Can't Fill the frame for 35mm adapter on a 2/3" Camera

[Forrest Schultz]

My friend Adam and myself are having a lot of trouble trying to get our setup to work. He is using one of the Sony Cinealta 2/3" HD cameras with a Canon HD zooom lens attached. I built a 35mm adapter for him using Daniels GG method. The 35mm adapter is fully functional and setup correcty as far as i know. But as i figureed, the size of the Canon lens for the Sony camera is keeping us from being able to completey zoom in on the 35mm frame and stay focused.

The setup is like this:

Sony 2/3" camera> Canon Zoom Lens> 82-72 step down ring> 72mm 10+ macro (we have 2 of these)> 72-55mm step down> 55mm Macro tube assembly spacer also in which the Canon EE-A screen is held> more space for FF then M42 mount.

I have tried using 2 of the 72mm macros butted up but it still doesnt allow us to fill the frame and also causes bad barrel distortion. We are a little exausted of the problem the only solution i could think of myself would be for me to make our own relay lens from a 2/3" CCTV lens. i already made a crude c-mount to Bayonet Mount adapter so i can attache the lens to the Sony. i just wanted to know if this is the route i would have to take?

[Bob Hart]

My personal preference would be to check the available zoom range of the Canon lens. If it is below about 80mm you might need for a test, to get hold of a compatable ENG zoom lens which mostly can go to 125mm without the doubler being selected in.

You may need 85mm or more to get a successful relay to 2/3" without the edges of the groundglass itself vignetting the image.

I can get relay to a 2/3" JVC camera head on Quyen Le's original Letus35 flip for Canon XL using the Minolta SLR relay unaltered except for matching the mount. This lens is a 50mm and has an about (guessing) 5+ to 7+ unmounted dioptre attached with compliant adhesive on front of it.

Quyen's relay system imaged a relatively small area of the total available from his still-image camera sized groundglass. So there was ample extra area for the wider field of view of the 2/3" image area with the existing 50mm relay lens.


The rough and ready initial formula to try might be :-


Your existing 2/3" camera lens zoom (if it is a zoom) set to between 50mm and 80mm.

2/3" camera lens focus set to about 1.5metres. If this causes a distortion, then go closer to infinity focus.

Achromatic dioptre of no more than 7+. (10+ and 2 x 10+ is a bit severe and given the wider field-of-view of the 2/3" CCD, I would expect edge softness and distortion to be a real problem.)

The achromatic dioptre needs to be as physically close to the front element of the Canon lens as you can get it, meaning the rear element of the achromatic dioptre being no furthur forward than about 8mm from the front element of the Canon lens.

Unless they fit with an indentical flush front face, stacking stepdown rings may force the dioptre to be too far forward of the Canon lens front element.

Front of your 2/3" lens setback from the groundglass initially may be about 5 inches for a non-flip design with the 7+.

The only 7+ dioptres available that I know of are the 58mm filter mount versions which on the 2/3" camera may vignette, but I guess at the zoom setting you will need to use, might just be wide enough.

Do not take this suggestion too seriously as I am no optical engineer and may be sending you on a goosechase.

[Ben Winter]

What Canon HD lens is there that is 35mm? And m42 mount at that?

[Forrest Schultz]

Originally Posted by Ben Winter

What Canon HD lens is there that is 35mm? And m42 mount at that?

Ben you should read slower, but the Canon HD Zoom lens is the large broadcast lens for the Sony cinealta (stock lens if you will), with a B4 mount i believe. it is not the M42 mount 35mm lens for the adapter.

[Ben Winter]

Ah, so it is. Bugger me.

Can you provide mm measurements from the front of the lens to the GG? And what macros are you using? Are they achromats?

Take a ruler with mm measurements, and put it in front of the camera with its stock lens. Get the ruler as close as you can while in focus, and record how many mm wide it is showing in the frame. With this information, you can use this spreadsheet to determine what power diopter you need to add to frame something at width R at distance D.

Good luck.

[Forrest Schultz]

Thank you very much Ben. I took a quick look at that spreadsheet and i imagine it is extremley helpful. ill have to come back to it after i get those measurements. we tried all types of distances between the 35mm gg and the camcorder lens.

The macros we have are the cheap Asain brand 72mm 10+ that you can get from ebay. we didn't spend big because i felt we would have this problem during testing. we tried close to the lens. very far. and all with no filling the frame. the macros didn't seem to be contributing at all either.

[Bob Hart]

Forrest.


Maybe step back from the actual adaptor for a moment and examine the lenses you have for the 2/3" camera.

Draw on a piece of stiff paper a rectangular target with some gridlines on it about the size of the 24mm wide x 18mm high image area or larger if that is your choice.

Use this to test your combinations of videocamera lens, dioptres and setback distances from the groundglass, a lot easier than trying to see images on the groundglass and doing a lot of unnecessary construction and dismantlement.

I just recently had a play with an EX1, which has a 1/2" imager system and a Mini35 for JVC. This I had working to a Z1 with a 4+ dioptre. My own Agus35 works about the same area off the groundglass as the Mini35 by this lens arrangement.

To barely frame the round groundglass of the Mini35, the EX1 camcorder lens had to be all the way zoomed in with a 4+ dioptre on front, which is 81.2mm.

