Looking for references on creating a rangefinder cam

[miles_s.]

I’m searching for any resources that will aid me in creating a properly formed rangefinder cam on a custom lens. Any

books, websites, previous threads are greatly appreciated. I doubt I have specified the task enough for comments.

 

Of course it is easy if the lens's focal length is normal --- that is 51.6mm for Leica, 52.3mm for Contax and the like.

However, how does the task change if the lens is shorter than normal versus longer?

 

If I were to modify a cam from an Industar or the like can one slope the cam through grinding? Does focal length

effect this?

 

Besides custom adjustments how does one account for the difference between the fiducial points of the lens's focus

(the posterior focal point) and that of the rangefinder (the baseline of the RF)? That is how does one account for the

difference between the angle in RF being inversely proportional to range while the lens extension is inversely

proportional to range minus twice the focal length plus the separation of the principal planes?

 

R. Kingslake Optics in Photography seems like a good start. However, I was hoping for some more concrete

direction with reference to LTM or M lenses.

[Brian1664876441]

I've not run across a how-to-do-it for making custom cams on "hacked" lenses. Sounds neat, though.

 

How far off from the "norm" are you looking for? Are you looking to modify an industar cam to accommodate a 45mm lens, or 35mm? I have a custom cut COntax to M-Mount adapter that cut the cam to correct for the 52.3mm vs 51.6mm focal length difference. It is a noticeable cut, about 0.5mm deep as you go through the 270degree rotation.

 

Sounds like an interesting project.

[john_shriver]

With some triginometry, and careful measurements of all the "arms" in the linkages in the rangefinder (easier on a screwmount Leica), you should be able to determine the formula.

 

That said, the "nature" of the formula is one where is it a "linear" function of the distance of the lens from the film. Look at (say) a Leica Elmar 35/3.5 or Summaron 35mm lens. There are two helicals. One is the right pitch to focus the lens. The other one, which has a higher pitch than the focusing helical, runs the rangefinder cam, and is rotationally locked to the focusing cam.

 

So what you need to do is compute the proper "slope" (compared to the focusing rate) for a given focal length. The intercept you already know, at infinity the cam protrudes 7.5mm on a LTM lens, or 6.5mm on a M-mount lens.

 

If you measure a "known accurate" lens at a known focusing distance (say 1 meter), you know how much your cam should protrude when you custom lens is focused at 1 meter. So you collimate your lens to infinity. Mark that focusing cam position for 7.5mm (presuming LTM). Now, you need to accurately focus your lens to 1 meter. Now, mark that point on the focusing cam to the "correct" protrusion for 1 meter. Now, mark a linear slope between those two points, and that's how you cut the cam.

 

Now, that said, the camming approach is really only appropriate for an M-mount lens. That's because the rotational indexing of the lens into the mount is consistent. That isn't the case with LTM mount lenses, the thread start angle is not precise, so the lens screws onto each LTM body with slightly different orientation. So for an LTM mount lens, you need to make a custom helical just for the focusing cam, with the right pitch for the focal length of the lens, and related to the pitch of the true focusing cam.

 

(Note that the formula is NOT linear to the subject distance to which the lens is focused. That's because the extension of a lens from infinity focus is proportional to the inverse of the distance you are focused to.)

[miles_s.]

How far off from the "norm" are you looking for? About +25mm I want a ~75mm lens. Interesting to hear that 0.5mm over ~270 converts a Contax cam into a Leica cam.

[miles_s.]

Thanks for the input. I can do at lot with this information.

 

 

“With some triginometry, and careful measurements of all the "arms" in the linkages in the rangefinder (easier on a screwmount Leica), you should be able to determine the formula.” The formulas are easy to derive and by having empirical constants set by "exterior measurements" -- distances and cam lengths.

 

“If you measure a "known accurate" lens at a known focusing distance (say 1 meter), you know how much your cam should protrude when you custom lens is focused at 1 meter. So you collimate your lens to infinity. Mark that focusing cam position for 7.5mm (presuming LTM). Now, you need to accurately focus your lens to 1 meter. Now, mark that point on the focusing cam to the "correct" protrusion for 1 meter. Now, mark a linear slope between those two points, and that's how you cut the cam.” Doesn’t this presume the lens “racks out” less than a normal lens? Of course I could reverse the procedure but then I need to know what the cam extension a1 1 metre is. Considering the range of the rangefinder differs from the object distance by about 10% this is tricky.

 

“the camming approach is really only appropriate for an M-mount lens. … with LTM mount lenses, the thread start angle is not precise, so the lens screws onto each LTM body with slightly different orientation.” Agreed, I was going to dedicate an adapter -- LTM to M mount.

[Brian1664876441]

A 75mm lens is typically implemented with a double-helical, an inner cam translates the motion of actual focus into motion of the cam.

 

Found this explanation of backfocus.

