Building a infinity corrected lens setup for FF camera

Kurt · Post by **Kurt** » Fri Feb 26, 2016 2:40 pm

Hi all,
this is my first post. Since a long time I'm following this forum and have found many great information ans inspiration for my work herein.

I'm about to take the next step in my macro photography by moving from finite corrected lenses (mainly Macro Nikkors) to infinity corrected lenses. My intention - based on so many great images posted here - is to go with Mitutoyo M Plan Apo lenses.
As camera I'm using a Nikon D810, which I use with raw-files. So I've got two questions, which I couldn't fully answer so far:

1. Tube lens: Which is the best tube lens to cover the entire FF format with the best overall performance?
I think Rik did a great test on that here:
http://www.photomacrography.net/forum/v ... +tube+lens
Based on these results I'd go for the Raynox. Is this still valid? Does the Raynox need to be mounted reversed?

2. Tube setup:
I've read many posts on that and I particularly like the setup Chris S has shown here for Mitus: http://photomacrography.net/forum/viewtopic.php?t=12880
But I think that more wide tubes are needed to prevent vignetting on FF. Is this correct? Are there any good ideas what kind of tubes could be used? I've found at thorlabs their 2'' tube system (http://www.thorlabs.us/navigation.cfm?guide_id=2070), which looks not to bad. But I actually struggle putting the parts needed together. I'm open to any idea. Throbs system was just the first tubes I found, which are wider than M42 or T mount.

Any help is highly appreciated!

Chris S. · Post by **Chris S.** » Fri Feb 26, 2016 5:45 pm

Kurt, welcome to the forum!

The tube lens setup you referenced (mine) does not vignette on full frame. As you’ve observed, it’s based on Edmund Optics Techspec T-mount components, which have a clear aperture of 37mm. The Thorlabs SM2 tubes you referenced have a clear aperture of 50.2mm, and so would also work well. A nice thing about Thorlabs tube components is that they cost a bit less than similar items from Edmund Optics. And Thorlabs keeps adding useful things to this line based on customer requests. Somewhere around, I have most of a parts list I put together for a tube lens assembly based on Thorlabs components, including a Thorlabs ITL200 tube lens. I haven't built this assembly (I don’t need it), but couldn’t resist thinking about it. I’ll see if I can lay hands on my notes for you.

Edmund Optics and Thorlabs are the two best sources I know for lens tubes, each with an extensive line of connectors, adapters, irises, beam splitters, and other parts you might need. Sadly, the two company's lines are largely incompatible with one another. Newport has a more limited line, at least some of which looks as if it might be compatible with Thorlabs lens tubes. This is something to keep in mind if you need a component that Thorlabs doesn’t make. (Though as mentioned above, they seem to be willing to make things at user request.)

If you build one of these assemblies, make sure to get some flocking material to cover the inside surfaces—this is very important to preventing flare. I use Protostar Hi-tack flocked light trap material (self-adhesive) for this.

Your use of a full-frame camera limits your choices for a converging lens (aka "tube lens"). Something Rik demonstrated in the tests you referenced is that the Raynox close-up adapter does a better job covering a full-frame sensor than dedicated tube lenses from Mitutoyo, Nikon, and Thorlabs. (I’m not convinced that the Raynox is quite as good on an APS-C sensor, however; nor do I like the way the Mitutoyo objectives cover a full-frame sensor—so my personal emphasis is on getting the best APS-C images possible.) So for a tightly-integrated setup like mine on full-frame, you’d go with a Raynox. The other possibility would be a 200mm Nikon camera lens; this would offer very high quality, but not the tight integration.

Integrating a Raynox into Thorlabs tubes is not trivial, however. I’ve spent a bit of time pondering how to do it. The Raynox is too big to fit inside a Thorlabs SM-2 (nominally two-inch) tube, and bumping up to an SM-3 (nominally three-inch) tube seems clunky. I’d be tempted to remove the Raynox retaining ring and see if the elements could be removed and mounted with retaining rings inside the SM-2 tube. (I don’t have a Raynox to experiment with, and taking the Raynox apart would entail some risk of destroying it.)

Your timing is interesting. A couple weeks ago, in private correspondence, I wrote as follows regarding a Thorlabs-Raynox integration:

It would be easy to add an iris, analyzer, and beam splitter. Or separate tubes could be set up for the DCR-250 (125mm) and DCR-150 (208mm), and arranged as drop-ins. This would, of course, necessitate putting the camera on a moveable stage. A small Velmex UniSlide with Rapid Advance, without a lead screw, comes to mind.

