accuracy of the affordable pocket ones seem to be 0.05deg at 0 and 90deg position, and 0.1deg in between.
here one I've been looking at (no idea about quality but looks like a reliable company to me)
https://www.sola.at/en/products/go-smart-p268240

on my setup I have a DoF of about 15um at 1:1 on a 36x24 sensor before I start to see a slight softening.
so did some calculations, if my math is right then at the far corner, an error of 0.05deg would result in an offset of roughly 18um (sind(0,05) * 21mm = 0.018mm)

so that might just about work out!
chris

seta666
Posts: 1071
Joined: Fri Mar 19, 2010 8:50 am
Location: Castellon, Spain

Re: Parallel sensor alignment / digital spirit levels

Post by seta666 »

RobertOToole wrote:
Mon May 03, 2021 8:29 am
..........what is that 25x3x3x2=225 stacks. Took all day and a half. Lost my note pad with all the test notes. Had to start over. That one almost killed me. #-o
............
{-( Will never running a test like that big again.
............
I used to run lots of tests too; I lost the piece of paper were I had the notes for one I did with 9 or 10 different lenses at 20X .
But in your case loosing the notes was much worse.. :D

chris_ma
Posts: 570
Joined: Fri Mar 22, 2019 2:23 pm
Location: Germany

Re: Parallel sensor alignment / digital spirit levels

Post by chris_ma »

I realised that Mitutoyo makes some spirit levels...

their digital version states 0.05deg accuracy on horizontal measurements and costs 660EUR:
https://shop.mitutoyo.eu/web/mitutoyo/e ... ndex.xhtml

so I think it's safe to say that those 30EUR levels on amazon will not quite reach 0.05deg accuracy.

they also have a analog version, which is a bit cheaper (460EUR) and states 0.02mm/m accuracy on horizontal measurements, which if I'm calculating this right would be around 0.001degree.

now that certainly would be enough :)
next thing to do is to find a good CNC service.
chris

elf
Posts: 1416
Joined: Sun Nov 18, 2007 12:10 pm

Re: Parallel sensor alignment / digital spirit levels

Post by elf »

You could easily build your own super precise level. :D

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: Parallel sensor alignment / digital spirit levels

Post by rjlittlefield »

elf wrote:
Thu May 06, 2021 12:37 am
You could easily build your own super precise level. :D
That's a very clever device: butane with bubble in a glass tube that is bent slightly by pressure from a couple of adjusting screws.

As shown in the video, the builder gets bubble movement of about 2.5 cm for an inclination of about 1 part in 1000 (0.006" shim in roughly 6" length of level). I doubt it would reach the 1 part in 50,000 sensitivity of the best Mitutoyo unit, but it's not hugely off and it's much cheaper (ignoring the time to build...).

--Rik

chris_ma
Posts: 570
Joined: Fri Mar 22, 2019 2:23 pm
Location: Germany

Re: Parallel sensor alignment / digital spirit levels

Post by chris_ma »

or, one could get a swiss made one with micrometer on a discount for a mere 3600USD :)
https://store.gaging.com/wyler-spirit-l ... ensitivity

But it definitely seems that as long I only need horizontal level, a level with bubble is much more accurate then an electric one. It looks like one can get 0.5mm per meter accuracy for about 25bucks:
https://www.rabo-bormann.com/wasserwaag ... 879_125599

So that would be about 10um on the corner of a full frame sensor, or just about enough for my target. I guess the tricky bit is to brush away all the dust first :)
chris

mightimatti
Posts: 28
Joined: Tue Feb 08, 2022 9:03 am
Location: Rome, Italy

Re: Parallel sensor alignment / digital spirit levels

Post by mightimatti »

Hi everyone,
this is my first post outside of the equipment exchange, but I came across this thread and wanted to add an idea:

I am building a automated rig to make reproductions of film, using a camera in a M42 bellows unit and to align my camera and surface I ended up buying these items:

Baader Planetarium Laser collimator A device for aligning the mirrors in telescopes
M42->1.25" telescope ocular adapter This adapter usually allows you to insert a telescope ocular instead of an m42 camera body,but I am using it because the interface is the same on the collimator. I actually have a c-mount -> 1.25" adapter so I can quickly slip in my camera or the collimator and the bore adapter can stay on the bellows permanently.
thorlabs 1"x1" first surface mirror.

The idea is that you place the mirror on whatever you intend to image, replace the camera with the laser collimator and manipulate alignment until the laser is reflected back "into itself" at this point(assuming the collimator itself is collimated which is easy to check by rotating it in the mount while shining at a distant wall) you have aligned the optical axis at a right angle to the surface you are imaging with accuracy proportional to the distance to the mirror.

This isn't the cheapest solution, but especially if you have a long working distance or for some reason Robert's trick isn't viable this should work pretty well. I have yet to test it, as I am still waiting for the m42-> ocular adapter, but I have high expectations.

mightimatti
Posts: 28
Joined: Tue Feb 08, 2022 9:03 am
Location: Rome, Italy

Re: Parallel sensor alignment / digital spirit levels

Post by mightimatti »

Hi,
I thought I'd post a couple of pictures, as I have been using the method described in my previous post for a couple of weeks and can confirm that it works very well and with great accuracy and repeatability.

I use a M42->1.25"(standard attachment for telescope oculars) adapter on my bellows unit and slide the laser collimator in when required. I use this setup for making reproductions of photographic film, so I have the option to just reflect the laser dot off the film(though the dot wanders significantly when you transport the film) or use a little 1" thorlabs first surface mirror(could be any other piece of mirror or even glass really). The setup can also be used with the optics in the light path to confirm their alignment, as any misalignment on a the eliptical lenses immediately becomes apparent.

I usually move the rear board of the bellows as high up as I can, as distance increases the accuracy of the laser. The reticle in the laser collimator visualizes to see the misalignent.
Attachments
5s.jpeg
3s.jpeg
2s.jpeg
1s.jpeg

Post Reply Previous topicNext topic