Help me spend money on a 20X objective and camera mount!

Images taken in a controlled environment or with a posed subject. All subject types.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

donkom
Posts: 8
Joined: Tue Nov 15, 2016 6:09 pm
Location: Ontario, Canada
Contact:

Help me spend money on a 20X objective and camera mount!

Post by donkom »

Hello Everyone,

I'm new to this forum, but my work in macro and microscopy has received a lot of attention for two subjects in particular: water droplets and snowflakes. My name is Don Komarechka, and with winter just about here, I'd like to take my game up a few notches with the smallest snowflakes this winter.

My current gear is a Canon MP-E 65mm, and with extension tubes + the Canon Lifesize Converter EF I can get to about 12:1 magnification. I believe I'm the only person out there that shoots focus stacks of snowflakes at this magnification - entirely handheld.

I'm looking to jump into a 20X setup, and I've got a very decent budget for it ($2-3K) for a microscope objective and the necessary hardware to properly mount it on my camera.

I was looking at a 20X Mitutoyo telecentric objective, as I know these sorts of objectives are favoured in focus stacking scenarios:
http://www.edmundoptics.com/microscopy/ ... ves/87228/

The Nikon TM series of lenses has been discontinued according to my Nikon rep, and they recommended this as a similar product.

A few questions:
- Would any other optic be better suited?
- How well does a telecentric objective handle vertical and horizontal alignment that is required due to handholding the camera?
- What hardware would I need to mount this lens (or any other recommended ones) to a Canon full-frame DSLR.
- Will such a lens fill the 36x24mm image frame, or is there a better option at this scale?

Diving into the deep end here I know, but any advice is appreciated!

(below is one of my snowflakes, a very rare variety of capped column crystal that measures roughly 1.5mm tall)


Image

Pau
Site Admin
Posts: 6051
Joined: Wed Jan 20, 2010 8:57 am
Location: Valencia, Spain

Post by Pau »

Hi, donkom, welcome aboard!

Your snowflake is extraordinary.

20X has been discussed several times at the forum and because you can afford it the Mitutoyo M Plan Apo 20 0.40 (avoid the SL) is likely the best option, no info yet about the telecentric lens linked.

Just a caveat: stacking handheld at 20X would be heroic if not just impossible, DOF is so small...

Other members will give you for sure more detailed advice, I don't own this kind of optics
Pau

soldevilla
Posts: 684
Joined: Thu Dec 16, 2010 2:49 pm
Location: Barcelona, more or less

Post by soldevilla »

Wellcome on board! I know your work, I seen it searching internet some years ago. Your Snowflake presentation is incredible!

donkom
Posts: 8
Joined: Tue Nov 15, 2016 6:09 pm
Location: Ontario, Canada
Contact:

Post by donkom »

Thanks Pau!

I appreciate your compliments on my snowflake work, and if I can handle 12:1 handheld, I think there's a good chance I can handle 20:1. I think I'll be the first to accomplish that goal, though!

With the objective you noted, it doesn't look like it's telecentric but without experiencing the difference myself, I am leaning to the advice of others such as yourself. Because the lens you suggest is an infinite objective, what would be the best way to mount it to my camera? A 100mm macro lens as the "tube"?

For clarity, there would be no better objective for my purposes than the one you mentioned: http://www.edmundoptics.com/microscopy/ ... ves/46145/

or would there be another option? :)

Thanks too soldevilla! Glad you're familiar with my work - let's see how far I can push it!

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

Re: Help me spend money on a 20X objective and camera mount!

Post by rjlittlefield »

Don, welcome aboard -- that's a great image of a spectacular specimen!

I know you've said that you're diving into the deep end, but I'm not sure you appreciate just how much deeper it is, compared to what you're used to.

I see from the EXIF data in the image you've posted that it was shot with the MP-E 65 set on f/3.2. At 12X through the Canon Lifesize Converter EF, the DOF using standard formulas is about 0.026 mm.

When you switch to a 20X NA 0.42 objective, DOF using the same formulas drops to about 0.0031 -- over 8 times more shallow. This is a direct result of the higher resolution offered by the 20X objective.

Now, it may be that the 20X objective offers more resolution than you really need. Compared to biological anatomy, freezing water is not so prone to producing small scale details.

If you don't need the full resolution of a 20X NA 0.42 objective, then you can trade reduced resolution for more DOF by mounting an adjustable iris just behind the objective so as to stop it down.