Much as I’d enjoy building and testing such a thing, I don’t need it. Which makes the expense something of a waste.

The possibility comes to mind that if you'd care to pay for the parts and any machining I need to send out, I could integrate this, test it, and send it to you. Not entirely sure I want to commit to this, but if interested, we could discuss it offline.

--Chris

jcb · Post by **jcb** » Sat Feb 27, 2016 12:31 am

Comparing Edmund T and Thorlabs S2 tubes on Nikon full frame cameras (24x36mm sensor, 43.3mm diagonal) :

Edmund T system requires a T2 ring to fit on the camera, the worse restriction appears to be the rearmost Edmund tube clear opening inside the T2 ring (36.2mm diameter, 52.0mm from the sensor).
With Thorlabs S2 tubes and a Nikon BR2A, the most significant restriction is the rear part of the BR2A (41.5mm diameter, 42.0mm from the sensor).

Some trigonometric computations later (similar triangles) I get the following results for the maximum non vignetting rear pupil size at 200mm from the focal plane :

With the Edmund T system : 27.8mm.
With the Thorlabs S2 system : 34.7mm.

So, Thorlabs will accept bigger lenses but if your lens has an exit pupil that is not bigger than 27.8mm Edmund T system wil work perfectly.
Figures look as if they are precise to 1/10th of mm, please take it with a grain of salt as an order of magnitude.

Added : If it is easier to adapt the lens to Edmund T system, it is possible to go up from T to S2 with an appropriate adapter from Thorlabs.

Kurt · Post by **Kurt** » Sat Feb 27, 2016 12:01 pm

Thanks for the quick and informative replies!

With respect to your setup Chris S.:
I thought that the MT-1 tube lens won't cover full frame. As far as I could find the image circle of the MT-1 is approx. 30 mm. This would be not enough for full frame (> 43 mm needed). Therefore I thought that you set was made for APS-C only.

My intention is to get the most out of the objectives as possible. Therefore I thought that the Raynox would be the way to go. If there's anything out there, which is better as a converging lens, I'm happy to take that into account.

In general:
Chris you wrote, that you aren't entirely happy with the coverage of the Mitutoyo lenses on full frame. Could you and / or somebody else comment on that with some more details. My approach is to get the best possible "optical" setup for my work between 5 x and 50 x magnification. I got the intention that the Mitutoyo M Plan Apo would be the best choice for that.

To the tubes:
I thought - as jcb calculated (thanks for that!) - that a wider tube system would help not to run into vignetting issues. Therefore my search yielded more or less only the thorlabs SM2 system as the tubes of choice.

Chris,
if you'd find your notes any help would be highly appreciated.

For the integration of the Raynox lens into the SM2, I already feared that this would be the tricky part.
Unfortunately I haven't found, if I need to integrate the DCR 150 reversed or non-reversed. For a non-reversed integration, I'd probably go in first instance for something like (starting from the camera):
Nikon BR2a (approx 4 mm thickness, if I remember correctly)
SM2A52 (for connecting the Br2a to the tubes, < 1 mm thick)
SM2L30 (77 mm extension)
SM2L20 (51.6 mm extension)
SM2V15 (13.5 mm + 33.3 mm variable extension)
SM2T1 (no extension added converts external SM2 thread to internal SM2)
SM2A27 (stepdown ring from SM2 to M49x0.75)
Ebay m49 to m43 step-down ring (http://www.ebay.de/itm/KAMERA-RING-STEP ... Sw9N1Vw2Vg)
Raynox DCR 150
m49 to Mitutoyo-thread adapter (http://www.ebay.de/itm/RafCamera-M26x0- ... Sw37tWErab)

This setup should lead to an adjustable tube length that should allow the DCR 150 to be focused to infinity, if I haven't calculated wrongly.

One weak point is that it's impossible to change the distance between the objectives and the converging lens.

Further ideas are highly welcome!

bobfriedman · Post by **bobfriedman** » Sat Feb 27, 2016 12:53 pm

you might check out my post http://photomacrography.net/forum/viewt ... highlight=

Kurt · Post by **Kurt** » Sat Feb 27, 2016 1:04 pm

That's helpful!
Do you use this setup with a FF camera and how well does it work? Especially corner performance compared to the middle?

