Exact distance between tube lens and sensor plane?

A forum to ask questions, post setups, and generally discuss anything having to do with photomacrography and photomicroscopy.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

anto74
Posts: 66
Joined: Thu Feb 06, 2020 11:48 am
Location: Italy
Contact:

Exact distance between tube lens and sensor plane?

Post by anto74 »

I am very confused about how to determine the exact distance between the Raynox DCR150 tube lens and the sensor. I use a Sony (19mm flange). Should I use 208.33mm minus 18mm = 190.33mm? With this distance if I focus on a distant object with the Raynox DCR150 it is not in focus. I'm very confused.
I use infinity objective 10X and 20X ( 180mm)
The distance from the sensor to the Raynox DR150 (front or rear lens?) must be that for which if I photograph a distant subject is in focus or 208.33 minus the size of the Sony flange?
Thanks
Antonio

rjlittlefield
Site Admin
Posts: 23564
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

In theory, the "ideal" situation is when the rear lens, by itself, produces a focused image for a subject located at infinity.

When the rear lens is set up this way (to produce a focused image for a subject at infinity), then when used with your objectives the magnification will be increased by a factor of rearLensFL/180. For the Raynox DCR-150, this is 208.33/180 = 1.157, so for example your 20X objective would actually give 23.15X .

However, for 10X and 20X objectives, you can change the rear extension by quite a bit and still get a good image. So for example if you want exactly 20X, you can shorten the extension by about 15% to get that.

To get the magnification exact, you'll have to experiment. Also, the change that is exact for 20X will probably not be exact for 10X.

Be aware that the number 208.33 is the nominal focal length of the Raynox lens. For infinity focus, that distance would be measured from the sensor to a special place called the "principal point". That point happens to be inside the lens, but it is not located at any particular lens surface or mounting threads of the lens.

--Rik

zed
Posts: 72
Joined: Thu Feb 07, 2019 1:35 pm
Location: Houston, TX
Contact:

Post by zed »

Hi Antonio,

From a practical standpoint - I would assemble my tubes so that the Raynox to sensor distance is as close to the 208mm as you can. Then I would attach this to the camera and point it at the moon on a clear night (any phase you can see will do). From there open Live View and adjust the tube distance until the moon is in focus as best you can.

For this to work - the easiest way is to have 2 tubes plus a threaded adjustable tube in between. I recommend Thor SM2 tubes - for example:

1x - SM2M30
1x - SM2M20
1x - SM2T20

In this scenario just assemble the T20 in between the M30 and M20 then you can just rotate the tube furthest from your camera body until your focus is correct - then use the provided locking rings to lock them down.

If you require precise magnification reporting - as Rik has stated you will need to measure this for your specific setup. Even the objectives will vary slightly from their stated magnification. For example a 20X objective might in reality be a 20.8x or a 19.9X.

Personally - I don't care if my magnification works out for a 20X to be 20.8X or something similar. In the case where I need to report this information I use a scale bar in the image - which can be calculated from an image of a stage micrometer slide.

Hope this helps!

-Jason

metebalci
Posts: 29
Joined: Fri May 01, 2020 1:32 pm
Location: Switzerland
Contact:

Post by metebalci »

Hello,

I am having exactly the same question myself for some time, and also asked this to Thorlabs (their support through chat) since I am using their tube lens (ITL200) and also an SM2 based setup.

For my case, from the tube lens to the sensor, it has to be 148mm (it is called working distance). Of course it is impossible to make this exactly with fixed size tubes so I also use an adjustable lens tube.

From the support I learned a few things - and I am not an expert in optics and not experienced using these setups, so only sharing what they said and what I understood -:

- 148mm has to be adjusted as accurately as possible. This is independent of the setup whether you use focus stacking or not. Because I thought, maybe this is important if you only use it like in an optical microscope with eyepieces, but with focus stacking, this may lose its importance, but it seems it is not that case.

- Although this distance has to be (theoretically) set accurately, it is hard to say something about the effect of any deviation without doing some simulation. So maybe, it does not matter that much, also given the fact that microscope objectives have a higher resolution than the consumer cameras attached to these setups.

- I tried to do the same like Jason said, looking somewhere far, moon or mountain etc., without the microscope objective, then adjusting the distance for the best focused image. Then I thought there has to be a better way to do this, so I asked, and the person at support actually said he is doing and actually he recommends to adjust this distance while using the microscope objective. Because, he said, although the idea of the tube lens is to have an object at infinity at focus, it is not designed to be used alone, so it is better to adjust the distance with microscope objective attached (so I guess there is a possibility these two conditions might result differently). So rather than looking to pseudo-infinity, I switched doing this and looking to a resolution target (I have a high-freq one, up to ~220 lp/mm, so it is enough for the camera I think).