However, when you do this, you will also destroy whatever telecentricity the objective originally had. This is because telecentricity is determined by the position of the limiting aperture, and it's virtually guaranteed that the added iris will not be in the proper position to make the combination be telecentric.

In addition, the DOF offered by any 20X objective, operated wide open, is so shallow that you can just pretend it is telecentric for purposes of stacking, even if the designer or very careful measurements would say that it is not.

Taking all this into account, I recommend not spending the additional money to get an objective that is sold as telecentric.

Instead, get the nominally non-telecentric "20X Mitutoyo Plan Apo Infinity Corrected Long WD Objective" that you pointed to at http://www.edmundoptics.com/microscopy/ ... ves/46145/ .

Then, combine that objective with an iris, a Raynox DCR-150 used as "tube lens", and a set of empty tubes, roughly as shown at http://www.photomacrography.net/forum/v ... 555#189555 plus the iris. (I'll provide more details in a later post to our FAQ: How can I hook a microscope objective to my camera? It will be a long overdue addition to that topic.)

To address your specific questions:

- Would any other optic be better suited?
No.

- How well does a telecentric objective handle vertical and horizontal alignment that is required due to handholding the camera?
Telecentricity does not affect this.

- Will such a lens fill the 36x24mm image frame, or is there a better option at this scale?
Yes, the 20X Mitutoyo objective with Raynox DCR-150 tube lens will fill a 36x24 frame with high quality image, when used at rated magnification with a ~200 mm tube lens.

A 100mm macro lens as the "tube"?
No, this won't work. With a 100 mm tube lens, the objective will only be delivering 10X. The field at sensor shrinks accordingly, and the objective will not be able to deliver high quality image clear to the edge of the sensor on full-frame.

I hope this helps!

--Rik

ChrisR
Site Admin
Posts: 8668
Joined: Sat Mar 14, 2009 3:58 am
Location: Near London, UK

Post by ChrisR »

I never would have believed a crystal would form like that!
(Are you sure you aren't related to the Victorian microscopists who "made" pictures of radiolarians according to how they thought they might exist?:lol: )

I don't think anyone in the forum has tried one of that family of objectives.
This
http://www.edmundoptics.com/microscopy/ ... ves/46145/
is the well known, good performer.

They aren't far (at all) off telecentric. They aren't much less expensive new, but are much more common used than the one you linked to, meaning that you could find the above for a fraction of the price. Unless you want to stitch images together, there's much less benefit in using a telecentric objective - if any at all.

Obviously, moving the camera a couple of microns at a time (at 20x) is what you're faced with. I daresay you've refined "hand held", but I suspect you'd do better with a rapidly movable rig of some sort.

I doubt the objective you refer to is designed to cover anything near 24 x 36, though you may "get away with it" - Pass. With the non telecentric 20x, the central definition is definitely better than the edges, for 24 x 36.

You'd need a 200mm tube lens to get 20x.
At lower magnifications with it, the corners would be unsharp or dark.

If you're using an MP-E with tubes and extenders, you're already into diffraction limited resolution. That means you'd be as well off using less magnification and "blowing up the middle". You'd be using fewer pixels, but they'd be used, instead of redundant (oversampled) as now.
If you're at 12x with an MPE, wide open that's
(12 + 1) x f/2.8 = eff 36.4. You're "diffraction limited" for more than about 8MP on full frame.

I've just run some numbers. Using the same criteria for sharpness, I got:
with what you're using, 2200 pixels wide is sharp.
With a 10x NA 0.28 objective it would be 4500
With a 20x NA 0.42 objective it would be 3500

That may make you scratch your nut, ...?
My point is that you may be better off with a 10x objective than pushing all the way to 20x. Wider field of view than a 20x, and a lot sharper than you have now.

If you really can control things, then for the tiny crystals, yes a 20x would be better for detail on subject.

The sharpness figure for the 20x drops on sensor because nobody makes a good enough lens with usable working distance.
Chris R

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

Post by rjlittlefield »

Don, please note that ChrisR and I were apparently typing at the same time.

Our posts take different directions on the same underlying issue: that an unadorned 20X objective will give more resolution and less DOF than appears to be suited to your application.

His suggestion of getting a 10X objective and then cropping to the especially good center portion makes sense. Alternatively, you could get the 10X objective and use it with a tube lens longer than 200 mm, and/or a teleconverter, so as to optically expand the center portion to cover your sensor. Either way, you'd be getting closer to 0.008 mm DOF, versus 0.003 mm for the 20X with no iris, and with about twice the resolution on subject that your MP-E 65 is giving.