With respect to the DCR 150: You're using it reversed. What is the benefit and how big is it?

bobfriedman · Post by **bobfriedman** » Sat Feb 27, 2016 1:33 pm

you could check some results here http://www.pbase.com/bobfriedman/new_macro_trials

setup looks like this..

Kurt · Post by **Kurt** » Sat Feb 27, 2016 2:28 pm

These images look promising, although they are quite reduced in size. Could you share some crop from centre and corners of the different Mitutoyo lenses? I'm planning to switch to M Plan Apo 5x, 10 x, 20x and 50x.

Based on your post, I'll try to redesign my tube setup for reversed mounting of the DCR 150.

What is optically better:
Placing the objective as close as possible to the Raynox or in a certain distance. I thought that there was a thread about that, but I couldn't find it.

ChrisR · Post by **ChrisR** » Sun Feb 28, 2016 2:01 am

Yes the posts can be hard to find

Reversed, and with a gap of 50mm+ is what I remember.
Finding that reversed is better, came after the initial trials which suggested the normal way round, but, I think, that was without the gap.

mawyatt · Post by **mawyatt** » Sun Feb 28, 2016 6:19 am

Kurt,

I have both a 42mm based tube/relay lens and 52mm. These use the Raynox 150 reversed. I created the 52mm because of slight vignetting I was seeing on my D800 with the 42mm based lens.

All the tubes and adapters are cheap eBay. The tube interiors are flocked with Protostar. I found the cheap eBay Nikon F mount adapters are weak and don't support the long lens well (see 42mm lens breakout), the Nikon BR2 52mm reverse lens adapter is much better and fits the 52mm tubes directly.

You can find the 52mm tubes on eBay sometimes in 7, 14, and 28mm extensions. I recall paying $6~7 for all three sometime ago. The threads are not precise, so work them well before using in your lens.

The 42mm has an extra section in the image below, so ignore the length.

Cheers,

Mike

mawyatt · Post by **mawyatt** » Sun Feb 28, 2016 6:37 am

Kurt,

The long 200m tube/relay lens is supported with the Hejnar 12" rail (G010-12) & LLSB support bracket. The multi-colored rubber bands keep the lens assembly firmly pressed onto the LLSB bracket 2 rubber wheels.

Best,

Mike

Charles Krebs · Post by **Charles Krebs** » Sun Feb 28, 2016 3:30 pm

Another supplier to consider:
http://sciencecenter.net/hutech/catalog ... g-2010.pdf

They have tubes with a 53mm inside diameter, and also camera mounts that are much "wider" internally than the standard T-mounts. Their product quality is high. I've never compared costs to Thorlabs or Edmund, but worth a look.

A few things to start your quest in their catalog:
Page 5, camera mounts
Page 8 tubes (part # 7601-7607)
Page 11 (7615, 7616)
Page 12 (7061, 7060. Ring clamps with 1/4-20)

Kurt · Post by **Kurt** » Sun Feb 28, 2016 3:50 pm

Thanks very much for all the input.
though, based on the last input, I changed the items list:

Nikon BR2a (approx 4 mm thickness, if I remember correctly)
SM2A52 (for connecting the Br2a to the tubes, < 1 mm thick)
SM2L30 (77 mm extension)
SM2L20 (51.6 mm extension)
SM2L05 (13.5 mm extension)
SM2V15 (13.5 mm + 33.3 mm variable extension)
SM2A26 (converts SM2 tube to M49x0.75)
----
This should sum-up to: 46.5+4+77+51.6+13.5+13.5 = 206,1 + 33.3 mm variable extension.
Is this okay for a reverses Raynox DCR 150?
----
Rayons DCR 150
Step-down adapter m43 to m52
M52 tubes (up to 50 mm - optional, couldn't find one on eBay or elsewhere. Any hints are welcome)
M52-M52 female->female adapter is missing.
Has anybody an hint for me, where to find such ring?
m52 to Mitutoyo-thread adapter (http://www.ebay.de/itm/RafCamera-M26x0- ... Sw37tWEqNV)

Kurt · Post by **Kurt** » Sun Feb 28, 2016 3:51 pm

Thanks Charles,
I'll have a look!

ChrisR · Post by **ChrisR** » Sun Feb 28, 2016 4:07 pm

M52-M52 female->female adapter is missing.
Has anybody an hint for me, where to find such ring?

eg http://www.photomacrography.net/forum/v ... 172#163172