Canon TV 50mm f/0.95 optical construction

[OlivierAOP]

Hi. First post here, in search of advice.

I found a Canon 50mm f/0.95 in my lab, it was used 30+ years ago in medical imaging. It is missing the rearmost optical element. That's based on appearance and from the back-focus distance (too long), FOV (too narrow) and image quality (quite poor at large apertures). It is also missing a mount but I made one for my Sony A7.

My hopes of finding a correct replacement lens element are pretty low but maybe Mr. Kitchingman or others have detailed info on that element?

It looks bi-convex from the diagram, and based on rough measurements and calculations it should have a focal length of approx. 50mm.

 

[Mackinaw]

With no rear element, no wonder the pictures are blurry.

Here's the optical diagram. No idea what type of glass Canon used, or the curvature of the rear element. Good luck.

Jim B.

 

[OlivierAOP]

Thanks. I did lift the curvatures and thickness from the diagram and scaled the results to what I think is the correct OD of the element. I got sensible results (r=75 & 85mm, t=5mm) which gives focal lengths of ~55-80mm depending on glass type. I doubt the diagram is to scale given that there should be a space between the concave doublets for the aperture.

 

[rfaspen]

You're so close to having a neato lens. I guess I wonder why the rear element is missing. Was it intentionally removed? Or some mishap? I know how labs can get filled with the remnants of so many past activities -- kind of fun to peek in my drawers sometimes. So...perhaps that element is sitting in a drawer? Worth a look?

 

[OlivierAOP]

It's funny. When I was researching the lens after it was found (I had no idea it existed nor that it was missing an element), I was thinking this lens might not be for me. Now that it's out of reach I have a bad case of the blue balls LOL.

Anyway. My colleague who was around back then has no recollection of that lens. Too bad. My best guest a a mishap... the lens it quite bad without the element.

I did find an element in the drawer that sort of fitted. Testing (freelensing) did improve the IQ and restored the FOV. That's why (+ some tests/calculations) I think a 50mm element will help but it'll be bad at f0.95 I think.

Short of finding spare parts or hiring a lab that'll have to do!

 

[Rayt]

The bokeh should improve without the rear element.

 

[retinax]

If you're so inclined, wouldn't it be possible to buy or borrow another specimen, measure the radii of that element, and together with the focal length that your incomplete lens has, calculate what glass properties are needed?

 

[OlivierAOP]

Well right now the images are a hazy mess with no contrast or detail at f0.95. More 'knocked-out' than 'dreamy'. Once I figure stuff out I'll post pictures to illustrate.

Retinax ideally I'd find a broken one for the parts. Buying or borrowing is an option but measuring curvature accurately requires specialised equipment.

I guess this thread shows my wishful thinking. Even if I knew the exact specs of the element I don't know that it could be sourced unless it's coincidentally off-the-shelf like 50mm FL symmetrical bi-convex in BK-7.

 

[retinax]

Well right now the images are a hazy mess with no contrast or detail at f0.95. More 'knocked-out' than 'dreamy'. Once I figure stuff out I'll post pictures to illustrate.

Retinax ideally I'd find a broken one for the parts. Buying or borrowing is an option but measuring curvature accurately requires specialised equipment.

I guess this thread shows my wishful thinking. Even if I knew the exact specs of the element I don't know that it could be sourced unless it's coincidentally off-the-shelf like 50mm FL symmetrical bi-convex in BK-7.

Yes, it seems more likely that you could find one with a broken front. Fortunately normally the fronts are destroyed in a fall or so.

 

[raid]

I would send the lens to DAG.

 

[OlivierAOP]

I corresponded with DAG, Japan Camera Hunter and Kamera Service (Netherlands). Asked an optical physicist as well. No luck. I'm a newcomer in my area regarding photography and I haven't found local people. Happy to receive further suggestions!

 

[KEVIN-XU 愛 forever]

I think there is no way to build the rear element in the right spec. If you don't want to waste this lens part, just sell it for parts. I guess some technician may need the rest of the element to fix the other Canon 0.95. Or you bought another lens part with the rear element and resemble together.

 

[OlivierAOP]

Small update. Via Marco Cavina's site I found the patent for the lens. It's by Oude Delft (US 3357776) which Canon may have licensed?

I modeled the specs on WinLens3D and without the last element the focal length matches that of my incomplete Canon (f=115-6mm, measured via FOV crop factor vs. a 90mm lens).

I also modeled the Canon f1.2 lens (US patent 283361102) and it doesn't match. So I can't recycle one of those.

So the rear element should be f=40mm. I can't get one in the same glass (high refraction & low dispersion) but I've ordered some in BK7. I'll update when I've tested them.

 

[retinax]

Exciting! Hope this works out.

 

[NY_Dan]

The problem is elementary my dear Olivier.

 

[OlivierAOP]

The problem is elementary my dear Olivier.

Because I need the right element?

Anyway. I tested some BK7 40mm and 43mm assymetric pieces. Correct FOV is right in between (i.e. close enough) but image is kinda poop. Guess I need high refraction glass as per patent.

A piece from the Canon f/1.2 would likely be decent as well.

 

[View Range]

Olivier, our pal Peter Kitchingman lists Mukai Jirou as the designer of this lens. Why do you think Canon might have used an Old Delft design?

 

[OlivierAOP]

It's only a theory. So far it *seems* to match the lens and while many Canon patents are out there (from H. Ito and M. Jirou), the Oude Delft one is the only one at 0.95 that I could find. It's my only lead basically.

 

[View Range]

But the patent was applied for in 1963, after being prepared in Nov 1962. The Canon 0.95 hit the market in June 1961. Again, I'm using Peter K's information.

 

[Mackinaw]

..... The Canon 0.95 hit the market in June 1961....

Which means it was designed in 1960. Dechert, in his Canon rangefinder book, says that rumors of the Canon 7 camera and 50/0.95 lens began, "more than a year before production began."

Jim B.