--Rik

donkom
Posts: 8
Joined: Tue Nov 15, 2016 6:09 pm
Location: Ontario, Canada
Contact:

Post by donkom »

Chris and Rik, thank you SO MUCH for the extra information.

From what I gather:

- the Mitutoyo Plan APO 20X will work, but will potentially need an iris/aperture to compromise resolution for DOF for my use (by the way, there's one on eBay - good buy? http://www.ebay.ca/itm/MITUTOYO-PLAN-AP ... SwmfhX5wsE )

- I'll need to build out 200mm of extension tubes for which these should serve a good purpose: http://www.ebay.ca/itm/Macro-Extension- ... xyu1FRk2i5


Questions still outstanding:

- Where does a Raynox DCR-150 fit into the equation?

- What would a recommended iris / aperture be constructed out of, and where in the tube would it be placed?


I think I'd like to use the 20X as a solution for this purpose, as it would also allow me to use the same magnification in studio scenarios and I've got some ideas for that as well.

I have no ability to take these images on any sort of rig, for a number of reasons. I need to rotate the camera and ring flash around the subject being the center of rotation to get surface reflection to play in my favour. Some snowflakes, due to layers of ice and air, produce vibrant colours via thin film interference. Often times these crystals are sub-millimeter in diameter, like this one (0.6mm across):

Image

Timing is also an issue, as these snowflakes are active sublimating in front of me - wait five minutes and the snowflake will be a ghost of it's former self. You need to work as fast as possible and deal with alignment issues in post, but I've gotten pretty good at that. :)

I understand now (thanks to the fantastic comments) that the DOF would be 8 times more narrow with the objective on it's own, and I'd like some ability to compromise resolution for practical usability. Even if the DOF is 2-4 times narrower I think I could make a decent go of it.

Thanks again guys, I look forward to more on this topic!

Lou Jost
Posts: 5942
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

Whatever you end up doing, I hope you post the results on this forum!!! Fascinating challenges.

ChrisR
Site Admin
Posts: 8668
Joined: Sat Mar 14, 2009 3:58 am
Location: Near London, UK

Post by ChrisR »

See the FAQ "How do I hook a microscope objective up my to camera"

There's a lot to take in.
You put an objective on the front of a 200(ish) mm lens. It could be a camera lens.

You can put an iris diaphragm in the adapter between the two, to increase DOF, at the cost of resolution.
It'll all be pretty dark - you're working at a small effective aperture, about Eff/20 wide open - though that's a couple of stops wider than your MPE at 12x!.
You DO NOT stop the tube lens itself down, so you get NO "auto diaphragm".

You CAN drive the camera with software to make it focus the camera lens, which shifts your whole focus, in steps. It can be driven from a laptop or phone. That may be the best way for you to get the stack done, while you hold still, on whatever support you can conjure. (Focal depth range is limited before the optics don't like it, doing it that way.)

Otherwise, I'd go for a "stick". Like a monopod, but you just hold the stick in the same grip as the camera, and use it to stabilise you while you gently rock forwards. I'm sure you have your methods!
I imagine you'll be near the ground, so the stick could be your other hand, in a fist on top of a dumpy table-top tripod, or something.
Something to slide along like a table mat, works too. There's a video here(to be found) of someone doing that at 10x.


Option 2:
I would consider a different approach too:
Several cameras will take a rapid series of photographs, sometimes in a focus-shifting stack, on a single press of the button. 60fps gets the series done without you moving much - which is what you want.
Olympus allow it in their recent mirrorless series, and many compacts do. My 1" sensor Lumix does it.
If you start with a small sensor, (some are 6mm x 4mm) then you don't need to magnify so much. You can use their lenses at their "long" end, with a close-up dioptre on the front, which can be eg a Raynox MSN-202 (40mm iirc) or a microscope objective.


It can be a battle to work out what magnification you wind up with.
One way is to dig out the actual (not "equivalent") focal length of the camera's lens at the long end, and the sensor width.
Then, find the FL of the lens you want to put on the front.
If that's a "M times" microscope objective designed for a 200mm tube lens, the FL is
200/M.
Call that Focal Length front or FLf

If the camera's actual "long end" Focal Length is FLc mm,
Then the Magnification you actually get is
Mact = FLc/FLf

If the sensor is W mm wide, the field of view you get is

W/Mact.

Hope that way makes sense, there are many ways to twist it.

