My Wild-Nikon-Leitz frankenscope

[viktor j nilsson]

Hi everyome.

I've been meaning to describe my Wild-Nikon-Leitz frankenscope in greater detail for a while. Here it is:



The base is a Wild M20. I got this stand and a mismatched set of objectives for free when I did my PhD at Umeå University in northern Sweden. It had been used by someone at the department of ecological botany - it has some engravings at the base saying so, and a nice 70s anti-nuclear power sticker. It came with a drawing tube and a Wild Phase condenser.

I'm very fond of the stand. In retrospect, it probably would have been a smarter move to get a Labophot/Optiphot with some objectives to begin with. But I've never had a grand plan for this, and I love the engineering and feel of the old M20. It also looks very nice next to my Wild M5.



This spring I found a Leitz Heine condenser on a local auction site and won it - for $20! It lacks the oil cap and has lost a bit of black paint around the light ring, but works well nonetheless.
I've adapted it for the M20 as follows:
I removed the Leitz dovetail.
A 38mm-m42mm adapter ring turned out to be a tight press fit over the shoulder of the Heine. It can be pried away with a bit of force and will only leave some very small marks.
I took apart a cheap Zeiss Jena 1.2 condenser, removed all optics, and epoxied it to the M42-M38 adapter ring.
The Zeiss condenser has the same outer diameter as the Wild condensers (39.5mm). However, I sanded it down a little bit exposing the brass because it was a little tight.
This allows me to keep almost a full range of movement of the centering mechanism.
Getting the Heine really opened up a lot of possibilities, which greatly inspired me to continue to invest in the frankenscope.

I've also aquired the very nice Wild Achromatic-Aplanatic condenser, with 0.95 and 1.30 top lenses. I only had to buy a second complete M20 with Wild achromats to get it! (The second scope was very cheap, though, so the condenser will eventually pay for itself).



My last bulb finally gave up this year, so I DIY'ed a [EDIT:10w] Cree XM-L2 (Cool white) LED illumination, with a 1000mA BuckPuck driver. It's ugly (I don't have a lathe) but works very well. After I acquired the light-eating Heine, however, I sometimes wish I had gone for the 2100mA BuckBlock.



I originally started buying Nikon CF objectives because they are good for macro photography. I don't know how or when it happened, but at some point they began to migrate over to the microscope. Over time, I became completely committed to Nikon optics and I've compiled a rather nice collection, mostly through opportunistic bargain-hunting on eBay. Nothing fancy; no PlanApo's.

At present I have a sixtuple nosepiece for brightfield/HMC (used with achr.apl. condenser):
4x E plan 0.1
10x E plan 0.25
20x CFN plan 0.50
40x E plan 0.65
40x HMC LWD 0.55 w/ correction collar (Hoffman Modulation Contrast)
100x E plan 1.25

And a quadruple nosepiece with phase contrast objectives (used with the Heine):
4x no-name chinese for scanning
10x E achromat 0.25 Ph1 DL
20x 0.40 Ph2 DL w/ correction collar
40x CFN plan 0.75 Ph3



My latest additions is Hoffman Modulation contrast. Just before christmas I was able to buy the 10x ph1 DL, 20x ph2 DL and the 40x HMC objectives for a grand total of £75! I recieved them today and they cleaned up really well. This evening, I constructed a DIY Hoffman Modulation Contrast Slit condenser by cutting a rectangular slit in an old library card. It slides right into the slot under the Achromatic-Aplanatic condenser. The lower part of the condenser rotates, and the library card can be slid in and out, which makes it super easy to align the slit to the modulator in the objective. Right-hand side shows the rear focal plane through the phase telescope.

Most of these things I've acquired over the last year or so. I have a 2.5 year old and a 0.5 year old son, so I currently have precious little time for hobbies. Putting this together - and reading up on things on here and other sites - has been an important outlet for me. I look forward to actually using it in some distant future!

[viktor j nilsson]

Forgot to mention the photo tube. It's very cheap but effective.

From bottom to top:
1) Wild HU photo tube (€135)
2) Asahi Pentax Microscope adapter (€20)
3) Nikon CF PL2.5x Photo eyepiece (€36).
4) M42 male-M39 female reduction ring (€2)
5) M42-M42 31mm focusing helicoid (€24)
6) M42-EOS adapter
7) Cheap EOS extension tube set (2x €5)

I only used the bottom silver part of the Asahi Pentax Microscope adapter (I actually bought one that only had this part). I glued the M39-M42 adapter on top of the Asahi adapter with the M42 male side pointing up. I've screwed the M42 focusing helicoid into this male M42 thread. The focusing helicoid is used to adjust parfocality (in fact, it is exactly parfocal when the focusing helicoid is at its shortest setting - good for stability!).

