Reichert U.S.A. and Leica. Why the undercorrection?

[apochronaut]

When Cambridge Instruments and Wild-Leitz merged in 1991, thd newly named Leica carried forward an AO design strategy of undercorrecting objectives slightly, then completing the correction downstream in either the telan lens or the eyepieces. Apparently, Leica still does this today.
Here is the history.
In the early 60's AO introduced it's infinity system. The 34mm parfocal objectives were completely corrected for ca wihin their colour correction category , using neutral eyepieces. Beginning in the early 70's, Reichert followed the lead of their then parent company and introduceed a 45mm parfocal system that is also completely corrected for ca, also using neutral eyepieces. In fact you can lower an AO 10 stage slightly using the aluminum lever on the side , install Reichert Austria 45mm objectives in the nosepiece and the image will be flat and perfectly corrrcted to the edge of the field and beyond using the stock AO #176 eyepieces, made since the early 60's.
However when AO developed their last system , the series 400, they engineered an undercorrection into the objectives that must be corrected downstream. Thus when you use either older AO 34mm objectives or Reichert Austria objectives in a series 400 microscope, that downstream correction causes them to be slightly over corrected and you have to use compensating eyepieces to neutralize the inverted lateral ca. When Leica began production of the Delta optical system in 1991, they patterned it after the newer AO/Reichert U.S.A.45mm parfocal system , rather than the older fully corrected system.
Does anyone have any idea of the technical reason for this deliberate under correction?

[PeteM]

I've never seen an explanation of this, but I can offer a hypothesis. The buyout occurred at an early time of financiers buying up companies and trying to assemble mini-conglomerates. That sort of management would very likely have wanted to invest as little as possible in R&D. Thus, bringing forward the old 45mm DIN RMS optical designs (e.g. MicroStar IV era) into a larger barrel and with minor tweaks might have seemed attractive. I suspect that you and/or Wayne B. would likely be the ones best able to sort out the design decision.

In other words, it may have been more a financial decision than a technical one?

[apochronaut]

Both design teams had the designs based on a complete correction model at their fingertips. The Reichert Austria wider field designs were 30mm f.n. too.
It seems it was a choice for a technical reason.
With Leica, perhaps it had somerhing to do with Confocal.

[PeteM]

I was thinking it more a matter of production costs - that they might be able to use some existing designs, machines, or processes. It's purely conjecture. Someone intimately involved would know better if any IP, factory preferences, production processes, or machinery carried over.

Don't know Leica's confocal history. I do know that Olympus' first confocal effort, at least on the BX40, was designed by a Bay Area optical research effort - I ended up talking to the electro-mechanical designers who offered me their first prototype. That work, done decades earlier, was perhaps a year or so past the 1993 introduction of the BX. If Leica was on that same time schedule, considerations of confocal might have come later than the DIN 45 to Delta change.

Cost-wise, I suspect confocal would have been so a small niche that the bean counters in charge at that time might not have cared. Were it the technical guys, it could be they were looking ahead. However, it seems to me that having corrections in the objective and tube lens would also be best for confocal.

Still a mystery. Perhaps Wayne will see this and know some of the details.

[apochronaut]

With the series 400, the planachros were 45mm barrel tweaks of the fully corrected 34mm objectives for some of them and really, that is just a matter of lengthening the barrel but they altered the corrections too and designed a group of new planachro and then planfluor objectives on top. They weren't saving any money at that point. Further, I have exactly the same objective, a planfluor 40X .70 in the Reichert U.S.A. incarnation and the Reichert Austria incarnation as a phase objective. The two in their own systems have identical performance but when you swap them over to the other system, one is overcorrected and the other undercorrected. Both made in the same factory around the same time. That isn't a money saving procedure. The difference in production costs between the two would be minimal. One doublet would be a tiny bit more dispersive than it's counterpart in the other objective. They still had to engineer two separate objectives, possibly procure a special glass for one of them, grind the two unique elements in a separate production, then cement the doublet in a separate production, then assemble and test two different objectives under two slightly different protocols. It seems like it would be more expensive to produce two overlapping systems with slightly different corrections. There must have been a reason : perhaps in engineering, perhaps in marketing but there was a reason beyond production costs.
With Leica, I can see that tweaking the reference length out to 200mm made sense. They were going to be designing a whole new system of next generation microscopes so why not increase the length closer to the theoretical maximum to add more room in the infinity space. That twenty mm could be useful. The second thing is : how long did Leica have to introduce a new range of microscopes? Not very long. Optical designs take time. Barrel designs don't. They can be designed and tooled in a few days.
Has anyone tried out any of those Delta objectives in a 182mm microscope to see if they need further correction?