- I also asked about the distance between microscope objective and tube lens. There is a large range for this in the specs. So the support person said within these range, it is expected to perform equally well. So I aim to be close to middle of this range, far from the border values.

Mete

Lou Jost
Posts: 5948
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

For our purposes, the distance between tube lens and objective DOES seem to make a significant difference. Many tests on this forum have shown that.

On the other hand, the tube lens to sensor distance is usually easy to set as described above, focusing on a faraway object, and that distance does seem to have a lot of leeway, according to tests on this forum.

rjlittlefield
Site Admin
Posts: 23564
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

metebalci wrote:I tried to do the same like Jason said, looking somewhere far, moon or mountain etc., without the microscope objective, then adjusting the distance for the best focused image. Then I thought there has to be a better way to do this, so I asked, and the person at support actually said he is doing and actually he recommends to adjust this distance while using the microscope objective. Because, he said, although the idea of the tube lens is to have an object at infinity at focus, it is not designed to be used alone, so it is better to adjust the distance with microscope objective attached (so I guess there is a possibility these two conditions might result differently). So rather than looking to pseudo-infinity, I switched doing this and looking to a resolution target (I have a high-freq one, up to ~220 lp/mm, so it is enough for the camera I think).
I am curious to know exactly what you are doing.

For sure if you place the tube lens at 148 mm from the sensor, you'll be able to get a nice focused image. But you'll also be able to get nice focused images with the tube lens at say 160 mm from sensor, or 130 mm from sensor, simply by tweaking the distance from objective to target.

So, how do you know when the "correct" distance from tube lens to sensor has been reached?

--Rik

Macro_Cosmos
Posts: 1511
Joined: Mon Jan 15, 2018 9:23 pm
Contact:

Post by Macro_Cosmos »

Thorlabs quotes the distance between the tube lens (ITL200) and objective should be 70-148mm, mine is usually 25mm away and it works just fine. I've added various stuff in the optical path and it doesn't seem to have an impact on CA. Resolution, yes, not CA.

Supposedly, there should be an optimal distance here anyway. I've said I wanted to find it before, never got around to actually set something up. I suspect the difference will be minimal. I also wanted to test retro VS normal for the ITL-200, never got that done either. I should have plenty time after the 20th of May.

Lou Jost
Posts: 5948
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

I suspect the difference will be minimal.
As I said, there are lots of tests on this forum showing that it is significant, with many tube lenses. It also varies from one tube lens to another. And probably also with the objective or reversed lens too. It may well be that the ITL200 is unusual in that regard and is insensitive to the distance. Purpose-built microscope tube lenses have to be optimized for a longer distance.

But for most re-purposed tube lenses, closer is usually better. There are exceptions though. Forum member nathanm did extensive optical bench tests on this.

metebalci
Posts: 29
Joined: Fri May 01, 2020 1:32 pm
Location: Switzerland
Contact:

Post by metebalci »

rjlittlefield wrote: I am curious to know exactly what you are doing.

For sure if you place the tube lens at 148 mm from the sensor, you'll be able to get a nice focused image. But you'll also be able to get nice focused images with the tube lens at say 160 mm from sensor, or 130 mm from sensor, simply by tweaking the distance from objective to target.

So, how do you know when the "correct" distance from tube lens to sensor has been reached?
--Rik
To clarify:

I have Mitutoyo 5x connected with two adapters to SM2, after 80mm or so (which I plan to increase to ~120mm) there is tube lens, ITL200, then two extension tubes and one adjustable extension tube and then M42 and Sony E mount (very similar to one shown here: https://www.closeuphotography.com/blog/ ... ube-system).

Since ITL200 is supposed to create an image of the object at infinity on the sensor, as far as I understand, it has to stay at a certain distance (148mm) to the sensor plane. Simply, if I remove the microscope objective, and look somewhere far (I mean look on the camera screen) only through the tube lens, and adjust the adjustable extension between the tube lens and the camera, there is a certain point that makes the image sharpest -this is not hard to see, but no idea about the accuracy -. That point is (assuming the far object is a good approx. of infinity) I understand the correct position of the tube lens in this configuration. (and what I have been told is instead of doing the adjustment like this, it is better to adjust it with the microscope lens mounted, and by looking to something like you plan to use it (in very close distance) and adjust the best focus like that)

My problem is, I still cannot understand clearly (I do not know enough about optics) the effect of this distance (the distance between the tube lens and the sensor) on the whole configuration. I simply want to see a ray diagram of the whole system for a few different conditions, but I haven't seen any such yet (I wish there is a very simple model of this so I can play with it).