The Asahi adapter has a rotating bayonet locking mechanism. Since I am not using this, I simply epoxied it together to make it stable.

[MichaelG.]

This spring I found a Leitz Heine condenser on a local auction site and won it - for $20!

^^^ I have the greatest respect for a man of your talent !!

MichaelG.

[zzffnn]

Viktor,

I like your frankenscope very much and respect your passion and talent as well!

I have some questions and comments for you:

1) How do you like the image quality of your Heine at NA 0.95 under circular oblique light vs your Wild Achr Apl condenser at the same NA under offset oblique light, for casual protist photography, subjectively?

I found that my Heine produces too much halos (donut bokeh) for most protists but the thinnest diatoms like Amphipleura. So much so, that in most cases, I prefer to use my Olympus Achr apl condenser at the same NA under offset oblique, because it produces a cleaner and better-looking photos, even though the Olympus condenser does not resolve as many diatom dots as the Heine. I still like Heine's ability to quickly change among phase, oblique and darkfield, even though its image quality is not always better than dedicated darkfield or phase condenser.

2) Your 2.5x photo eyepiece may be too strong for your Canon 7D' APSC sensor? Have you tried using a visual 10x eyepiece (Nikon CFW10) in the trinocular port, then use a 1.4x camera teleconverter lens, along with a camera lens of about 50mm in focal length? That may give you a view field closer to your binocular visual eyepieces.

3) 3W LED is not a lot for Heine at high magnification.

I use a 30W LED that runs at close to 30W, but I am still not getting a shutter speed that is fast enough to freeze motion of fast protists (unless I use flash), at a 65x NA 1.1 objective with Heine+ immersed NA 1.4 oil lens. Admittedly, my LED is slightly too big (not all 30W is efficiently utilized in my system) and my illumination system is not optimized for efficiency (for example, my camera gets only 50% of light from head prism). With Heine, when I add a set of polarizer +analyzer or color filter, my shutter speed gets slow quickly, even at high ISO. With my Olympus condenser, I get better light efficiency and much faster shutter speed.

[viktor j nilsson]

Thank you for your comments, Michael and Fan.

Fan,

1) To be honest, I am also not totally impressed with the image quality at the "intermediate settings" with the Heine. I have gotten my trinocular setup working only very recently, so I cannot really tell you how I like it for photography yet. However, while I enjoy the effect of swooping through phase-oblique-darkfield, I don't think I will use it for COL very much. Like you, I find the halos and "glow" to be disturbing. I think that I will use the Heine much more for phase contrast and darkfield than anything in between.

2) Yep, 2.5x is a bit too much, but as you know its the lowest magnification photo eyepiece that's realistically available. I really like the idea of direct projection, I think I'd rather get a full-frame body if I find that the 2.5x crops too much. Again, only got it running so recently I can't tell for sure. But please tell me, are you describing an afocal setup? Why use a teleconverter? I don't think I've stumbled upon that combination before. Please explain!

3) Sorry, it was a typo (now fixed). It's a 10W xm-l2 T6 led. When I run it at 1000mA, it outputs ~380 lumen. If I were to run it at 2000mA, it would put out ~680 lumen. But right now it's actually a bigger problem that it's a bit too bright at its lowest setting. I have a PWM dimmer lying around that I could use to get the full output if I need to.

Your 30W is not a single die, is it? Having multiple dies might really limit its efficiency.

Not sure if it is feasible to cram more than a 10w led into the small light holder of the M20. I'm quite keen to experiment with flash, actually. I think that that might be the next project for me, a beam splitter between the field lens and the condenser.

[zzffnn]

Viktor,

1.4x teleconverter simply enlarges image circle 1.4x. So it can be used with either afocal or direct projection. With your Nikon CFN objectives or any "self-corrected" objectives, direct projection can be used (eyepiece is not necessary). For objectives that requires eyepiece compensation, afocal (through compensating eyepiece) works better.

Forum member kit1980 has used 1.4x teleconverter on his microscope trinocular port:
https://www.photomacrography.net/forum/ ... microscope

Yes, my 30w LED is not a single die (it has a single phosphor, and I prefer to further diffuse its light futher for better evenness before it gets to field iris). A single die high power LED that more precisely matches original halogen filament would work better, had I knew this before. My effective illumination size (after I changed location of illumination collector lens) is close (though probably not identical) to original though.

