UV photography with a tilt lens

[Kai]

Tilting the depth of field has been a method used in architectural and landscape photography since the early days of photography, in order to be able to image areas of the foreground and background sharply at the same time, especially with large film formats and long focal lengths. Theodor Scheimpflug's name is often associated with this method today.
Early field cameras and cameras based on the principle of the optical bench provided the necessary mechanical requirements. Modern (digital) cameras require special tilt lenses or tilt bellows devices.

An alternative that is used intensively today and does not require such special accessories is the focus stacking.
 

Particularly suitable for UV photography with system cameras is the use of a tilt bellows (e.g. Yashica), which is combined with a UV-transmissive enlargement lens head.

 

Today I would like to present the variant that does not require any handicrafts or adaptations. I simply combined my Canon EOS 6D-FS (UV sensitive up to approx. 365 nm) with the Canon TS-E 90mm 1:2.8 (the old non-L version). For filtering I combined a ZWB2 with 2 QB21.

Even without post-processing, all images show very good sharpness and excellent contrast. I slightly increased the color rendering.

 

 

In addition to grass, Glechoma hederacea (violet flowers in the VIS, colorless dark in UV), Bellis perennis (white leaves and yellow center in the VIS, blue with a dark center in the UV) and Ranunculus acris (completely yellow in the VIS, yellow edge with dark center in UV). Both shots were taken from a tripod at f/8 and 3200 ASA. The exposure time was (in the evening when the sun was low) 0.7s.

 

 

 

 

The other two images show a violet potted plant in the VIS. When shooting at aperture 11, once the plane of focus through the flowers, in the second picture with the plane of focus tilted for maximum blur in the background.

 

 

 

 

[photoni]

Good memories @Kai I started using the large format 10x12cm in 1977

 

I bought for the Sony A7 f.s. a NEX-AI flip ring,

it's easy and cheap and works with all Nikkor lenses ... but ... manual aperture and focus only

(I also have manuals, Nikkor PC-E 85 and 45mm for work with Z7)

[dabateman]

Kai your brilliant.  

I haven't thought of this but could easily use my igoriginal 35mm f3.5 T-mount to nikon mount, old metal EL 80mm f5.6 or old metal EL 105mm f5.6 lenses on my Nikon F-mount to m43rds tilt shift adapter to gain some fun UV tilt images. The smaller image circle of the m43rds sensor, allows quite a bit of tilt or shift.

[photoni]

12 minutes ago, dabateman said:

The smaller image circle of the m43rds sensor, allows quite a bit of tilt or shift.

remember that when you do macro photography, all photographic lenses, even DX, can be tilted with a full format

[dabateman]

3 hours ago, photoni said:

remember that when you do macro photography, all photographic lenses, even DX, can be tilted with a full format

There is a limit to the circle depending how far you have moved the lens away from the sensor. 

I found that out with various lenses mounted to a Fujifilm gfx sensor.  Other stuff gets in the way. 

[colinbm]

These look great Kai.

[Kai]

10 hours ago, colinbm said:

These look great Kai.

Thank you. Colin :)

[Kai]

12 hours ago, dabateman said:

There is a limit to the circle depending how far you have moved the lens away from the sensor. 

I found that out with various lenses mounted to a Fujifilm gfx sensor.  Other stuff gets in the way. 

Is the image circle just as relevant when tilting as it is when shifting?
But even if that were the case, the image circle increases with the magnification. In the case of macros, the image circle shouldn't be a problem, even in the case of full format.
I'm looking forward to the results/experiments :)

[OlDoinyo]

Image circle matters for either tilting or shifting--think of the lens as a spotlight. Tilting means it is no longer aimed at the center of the sensor, so one can encounter the edge more easily.

[Kai]

23 hours ago, OlDoinyo said:

Image circle matters for either tilting or shifting--think of the lens as a spotlight. Tilting means it is no longer aimed at the center of the sensor, so one can encounter the edge more easily.

@OlDoinyo
What makes you think that?
It depends on the tilt mechanism whether the image circle is shifted at the same time when tilting (and whether this can be compensated mechanically) or not.
When tilted, the image circle becomes an ellipse. In theory. In fact, the described commercial TS-E from Canon can (only) be tilted by +/- 8°. This means that the area of the "ellipse" is less than 1% smaller than the area of the full circle (the cosine function gives the connection). The reduction in area of the image circle of less than 1% is certainly irrelevant!

If you have other results from practice, I'm curious...

[OlDoinyo]

It is not primarily about the area. It is about how well the sensor is centered within the image circle. Tilt the lens and (unless you do a compensatory shift at the same time) the edge of the image circle is now closer to the edge of the sensor. Tilt far enough and the edge of the image circle will actually fall within the sensor, resulting in truncation.

[ulf]

13 minutes ago, OlDoinyo said:

It is not primarily about the area. It is about how well the sensor is centered within the image circle. Tilt the lens and (unless you do a compensatory shift at the same time) the edge of the image circle is now closer to the edge of the sensor. Tilt far enough and the edge of the image circle will actually fall within the sensor, resulting in truncation.

That is true, but lenses designed for T/S are designed with much bigger image circle.

The same thing can be obtained by using enlarger lenses that are designed for bigger format film types.

Some of them have a better UV-pass than I would expect the typical modern T/S to have.

[Doug A]

Great idea and results @Kai . The purple and yellow go well together. Hadn't considered trying the EL-NIKKOR 80mm as a shift lens. There is plenty of room, to make it work, on the Pentax K-1 DSLR. Putting this on my list to investigate. 

Thanks for sharing,

Doug A

[Kai]

On 5/8/2022 at 5:19 PM, Doug A said:

Great idea and results @Kai . The purple and yellow go well together. Hadn't considered trying the EL-NIKKOR 80mm as a shift lens. There is plenty of room, to make it work, on the Pentax K-1 DSLR. Putting this on my list to investigate. 

Thanks for sharing,

Doug A

Thank you, Doug - and have fun and good luck with your own experiments. I'm curious!

[Kai]

Here is a supplementary comparison test between the two versions of the TS-E 90mm from Canon. The "new" is called "L Macro".

 

[StephanN]

Great stuff, have a couple ot T/S-lenses, all of them pretty new L-versions, so they most like won't work, but will give it a try. I might even risk getting an old T/S-lens on ebay

[Kai]

Optically, only the old 90s TS-E is usable (I took nice industrial and architectural detail shots with it, even im macro scale). I tested the old 24er TS-E myself and it's much too blurry for me. I haven't tested the 45mm, but it's supposed to be much blurrier than the old 90mm.
The new 24 had led to clearly visible IR hot spots with my modified 6D. In addition, there was no correction module for my RAW development with DxO (the mechanical adjustment data is not registered). In the end, comparison images with the EF 1:4 24-105mm (old version) were more detailed and sharper than the images with the new TS-E 24mm. So I sold the 24 again. My IR favorite is the old EF 1:4 16-35mm. The depth of field at aperture 8 is usually sufficient for me.

[StephanN]

I see, I've got the 17 and the new 24. With neither did I experience hot-spots in IR, but I once took a couple of shots in UV with both of them, I remember now. If you're interested, I can post them here?