Basically what I don't understand is, even if ITL200 is said to be put 148mm from the sensor, lets say I put it to 153mm. What is the effect of this 5mm deviation ? 1) Do I harm any optical performance (resolution, ca etc.) of the whole configuration or 2) do I only change the magnification and/or working distance of the whole configuration (which is 5x and 34mm for my 5x objective) ?

If the answer to (1) is yes, then this distance really (theoretically, practically the impact might be limited this has to be evaluated) has to be set as accurate as possible.

If answer to (1) is no, and answer to (2) is yes, then I really do not care the accuracy of setting this distance (OK to be within a few mm) because I do not care if I have exactly 5x or not, and I do not care if the working distance (of the microscope objective) is exactly 34mm or not.

I have the same questions for the distance between the microscope lens and the tube lens (called pupil distance in the specs), but I have less concern about it because I see in the specs it has a somehow large range. However, Lou said above, it might have a visible impact, which I understand but I also do not understand if this would have a visible impact why it would have a large distance in its spec.. I understand that this tube lens can be a special case, it might be less sensitive to this distance (I have also been told by the support something similar), I haven't checked many other tube lenses yet.

I understand all above can be set just by evaluating the outcome/the image, I am just curious what the "correct" setting is, correct in the sense how these pieces (objective and tube lens) are designed to work together.

Mete

Macro_Cosmos
Posts: 1511
Joined: Mon Jan 15, 2018 9:23 pm
Contact:

Post by Macro_Cosmos »

Lou Jost wrote:
I suspect the difference will be minimal.
As I said, there are lots of tests on this forum showing that it is significant, with many tube lenses. It also varies from one tube lens to another. And probably also with the objective or reversed lens too. It may well be that the ITL200 is unusual in that regard and is insensitive to the distance. Purpose-built microscope tube lenses have to be optimized for a longer distance.
Yeah but that's for the Raynox or most 200mm manual teles, I'm talking exclusively about the ITL-200 here.

I've had it very close to the objective from 100mm apart. I've never seen a drop in IQ that makes me wonder if anything is wrong.

Lou Jost
Posts: 5948
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

Yes, that's my point. The ITL200 (and probably other dedicated tube lenses) are exceptional in this regard. Many of us use makeshift tube lenses, and the distance does seem to make a difference for them.

Macro_Cosmos
Posts: 1511
Joined: Mon Jan 15, 2018 9:23 pm
Contact:

Post by Macro_Cosmos »

metebalci wrote: Basically what I don't understand is, even if ITL200 is said to be put 148mm from the sensor, lets say I put it to 153mm. What is the effect of this 5mm deviation ? 1) Do I harm any optical performance (resolution, ca etc.) of the whole configuration or 2) do I only change the magnification and/or working distance of the whole configuration (which is 5x and 34mm for my 5x objective) ?
Theoretically, 1) yes, 2) yes.
Practically, this could be entirely wrong, hence why the Thorlabs engineer suggested you to couple the tube lens with an objective and extrapolate the best configuration via trial and error.

Rob has demonstrated that for 5x and perhaps 10x, reversing the ITL-200 gives better results. This certainly turned out to be true for me, but when it comes to 20x, reversing yielded worse images.

This result is simply baffling. I do have a 5x coming to be soon (Edmund Optics, not Mit), so I might put together a test of my own.
metebalci wrote: I have the same questions for the distance between the microscope lens and the tube lens (called pupil distance in the specs), but I have less concern about it because I see in the specs it has a somehow large range. However, Lou said above, it might have a visible impact, which I understand but I also do not understand if this would have a visible impact why it would have a large distance in its spec.. I understand that this tube lens can be a special case, it might be less sensitive to this distance (I have also been told by the support something similar), I haven't checked many other tube lenses yet.

I understand all above can be set just by evaluating the outcome/the image, I am just curious what the "correct" setting is, correct in the sense how these pieces (objective and tube lens) are designed to work together.
The ITL-200 is a specialised tube lens, designed to offer a long infinity space (distance between objective and tube lens when TL is focused at infinity), so the difference as I've noted -- is not visible enough for me to raise an eyebrow. I've inserted stuff in the optical path for various reasons and applications. While resolution in the corners usually takes a hit, CA and the alike is identical, unless I screwed up badly somewhere (wobbly tubes, Thorlabs' variable tubes are hard to lock in, I'm going to get their slotted locking ring and wrench soon).