[wabutter]

First let me clarify from my perspective the relationship the different factories had in the Leica family of products. Lots of points to reference here, I hope I can keep it straight.

The Buffalo infinnity based products. those that carried the AO/Reichert-Jung branding, used uncompensated eyepieces. even the Series 400 was uncompensated. The corrections were in the objective and the tube lens. Moving from a 34mm parfocality platform to 45mm gave the factory even more roon internally to make corrections for ca and flatness. Nikon CF optics (Chrome Free) marked a movement toward uncompensated eyepieces with the correction in the objective and tube lens. As Phil pointed out, AO was first to make the committment to an infinity corrected system.

Reichert (Vienna) being owned by AO followed suit with infinity systems with the introduction of the Univar in the early 70's and subsquently the Polyvar and MeF3 for the inverted platform. They followed the same reference focal length as the AO platform, but the platforms were widefield photo scopes with a FN of upto 30mm, the eyepieces were Wide field Plan Compensated. (WPK). Most of the ca was corrected in the tube-lens, that was integrated with the magnichanger. I think the compensation was really part of the Plan correction for the huge field of view. BTw, the FN to the camera back, if I remember correclty was 24mm. Unheard of in that time frame. Point of information: The objectives were hand fitted to the nosepeice of the Polyvar and Mef, so the parcentration was less than the diameter of white blood cell. 100 micrometers and the parfocality was less than 10 microns. Truly incredible specs.

Leitz also had an infinity correctly compound system as early as the 70's for their industrial platforms such as the Ergoplans, Ergolux and MM5/6 as well as the metallurgical upright microscopes all used infinity optics that were based on a 250mm reference focal length. In the early 90's under the Leica brand, the DM L, and DMR platforms were introduced using the Delta optics at a 200mm reference focal length. The eyepieces are not marked as compensating, so I beleive most of the compensation takes place in the tube-lens. Later in the 90's the HC varient was offered, HC meant Harmonic component. I was meant to provide some additional development opportunities some of which were driven by Laser Scanning Confocal technology. Leitz and later under Leica, Confocal systems became increasingly important. So much so that there was rapid development in multiphoto systems and specialized objectives. The Wetzlar factory in some cases had two different catalog numbers for the same product, on for widefield applications and one for confocal. This created two separate business units that could operate under their own P&L

Lastly, it might be helpful to understand the political aspect of the different factories. As consolidation of the product lines occurred. each of the factories struggled for dominance and survival. Stereo production got moved to Herrbrugg, and to Leica owned factories in Singapore and China. Buffalo heard the bell tolling. Wetzlar was dominating the compound microscope platforms, and was arrogant enough to not even consider dual and multiple viewing options that were paramount to success in the clinical pathology market. Wetzlar wouldn't even consider integrating a Buffalo based, turn key diagnostic solution for gout analysis for the new Laboratory scope. I can assure you that the optical design concepts were not shared or incorporated in Wetzlar. Even less so, with Vienna. Once the DM L was introduced, the Microstar/Diastar line were phased out. Only Special products and some OEM products were left and Buffalo was eventually closed. Vienna is still in operation, they have a robust Electron Microscopy Specimen Prep product line. The worlds best ultra microtome.

Side bar, the story goes that K Reichert was related by marriage to the Leitz family had a falling out and created Reichert Optics in Vienna. Long history of animus there. In some ways it is really too bad. Many of you know that both AO, B&L and Reichert were recognized form many first and certainly innovation in the field of microscopy. LIke many other product areas, we have lost that capabilitiy in the good old USA.

[apochronaut]

Well, I'm still up in the air why the 34mm Buffalo objectives and presumably those from Vienna made for models like the Neovar 2 and the 45mm Vienna objectives have the same corrections but the later 45mm Buffalo objectives don't. In fact I have used 34mm AO objectives on 11mm extensions, 45mm Reichert Austria and R..M.S. Nikon and NIS objectives in the same stand and they have all performed equally well corrected with the same neutral eyepieces. I can set up a Nikon 25M planapo independent of the others and the results are the same.
If I put those objectives in a Series 400 stand from Buffalo. a compensating eyepiece is needed, otherwise all four objective types are over corrected. Same with the Nikon planapo. This is in actual tests , using highly refractive diatoms to display as much ca as possible. If I put the Buffalo objectives in the former stand, they are under corrected. The difference is slight but it is there. Just trying to figure out the reason why that small under correction was installed in the Buffalo system .

It seems the same is true in the Delta system. I would be interested to know if anyone has used a Reichert Austria objective in a Delta based stand. By my reckoning, they should require compensating eyepieces.

[PeteM]

Here's a shot at summarizing the discussion of eyepiece corrections in a table. There are surely omissions and likely errors. Let me know and I'll update it.

.