I am using flash with this beam splitter prism between the field lens and the condenser on my Nikon Optiphot frankenscope: https://www.surplusshed.com/pages/item/l2048d.html It works quite well that way and is much more efficient that I expected.

[viktor j nilsson]

Fan,
thanks. I still don't quite understand why an extender would be needed in an afocal setup with a 50mm lens? This lens by itself is already too much for an APS-C sensor.

I've tried afocal before, but your comments inspired me to run some more tests.

Direct projection
First I tried direct projection without an eyepiece. However, I knew this wouldn't work. The Wild HU phototube is non removable, and the primary image is located inside the tube. So that's a no-no.
Actually, I've previously made a Wild dovetail-to-M42 adapter that allows me to remove the binocular head and do direct projection. Although the quality is good, this is not a very convenient solution.

Afocal
Second, I put an 10x CFW eyepiece in the trinocular tube. However, a problem with this is that the CFW eyepiece is too wide for the Asahi Pentax adapter. It can't go all the way down, so it sits approx 7mm higher than it should. So the image can't be made parfocal with the bino head (and the increased tube length might add a little bit of abberrations, but probably not much). Anyway, I placed the Canon 7D with a nice Olympus Zuiko MC AUTO-S 50mm 1.8 objective over the CFW10x (using the focusing helicoid as an adjustable spacer). This Olympus lens is known to be very good in afocal setups, which is why I got it in the first place. Here's the result:


(fungal hyphae with CFN plan 20x)

Third, I put back the CF PL 2.5x photo eyepiece in the trinocular tube set it up to be parfocal with the bino head. Here's the results:


I then took a picture with my mobile phone down the binocular head. Here's a comparison of the FOVs:


The gray FF area is hypothetical. I just scaled it up using http://photoseek.com/2013/compare-digit ... inch-type/ as a reference.

Some thoughts:
- Am I doing something wrong? I thought a FF view with 2.5 eyepiece should almost exactly cover the field of view. And I thought that the APSC FOV should be smaller. More like this:

Could this be due to the 1.25x magnification factor of the Wild phototube? Hmmm...

- The 50mm on its own is clearly too much. A 40mm lens would perhaps be ideal. Maybe even a shorter FL? Image quality certainly looks quite good and the shorter extension is nice. Would still need a new lens and a new microscope adapter, so I don't think I'll do that any time soon.

- I honestly think that the 2.5x field of view is quite tolerable.

[Hobbyst46]

Viktor, the constructuion you described is impressive. Where does the "frankenscope" name derive from?
About the photo setup:

In a macrophotography post a couple of years agor, the following formula was shown for afocal setups:
(camera lens FL) * (photo eyepiece magnification)/250 = crop factor
for example:
if the camera lens is 50mm and the photo eyepiece is 8X,
then 50*8 / 250 = 400/250 = 1.6
which is the crop factor of the APSC sensor. So this should work fine.
If the eyepiece is 10X though, the formula gives 500/250 = 2
so, acceptable for a camera sensor with a CF of 2.

If the optical train includes additional components, they count. For example, with an optovar set to 1.25, a 50mm camera lens and 10X eyepiece, the calculation is
50*10*1.25/250=625/250 = 2.5
so, acceptable for a camera sensor with a CF of 2.5.

For a long time I used a 15-45mm zoom lens in the afocal, set the zoom to 45mm, focused at infinity, used a 8X photo eyepiece, and the FOV coverage by the APSC camera sensor was good. Recently I changed to a Canon EFS prime, 50mm-1.8, installed on the camera via an inexpensive Chinese Canon-clone adapter.

What I am not sure about is how the above depends on the camera lens to sensor distance. For example, using any prime lens on an M-series Canon mirrorless camera is only feasible with an EOS-M to EF (or other, depending on the prime lens) adapter. The adapter increases the lens-sensor distance.

[zzffnn]

Viktor,

Sorry, my numbers were off. I will blame it on my jet lack

You and Hobbyst46 are both right, you need either:

1) APSC Canon sensor - 40mm camera lens coupled closely over CFW10 eyepiece - distance to objective shoulder that makes the CFW10 in trinocular tube parfocal with the CFW10 eyepieces in binocular tubes - objective; or

2) APSC Canon sensor - 1.4x camera teleconveter - distance to objective shoulder that makes projected image circle filling the camera viewfinder reasonably well and be parfocal with visual (bino) eyepieces

But not both.

