Anatomy of my FS PCX lens build

[ulf]

Here is the anatomy of my lens build around one of my Fused Silica PCX-lenses.

I bought some quality surplus lenses in the beginning of last year and have made an assembly that I think will work reasonably well.

There will be some time until I can make proper practical tests with the lens , so please be patient about that.

 

First all components I have used:

 

All parts are just screwed together.

First the Iris-module and lens-mount adapter on each side of the helicoid.

 

As the lens has a diameter of 40mm it will be kept flat and well centered into the M42-structure.

 

If I should buy components from scratch I would have bought a smaller lens and mounted it with step rings as the full diameter is never used optically and a smaller lens is cheaper.

 

 

I found out that the construction is very sensitive to flare and the contrast is improved considerably by an optimal lens shade blocking everything except the needed area for the image.

The solution was to 3d-print a suitable baffle and place that in the end of an extension tube, serving as lens hood. Flocking both extension tube and helicoid where possible also improved the contrast

[Stefano]

Very nice, you built a complete lens from scratch. This is your lens, only you have it.

 

Can't wait to see how it performs. My guess is that it will have a significant amount of chromatic aberration, for the reasons you pointed out here, but hopefully not too much. Is this an aspheric lens? If so it should be already corrected for most if not all spherical aberrations.

[dabateman]

Yes 25mm, 1 inch elements are cheaper and They just cover the 135 format sensor. I should run a test on my Df.

They don't cover the instax film though, even in macro.

 

[Andrea B.]

Wow, Ulf !! This is great to see. Thank you for this. I'm really looking forward to seeing some tests of your FS PCX lens.

[colinbm]

Looks really good Ulf, I look forwards to the images.

[ulf]

Yes 25mm, 1 inch elements are cheaper and They just cover the 135 format sensor. I should run a test on my Df.

They don't cover the instax film though, even in macro.

How do you think here David?

I do not understand.

If it would be so, then a pinhole would only cover a micro-dot format supposedly used by spies.

 

A 25mm- or 1"-lens element would work well on full frame sensors as long as the aperture is placed close by.

If you accept a smaller maximum 1/f the even a smaller lens would do.

See the aperture diameters here: https://www.ultravio...dpost__p__45608

[dabateman]

Just tested my 100mm plcx and 100mm blcx 25mm elements. They fully cover the sensor on my Nikon Df camera. I tested them wide open so roughly F4. I didn't test them at infinity though, only at 1 to 2m focus distance from a toy.

 

Also my 110mm lens (500mm bicx + 200mm plcx + 300mm bicx elements in that order) was better for sharpness and aberrations. All elements are fused silica.

My comparison Nikon 105mm f2.8 macro was best, but it can't see UVC. Tested under white LED light.

 

Edit just reread what you wrote. What I have found with my instax film camera, which has largest test surface is that focus distance matters. At 1m my Rokinon 85mm f1.4 will fully cover the medium format film. But not at infinity at the same aperture. This should make sense as the closer the elements are to the film/sensor the less coverage. If you move the elements away for close up macro photography than you will cover more of the sensor.

When its light outside I will see if the 25mm diameter 100mm focal length element covers the sensor. I think it still will as the 100mm focal length needs the element to be far away from the sensor. A 35mm element would have a harder time, as it needs to be very close and wouldn't work with my F-mount DF.

[ulf]

What you are writing above about coverage and focus distance is true for complex lenses, as they have an image projection cone from the internal design.

 

It is not true for single lens designs like these as long as there are no mechanical obstructions after the iris, like from a badly designed narrow lens mount adapter, with a narrow opening.

 

The refraction is happening in the surface in the lens-curvature.

The radius of the spherical curvature is the same for both a Ø40mm and a Ø10mm lens, of the same focal length.

The bigger lens's surface stretches further and becomes steeper towards the edge, making the lens thicker.

 

As the iris is placed directly after the lens, it just changes how big part of the light from the domed surface that is used to project an image onto the sensor.

It is fully comparable with a pinhole.

 

Naturally you can define the coverage by how acceptable the image quality will be, but that will be the same for different lens diameters with the same shape quality and FL.

The only difference will be how wide you can open up the iris as that is limited by the lens diameter.

For reasonable image quality these builds has to be stopped down to f/8 or more, anyhow and then even a 1/2" lens element would be enough.

 

It seams like to short FL gives problems with mage quality and for full frame sensor cameras it is better to stay at or above 50mm or more.

The sweet spot might be in the 70-100mm range

[JMC]

On my build I've mounted the iris in front of the lens, not behind it, which I understand was not unusual in older designs of lenses.

 

With my 79mm aspheric element (only 25mm diameter) it seems to have no issues covering a full frame sensor. My 40mm PCX (again 25mm diameter) lens will cover a full frame sensor, but with vignetting and huge distortion at the edges. Ulf like you I think 70mm and above focal length singlets are better bet.

[Stefano]

Aspheric lenses look very promising for DIY builds. In theory, they should be aberration-free, as I understand (but still with chromatic aberration). It would be interesting if this can allow building wide angle lenses with reasonable image quality.

[Stefano]

Also, sorry that I am asking again, but is your lens an aspheric lens or a "normal" spheric lens?

[ulf]

My lens is a normal plano-convex spheric lens.

Perfect aspheric lenses are very difficult to make.

That is why any aspheric lens is rather expensive.

[Stefano]

OK, thanks. I know they are expensive, but if the quality difference is a lot then maybe it is worth it to spend the extra money. I would still suggest to use a cheaper lens in the beginning to experiment with.

[dabateman]

Ok just tested my Bicx 50mm focal length 25mm diameter lens on the Nons instax film camera and at f2 wide open it does cover the film close and at infinity. The shutter maybe in the way at extreme edges, but I am surprised it fully covers.

The 50mm diameter 39mm focal length lens also fully covers, but that wasn't a surprise. It can only close focus as the Nons is just like a Canon camera with 44mm flange back distance.

 

The film is 46mm x62mm so a worst case sensor size.

[colinbm]

Your 50mm diameter Fused Silica PCX-lenses, have you got any photos yet ?

Is this a LED light lens ?

[ulf]

No, I have no photos yet.

 

No, the lens above is much better quality than a LED lens.

It is a Ø40mm, focal length 50mm quality lens that came from a scientific optical parts-company that shut down.

I think it was a smaller independent company in the same type like ThorLabs, but much smaller and that the owner retired.

I bought several other FS-lenses from them for my optical experiments.