AO, Reichert, Leica eyepiece corrections.png (259.07 KiB) Viewed 677 times

[Scarodactyl]

I understand why this was necessary in the 60s and earlier, but I do find it odd that Nikon was able to manage so well without additional corrections in the 80s, followed by Olympus and Mitutoyo, while Leica and Zeiss cling to it still. I kind of assume Leica and Zeiss should have the best engineers. Certainly if production is cheaper or more efficient this way though the savings aren't being passed on to the customer. I have just assumed that it is a purely cynical move to keep their systems proprietary (similar to moves towards making products more disposible), but I could easily be missing something critical on the optics engineering side.

. BTw, the FN to the camera back, if I remember correclty was 24mm. Unheard of in that time frame.

That is really cool to hear. It amazes me how many companies made and make objectives with huge corrected fields and no way to out them on a camera, so I'm glad at least one engineer somewhere in Vienna had some good ideas. Way ahead of his time.

[PeteM]

"I would be interested to know if anyone has used a Reichert Austria objective in a Delta based stand"

Here's a quick test using a Delta-era Leica 501055 head with L Plan 10x/20 eyepieces. The stand is a DMLS that just came in. It has a yellow hot spot in the illumination—maybe the wrong bulb or illumination in need of collimation. You'll also see some barrel distortion in the higher magnification Android phone pictures. As far as I can tell, this is due to the phone camera—the view through the L Plan 10x eyepieces shows straight lines of the micrometer slide out to the edges.

I tried the following Reichert Austria infinity objectives

Reichert Austria Plan 10x 0.25na - with or without cover slip
Reichert Austria Plan Ph 40x 0.75na 0.17 cover
Reichert Austria SPl 63x 0.80na 0.17 cover

Visually, they looked reasonably plain and free of chromatic aberration. Attached are some small .jpg cell phone shots of the 10x, 40x, and 63x.
. .

[apochronaut]

Here's a shot at summarizing the discussion of eyepiece corrections in a table. There are surely omissions and likely errors. Let me know and I'll update it.

.
AO, Reichert, Leica eyepiece corrections.png

edit. I wrote all this before I saw your pictures but they confirm my suspicions.

I don't think the table represents the situation the way it comes out in my actual testing. There is also some information in various patents.

Firstly the F.N. of the various systems. F.N. is a confusing spec. It seems that in some references to it, the F.N. is objective dependent but in other references it is eyepiece dependent. Then there is the issue of the apparent field vs the actual field, not to mention the limitation of the ocular tube or the magnification of the eyepiece.
The B & L 15X U.W.F. eyepieces are a good case study. If you look at B & L literature related to their 15X U.W.F. eyepieces, the F.N. is listed as 20mm. They see the same field as the 20mm F.N.10X eyepieces do, yet becsuse they are 15X in order to achieve that, the image circle or apparent field is 1 1/2 times that of their companion 10X 20mm F.N. eyepieces. It serms then that to equalize the ground in terms of F.N., it should be the field stop of the 10X eyepieces that determines the F.N. , not the ocular tube size or apparent field , or theoretical objective image circle, and certainly if the objectives are being limited by the diameter of the optical tube.
So, in the case of each system you portray in your table I think it makes sense to establish an F.N. for the system plus a theoretical F.N. or objective image circle.
Secondly, the corrections. Corrections are very subtle because there are numerous aberrations that need to be corrected for and each one requires specific corrective attention that is potentially deleterious to another aberration . i think if any one of us saw the math associated with the evolution of a single eyepiece highly tuned to an
objective family, our jaws would resemble rhat of Eben Byers. When there is talk of correction, it almost always means correction for chromatic aberration. Nikon has carved a career out of trumpeting "chrome free" but did they ever say spherical free or curve free? No, because in order to correct completely in the objective for chroma, you have to undercorrect for spherical aberration and visa versa. Chroma correction gets one's attention because it is obvious. If you put an objective in a system that it isn't ca corrected for, anyone would notice that. If you put one in a system that it isn't sa corrected for, almost no one would notice that. They would go "wow these objectives aren't as sharp as the ones that came with the scope". Curvature of field? It is generally overlooked if slight but : in order to correct for ca in the objective , you would likely have to install other aberrations that would need to be corrected in the telan lens or a combination of the telan lens or the eyepiece. Leica is just being honest in stating the truth of optical design.
Your table needs some refinement.
The AO Cycloptic and Series 2/4 were not 100% harmonious in their eyepiece requirements and the series 10 needn't be in that section at all. The Cycloptic used a 147B 15X eyepiece, so there is something different about it's optics. For one thing it is infinity corrected. I would limit the section on the series 2/4 to just that,, the 2/4. It is relavent to the general pogression because it uses a corrective tube lens and has a tube length the same as the infinity systems, probably 7 3/16" originally, so just a bit over the 182mm mark. They didn't serm to be aiming for a f.n. target, probably doing as best they could with what they had. The objectives and eyepieces evolved in the late 1940's with the tube lens a further refinement that installed ca and curvature corrections, while extending the tube length without changing the magnification.. The eyepieces are close to neutral, maybe slightly corrective for ca and curvature. The field is about 19 .25 mm. A 34mm parfocal system.