Therefore if you use a Raynox MSN 202 (40mm) with a superzoom compact, you get a very small field of view. Your compromise on sensor size, can be anywhere!
Chris R

Lou Jost
Posts: 5942
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

I'm sure you've thought of this before, but wouldn't it be a million times easier to rotate the object (or its support) instead of the camera? You could then use all the cool toys....

donkom
Posts: 8
Joined: Tue Nov 15, 2016 6:09 pm
Location: Ontario, Canada
Contact:

Post by donkom »

Chris, thanks for the continued information!

I have found a novel way to test my magnification in a practical sense.... just photograph a ruler and count the millimeter markings. I know my camera sensor is 36mm wide, so if I see 10mm, then I know 36/10 = 3.6, which is my magnification. It's simple, but it works!

So let me revise my understanding:

The objective is not attached to 200mm of extension tubes, it's connected to an actual lens. I have a Canon 180mm macro lens, so I'd set this lens to infinity focus and go from there?

I still don't know where a Raynox DCR-150 would fit into the mix, nor the name/brand/place to purchase an iris for the microscope objective (unless this is already built into the objective and I am misunderstanding).

I've got a system down pat for getting focus, and for taking hundreds of images of the same snowflake. No worries there - my main concern is about getting the equipment to function in a practical way.

Lou, Thanks for the compliments and I'll be sure to post the results here. I have some equipment capable of rotating the subject, and I've done some successful experiments with that... but it still takes more time than I would like. Typically when "good snow" is falling, the conditions last only 10-20 minutes. I need to maximize the number of crystals I can image over that time! I've done the shooting process well over a thousand times, so I've got a system that works. :)

donkom
Posts: 8
Joined: Tue Nov 15, 2016 6:09 pm
Location: Ontario, Canada
Contact:

Post by donkom »

Okay, so I'm learning more! But not in the right direction.

Some research shows that the iris diaphragm you're referring to is for transmitted light microscopy and is built into the stage. All my work is reflected light, so there would need to be some mechanism in the lens to stop things down. Any suggestions - is this even possible?

ChrisR
Site Admin
Posts: 8668
Joined: Sat Mar 14, 2009 3:58 am
Location: Near London, UK

Post by ChrisR »

Photographing a ruler is the standard way to check magnification. If you use a more clever camera with a smaller sensor, you need to know what supplementary to put on the front, hence the calculation.
I still don't know where a Raynox DCR-150 would fit into the mix
See that FAQ I linked to.
Think of the bundles of light coming from the back of the objective as being parallel ( to focus at "infinity") so they need to be focused into a sensor. If you use a 200mm lens to do that, you get the marked magnification. (A DCR 150 is 208mm FL)

A camera lens is designed to do just that.
Your 180, focused at infinity end, should be ok - results can be a bit unpredictable, & I don't remember someone quoting results from that one. Being shorter than 200 you may start to lose more of the corners. Some camera lenses cause vignetting anyway, but primes are usually OK.

If you used something physically short like a 200mm enlarger lens, or 200mm FL close-up lens like a Raynox, you'd have to put it at the right distance from the sensor so it would focus at infinity, that's where tubes would come in. Then you use adapters to fit the objective on the subject side.

Whichever you use, this
the iris diaphragm you're referring to is for transmitted light microscopy
is totally wrong :) -.
Ideally you'd like to stop down the objective, but few allow that so you have to put the iris as close as you can, which is right behind it. The usual ebay seller is "jinfinance" who sells a few different adapters with an adjustable iris, which take objectives or M42 one side and are either M42 or a filter thread such as 52mm the other.
You need to research to find the right adapter rings - most relevant ones are given in the FAQs. Raynox mounting has its own thread - somewhere...


No connection with microscopes, stages or condensers!
Chris R

donkom
Posts: 8
Joined: Tue Nov 15, 2016 6:09 pm
Location: Ontario, Canada
Contact:

Post by donkom »

Perfect! This is such an eye-opening conversation.

So then, what is the filter thread size of a microscope optic? Something so basic eludes me.

I'll need:

- Microscope Objective
- Thread adapter from objective to 58mm
- Assuming this is 58mm on both sides, I need this guy: http://www.ebay.com/itm/IRIS-DIAPHRAGM- ... 0680323665

- 58mm to 72mm filter thread

and then I'm off to the races? Ideally I'd like to find some way to mount my ring flash at the end of the microscope lens, and I've got some ideas to make that happen unless you've got some sort of secret you can share! :)

Post Reply Previous topicNext topic