For other "repurposed" TLs, such as the Raynox and the Nikon 200mm F4 Ai-s, the objective should be placed close.

There isn't a "correct" setting. You can even shrink the FL of the ITL-200 to shrink the mag of your objective, you can make a 10x MPlan Apo 5x, NA of 0.28 stays the same and the centre will outperfom a 5x Mplan Apo which has an NA of 0.14. I have yet come across a researcher/scientist in the field of photonics that would do that. Even reversing the ITL-200 was seen as "wrong" when Rob asked Thorlabs about it. This is anecdotal though. As long as it works, it shouldn't be considered wrong just because one isn't following a manual.

Lou Jost
Posts: 5948
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

Macro_cosmos, have you been able to use epi illuminators between the objective and tube lens? Or do they vignette?

metebalci
Posts: 29
Joined: Fri May 01, 2020 1:32 pm
Location: Switzerland
Contact:

Post by metebalci »

Macro_Cosmos wrote: Theoretically, 1) yes, 2) yes.
Practically, this could be entirely wrong, hence why the Thorlabs engineer suggested you to couple the tube lens with an objective and extrapolate the best configuration via trial and error.
OK, thanks for clarifying this, I cannot explain but "yes and yes" was also my expectation and it is good to hear it from someone else.
Macro_Cosmos wrote: Rob has demonstrated that for 5x and perhaps 10x, reversing the ITL-200 gives better results. This certainly turned out to be true for me, but when it comes to 20x, reversing yielded worse images.
I also find this result quite interesting actually, and want to consider it as an unexpected side-effect. I mean of course it is possible but I do not want to think there is something wrong with the product (or its specs) or it is even considered to be used like this.
Macro_Cosmos wrote: (wobbly tubes, Thorlabs' variable tubes are hard to lock in, I'm going to get their slotted locking ring and wrench soon).
Slotted locking ring will be definitely in my next order too.
Macro_Cosmos wrote: There isn't a "correct" setting.
Sorry, "correct" was not the right word :) I meant what lens designer had in mind like assumption about the whole system, or most probable use of such products.

May I ask is there any particular reason you decided to use ITL200 and not TTL200-A ? They seem to be similar if not same in performance and use and price, and the latter is ready to be used in SM2 based setup without any adapter.

rjlittlefield
Site Admin
Posts: 23564
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

metebalci wrote:I simply want to see a ray diagram of the whole system for a few different conditions, but I haven't seen any such yet (I wish there is a very simple model of this so I can play with it).
Perhaps these diagrams will help. 10X objective modeled as a 20 mm FL "thin lens"; tube lens modeled as a 200 mm FL thin lens; distance from tube lens to image plane adjusted to be 200 mm (nominally correct), 150 mm (short), and 250 mm (long).

Please note the table at upper left, showing
"Object Distance" (from target to objective),
"Image Distance" (from tube lens to image plane), and
"Magnification" (negative because the image inverts).


First, with nominally correct 200 mm extension. Note that the space between the two lenses is an "infinity section" -- all the rays in this bundle are parallel. The magnification is 10X, exactly as it should be based on magnification = rear FL / front FL. Remember the Object Distance, exactly 20 mm in this case.

Image


Now, use a shorter than normal 150 mm extension. In the space between the lenses, the rays are now converging, not parallel, and the magnification has dropped to 8.5X. (Aside: notice that 8.5X is not equal to 150/20, as you might expect by casually misapplying rear FL / front FL.) The object distance is now longer than with 200 mm extension.

Image


Likewise, use a longer than normal 250 mm extension. In the space between the lenses, the rays are now diverging, not parallel, and the magnification has increased to 11.5X (again, not equal to 250/20). The object distance is now shorter than with 200 mm extension.

Image


I made these diagrams using the current version of WinLens3D Basic, version 1.2.11. It is a free download from http://www.qioptiq-shop.com/en/Optics-S ... Basic.html . The link is in the middle of the page and may be hard to find. It goes to http://www.qioptiq-shop.com/out/picture ... 1.2.11.zip .

The .zip expands to include a Windows installer that is not signed, so you have to grant approval to install an application from an "unknown publisher". (Good grief! Signed installers have been a thing for what, 8 years now? I guess all their people must be busy writing optical design software.)

I've placed a copy of the SPD model file at http://janrik.net/MiscSubj/2020/topic41 ... beLens.SPD . The lens model does not include the window layout, but it does seem to include the "Lens Drawing Options" that preserve the scaling as the tube length extension is changed.

--Rik

Post Reply Previous topicNext topic