With a 7+ for the same framing, the zoom came back to about 58mm but sharp focus could not be achieved because the dioptre was prevented from moving closer by the flip structure and bayonet adaptor on rear of the Mini35.

( When the camcorder iris was closed the image became almost acceptably sharp so the 4+ was pretty close to position. )

By the same arrangement, the Sony Z1 with its 1/3" imager system needs 54mm which is the lens zoomed in all the way with the 4+ dioptre.

With a 7+ for the same framing and closer to the groundglass, the zoom can come back to about 32mm.

In my rough math ( I failed in school ) :-

1/3" = 8mm approx.
1/2" = 12.5mm.
2/3" = 16mm approx.

So

54/8 (1/3") = about 7 ( counting on fingers and toes and visible nose whiskers. )

82/12 (1/2") = about 7 ( ditto. )

Therefore ??/16 (2?3") if = about 7, becomes about 112mm zoom position for your 2/3" camcorder zoom lens to be at with a 4+ dioptre on front.

The 7+ may allow you to move closer and back off the zoom a little to perhaps 85mm or so.

Setback from the groundglass to front of lens for the 4+ is about 8 inches or 200mm.

Setback from the groundglass to front of lens for the 7+ is about 5 inches or 125mm.

The 10+ will be closer but distortions start to creep in.


Your path may be someting like this :-

Videocamera lens set at about 112mm zoom-in >> 10+ dioptre mounted as close to front of videocamera lens as you can get it >> between 2" and about 3" distance from front of dioptre to groundglass.


There it is, a whole lot of vague guesswork but something that may get you in the ballpark of something being visible.


For curiosity sake, I ask, what is the available zoom range on your 2/3" videocamera lens?

[Bob Hart]

Deleted - Double post somehow happened

[Forrest Schultz]

Bob, i have been fidling with 3 different types of c-mounts in attempt to build a relay system. we are comign very close.

As you know c-mounts focal flange distance is very short compared to what the sony cinealta is set at. the sensors are further away and slightly out of range from the c-mount ffd. pushing the c-mount lens right up against the uv filter glass on the cinealta (allowing me to get as close as i can to the sensors without removeing that glass) i can frame and object approximetly 10mm in width and focus perfectly. if i move the object further back, it will obviously go out of the focus range. (PS: i built my own c-mount to b4 mount and am planning on buildign a cleaner version) so oviously 10mm is very close up and macro, and thats all dandy but we of course need at least 36mm of width to fill the frame. thats about 3 times more than what i can get so far.

what helped out was when i took that c-mount lens, detached the back focus portion of it and slid its backfocus optics further away from its front element. that allowed a good improvment in maximum focus distance. just not enough. i experiemnted with removing various elements of the lens or flipping them around, hoping to lengthen the focal flange distance, but that did not work.

is there an easier way to make the c-mount have a longer focal flange distance? I love you Bob

[Bob Hart]

That's an interesting choice of wordstuff which makes me a bit nervous, however back to the unscientific science :-

The best I can do on the "2/3" JVC KY-F50 is a Cosmicar 25mm C-mount lens with focus set at about 35.M, yielding a 26mm wide frame within TV safe area at about 122mm forward of c-mount flange face, c-mount flange forward of actual flange face by about 7.5mm.

The 122mm may be a bit too short to enable a sharp focus on the Letus groundglass without front of lens being placed forward of the rear face of the flip enclosure.

There may be a better result with a 28mm to 30mm c-mount prime lens if such can be found.

The front edge of the Cosmicar 25mm c-mount lens barrel is about 31mm forward of the front of its forward optical element.

The rear edge of this lens barrel (nose of c-mount thread) is about 2mm rearward of the rear of the rearmost optical element which is just a few hairs rearward of the flange face so there is no gain from re-barrelling it except to get that last 1.5mm to 2mm of closer position within the B4 mount.

If you chose to rebarrel this lens to physically shorten it, you would lose the ability to set aperture as the aperture ring is forward on the lens barrel. The front edge of the aperture ring to front edge of lens barrel distance is 11mm.

Old Schneider-Kreuznach lenses for Bolex have too small an exit pupil for this application.

Changing the relationship between the individual elements of your c-mount lens may alter its flange to focal plane distance but may introduce abberations such as barrel or pincushion distortion or soft edges. You may find that going the opposite of what you have been doing and placing the front moving group closer to the rear fixed group may enable you to then move the c-mount lens body furthur forward of flange position enough that you can get it to fit.

To get the lens far enough forward in the Cine-alta camera body to clear the internals, there might need to be an added optical element between the groundglass and the c-mount lens in order to have it see the groundglass as being furthur forward and the c-mount lens then having to be positioned furthur forward of flange face in order to restore focus sharply on the groundglass. This is the very opposite of what the achromatic dioptres have been used for, that of enabling sharp focus on close objects.

The P+S Technik Pro35 has what is called a "field lens" between the groundglass and the relay lens. I don't know the theory of field lenses. I think it might confer an extended depth-of-field which might "stretch" the focal range of the relay lens. On the surface the Pro35 relay appears superficially the same as the Mini35 but it could be a far different story under the relay lens barrel.



Sorry I cannot be more helpful.