You need full control of that distance (tube length) though. 7mm difference sounds slightly too much for objective NA of over 0.5.

Any additional optics (magnification) in photo tube does change image circle size and complicates things more.

Hobbyst46's formula looks good and I have seen the same before, but I did not remember it exactly.

[viktor j nilsson]

Viktor, the constructuion you described is impressive. Where does the "frankenscope" name derive from?
About the photo setup:

In a macrophotography post a couple of years agor, the following formula was shown for afocal setups:
(camera lens FL) * (photo eyepiece magnification)/250 = crop factor
for example:
if the camera lens is 50mm and the photo eyepiece is 8X,
then 50*8 / 250 = 400/250 = 1.6
which is the crop factor of the APSC sensor. So this should work fine.
If the eyepiece is 10X though, the formula gives 500/250 = 2
so, acceptable for a camera sensor with a CF of 2.

If the optical train includes additional components, they count. For example, with an optovar set to 1.25, a 50mm camera lens and 10X eyepiece, the calculation is
50*10*1.25/250=625/250 = 2.5
so, acceptable for a camera sensor with a CF of 2.5.

Thanks for this formula. Just a note: the calculation tells us the relay magnification of the system, not the suitable crop factor. It just so happens that the APS-C format needs approximately a 1.6x relay magnification, AND has a 1.6x crop factor.

A micro 4/3 sensor, on the other hand, needs a relay magnification around 1.2, but has a crop factor of 2.0.

What I am not sure about is how the above depends on the camera lens to sensor distance. For example, using any prime lens on an M-series Canon mirrorless camera is only feasible with an EOS-M to EF (or other, depending on the prime lens) adapter. The adapter increases the lens-sensor distance.

I don't know either, but I am assuming that a prerequisite for good image quality is that the adapter allows the lens to be focused at infinity. I'm fairly sure that the adapters you are referring to achieves exactly this, by ensuring that the distance from the shoulder of the EF objective to the sensor is exactly 44mm, i.e. the flange focal distance of the EF system.

If you use an objective can't reach infinity focus (e.g. if you use a short flange focal distance objective on an Canon EF camera), then I'm not sure what would happen.

[Hobbyst46]

Viktor, thanks for the correction, my mistake. So, according to the formula, and the post by Pau in photomacrography:
for example:
if the camera lens is 50mm and the photo eyepiece is 8X, then 50*8 / 250 = 400/250 = 1.6 relay magnification suitable for APSC sensor.
If the eyepiece is 10X though, the formula gives 500/250 = 2 . Between the recommended relay mags for full frame and APSC.
With an optovar at 1.25 and the above, 50*10*1.25/250=625/250 = 2.5 suitable relay mag for full frame.

And yes, I can and do set the focus of the EF prime lens at infinity through the adapter.

[zzffnn]

Viktor,

When you said APSC Canon sensor "needs approximately a 1.6x relay magnification", did you mean this:

26.70 mm diagonal length of sensor / 18mm (assumed) view field of microscope objective = 1.48 (relay magnification)

Cheers,
Fan

[viktor j nilsson]

Viktor,

When you said APSC Canon sensor "needs approximately a 1.6x relay magnification", did you mean this:

26.70 mm diagonal length of sensor / 18mm (assumed) view field of microscope objective = 1.48 (relay magnification)

Cheers,
Fan

That's what I was thinking, yes.

[viktor j nilsson]

This is still a little puzzling.

With afocal 50mm, I should get: 50*10*1.25/250 = 2.5x on sensor

With photo eyepiece 2.5x, I should get 1.25*2.5 = 3.125x on sensor.

Yet clearly I get less magnification with the projection eyepiece.

Some possible sources of error:
1) the CFW eyepiece is raised up ~7mm in the afocal setup. I'm not sure if or why this would increase magnification compared to the theoretical expectation.
2) maybe I accidentally re-focused the 50mm objective so that if was not at infinity? Again, I'm not sure if or why this would change the magnification this much.
3) did I mess up my notes and confuse the pictures? I don't think so but it's always possible.
4) maybe my Olympus-to-Canon adapter messed things up? If I remember correctly, it did focus at infinity just fine in normal use.

That's all I got right now.

Unfortunately I'm away for a week or so, so I can't follow up on this in a while. But I'll look at it when I come back.