The series 10/20 was the first professional infinity system and used 34mm parfocal objectives. They also introduced student lines as well. Originally, the objectives were close in design and planarity to those of the 2/4 but they seemed to know where they were headed because the #176 eyepiece was tuned to plan objectives and is very slightly compensating. It is about as compensating as the 146 is correcting but I would have to do a really careful and thorough assessment of all the AO Reichert eyepieces because they are all very close. The refinements are extremely subtle and lean into field curvature adjustments heavily, once the era of expected planarity arrived in the 60's.
Later in the decade, AO introduced the #176A eyepieces that were teamed with the standard achromats while the 176 went with the plan achromats.The ca corrections seem identical but there is an attempt to reduce the field curvature of the standard objectives with the 176A. .There was a limited release of infinity apochromats in the early 60's too and the 176A work well with those giving an unusually large 19mm field, which was a lot with apos in that era. The 176 are also 19mm.

In the early 70's Reichert Austria developed a 45mm parfocal infinity system which from my tests use the same corrections as the AO 34mm system. The Austrian system, since in the day Austria had about 1/2 the per capita GDP of the U.S. was created much more economically than it could have been done in the U.S. In fact Austria also introduced a 34mm parfocal system in the mid-70's. It was on the Neovar 2 platform, wheras the similar but a little more elaborate Diavar 2 used the 45mm objectives.The two systems might be roughly compared to the 100 and 400 series out of Buffalo. I don't know off hand what eyepieces were used in those but they were 23.2 mm tubes and likely WP 10/20. With the larger microscopes, the Univar and Polyvar using the same objectives , the eyepieces were WPK 10/24. I doubt if the K addition on those refers to compensation for ca. More likely it is teamed with plan as in plan compensating, due to the extra 2mm of peripheral field that the same objectives as used in the 20mm Diavar 2 had to cover.in the 30mm ocular design. There may have been a tweak of the curvature required. I can't really test that at the moment.
Anyway, despite the Uni and Poly microscopes having a 30mm eyepiece tube, they produced only a 24mm f.n. visual field with the 10X eyepieces. There may be field curvature limitations beyond that , even though the eyepiece tube would allow that.
In terms of corrections, when I install objectives from either the Univsr or the Polyvar in an AO 10 stand and use #176 eyepieces, they are perfectly corrected and flat, so my conclusion is that the corrections are the same. If I use Nikon CFW 18mm f.n. eyepieces, same result.
The AO 34mm plan objectives I have dug into by the way, Pete have an f.n. of 24mm, as per patents for them. The optical tube and eyepieces limit their plan performance. I suspect , that you could put an 11mm extension on a #1309 40X .66 advanced planachro for instance, thread it into a Polyvar nosepiece and it would be plan to the edge of the 24mm field and perform quite well. I don't know for sure but it would be interestingto try that.

The series 110/120 is just a wider field clone of the 10/20. The optical tube was widened and the # 180 eyepieces equally corrected to the levels of the #176, just with a 20mm f.n.

With the Series 400, things get odd and that was the main reason for the original post. While the eyepiece ca corrections are so close in the #181s used in the series 400 to the 176 or 180, they differ in curvature correction. Using a pair of 176 in a series 400 microscope results in field curvature. and the series 400 objectives are not fully compatible with the earlier infinity systems in terms of peripheral ca. Those objectives are also a 24mm image circle but possibly more fully corrected for curvature of field and or spherical aberration needing less in the telan lens but further ca correction in those corrector plates in the head. I am still assessing this. There has to be a reason for the change in engineering strategy and why Leica has continued so. This ca/sa/fc relationship in objective corrections is the axis of evil that all optical designers juggle. You correct for ca, then you have to deal with spherical and curvature down the line. Corrrect for curvature and you have to deal with ca and sa and so forth. This is the reason for ever increasing objective lengths. More glass choices make it easier but uf you look at some of the older designs, the AO 1029 100X planachro had lenses hanging out the back up into the nospiece and highly corrected even 45mm objectives are stuffed full, so the strategy to get as much correction in the objective results in bigger objectives.
Has anyone used a Nikon objective straight to the sensor at 200mm without a tube lens? I would be very surprised if they aren't correcting for curvature and or spherical aberration in the tube lens and they make heavy use of the embedded front lens design that they probably licensed from AO back in the day?