 

http://hyperphysics.phy-astr.gsu.edu/Hbase/geoopt/image.html#c1

 

Examine the difference in the image formation between a 7.5cm lens and a 5cm lens. Look at how it changes from 1m to 300m. It is the difference in backfocus between a 5cm lens and 7.5cm lens that you will have to design the cam.

 

As an alternative- lots of 9cm optics around that can be mounted in a beat-up 9cm Elmar focus mount, and lots of 8.5cm optics that could go in a beat-up J-9 mount.

[Brian1664876441]

Just some "quick Clicking", the 7.5cm optics move ~0.6cm to across the range of 1m to "infinity", and the 5cm lens moves ~0.263cm from 1m to "infinity". You have to reduce the 0.6cm to 0.263cm as you focus across the range.

[winfried_buechsenschuetz1]

It is not possible to create a sloped coupling cam for lenses with a screw mount such as LTM. The reason is simply that the relative positions of the ramp and the cam follower are not precisely defined - with some higher torque you will be able to turn the lens a bit more.

 

It's a bit different with bayonet mounts, I think some Leica M lenses use a sloped cam.

 

As mentioned by others, usually you need a "transfer helicoid". You need a supplementary helicoid which is coupled to the focussing helicoid in turning direction but which is free in longidutinal displacement. This helicoid has to move the fixed distance between the cam position at close-up (say, 1 m / 3 ft) and the cam position at inf. when it is turned the same angular distance as the focussing cam. Creating such a helicoid might not be easy.

 

You can also use an "internal ramp" which is not coupled directly to the cam follower but uses an in-between tab (with a wide flat end towards the camera body) avoiding problems with proper angular positioning of the cam relative to the cam follower.

[winfried_buechsenschuetz1]

I just remember that both design principles I mentioned were used by some manufacturers.

 

The russian Jupiter-9 f/2 85mm lens was originally designed for the Contax/Kiev cameras which do not use an axially moved rangefinder cam. So in the LTM version a secondary helicoid was added which has just the displacement needed to move the cam correctly. It is coupled to the main helicoid by a tab of the latter and a groove in the secondary helicoid.

 

I have two LTM lenses by A.Schacht of Ulm which use the second priniciple. They have a sloped cam on the main helicoid which drives a tab with an even edge. The slope is designed such that the tab has correct displacement.

[kelly_flanigan1]

The cam movement just varies with the focus distance; its the same movement whether its a LTM or M lens; of any focal length. The concept is such a noraml lens of about 52mm has a unity movement; ie the lens block and cam are ONE for this focal length; its been this way since the 1930's. For a longer or shorter focal length than 52mm the cam ie roller still moves about roughly 1/10 inch or 2.5mm when one focuses from infinity to 1 meter; except the lens block moves less or more; since its focal lenght is NOT 52mm. With a non 52mm the roller cam moves more or less than the lens block; depending on its focal length. The "gearing" ie lever ratio can be thru a internal helix with a fixed stub cam surface; or a sloped cam on a revolving cam ring; or a combo of both. One can make a custom cam if one has an open mind; and one can lap or machine any surface such that a LTM or M camera can focus with any lens. The cam on a Leica moves the same amount when focusing from infinity to 1 meter; its etched in stone.

[kelly_flanigan1]

From a Optical Mechanical/Mechanisms engineering model one might imagine you are a "little man"; ie an observer watching the roller cam's movement on a LTM or M body.<BR><BR> One would see and measure the roller cams postion to be the SAME if a lens; ANY LENS was set to 3 meters; no matter what proper lens on the planet was placed on the camera body. <BR><BR>The cam's position HAS to be the same; since its tied to the converging rangefinder assembly thats got an aligned image at 3 meters.<BR><BR> IF one has a screwed up lens; a hacked lens and the cam is off; one gets missfocused image; since one is "closing the loop" around the converged images. This scheme goes back to the 1930's; its over 7 decades old now. Its the basis of "creating a rangefinder cam". <BR><BR>

[kelly_flanigan1]

The same cam scheme and cam movement is used for LTM and M bodies; thats why a LTM lens can be used on a M body. Both a LTM and M body if aligned have the same movements of the roller cam when one focuses from infinity to 1 meter; or any other distance.<BR><BR> The real LTM versus M differnce us that its harder to control where the start of screw mount lens starts; thus scheme that uses a revolving sloped cam ring on a LTM lens is more dicey. Here I have made a modifed custom Jupiter-9 in LTM where the circular ring is lapped to correct its focus woes; BUT I use the same LTM to M ring when its on my M3 such that the "thread start angle problem is quashed".

[kelly_flanigan1]

An old body or test lens flange can be used with a dial indicator as a test settup to measure the roller cam position's contact point versus actual lens scale focus position. Here for LTM I use an old Zorki body with a Starrett dial indicator that reads to 50 millionths; ie one half 1/10000 inch. For a revolving cam I pull the contact point out to not ruin the point of cam when focusing.