Cirkut "Gears" program discussion

[Leigh B]

Setting up the nodal point to measure it is a little annoying with my rig, because moving the nodal point moves the lens, which means I need to keep moving the tripod with the ruler I'm focusing on, to compensate. But I did all that. And in both cases it's off from the theory that says it's midway between the object and the GG.

Hi Duncan,

Forgive the interruption, but I think there's a major misconception here.

I don't understand your reference to THE nodal point, and your description of its location.

Every lens, even a simple single-element lens, has TWO nodal points, front and rear, noted as H and H'.
The front nodal point (H) is important in the design of the lens, but not of much practical use.
The rear nodal point (H') is VERY important.
By definition, the distance from it to the image plane when focused at infinity is the lens focal length.

The two nodal points can be anywhere relative to the physical lens, even completely outside of it.
For example, in a long telephoto lens, the rear nodal point H' will be far IN FRONT OF the front element.
Conversely, with a short FL lens, the rear node will be behind the rear element.

- Leigh

 

[frobozz]

Leigh,

Yes, sorry, Just using "nodal point" as shorthand for "rear nodal point" the one that is important in figuring out how to set up a Cirkut camera properly. At least with the lenses I'm dealing with (large format lenses that are not trying to compact all of the focusing functions into a discrete unit, but rather are depending on external focusing and light-blocking apparatus) the rear nodal point seems to always be behind the physical lens. And proportionately farther behind, the longer the focal length.

...and I was late to realizing that the rear nodal point at infinity focus IS precisely one focal length away from the ground glass... but now I get it.

Thanks,
Duncan

 

[Leigh B]

the rear nodal point seems to always be behind the physical lens. And proportionately farther behind, the longer the focal length.

Hi Duncan,

Of course that's not impossible, but I've never encountered it.

On long FL lenses, the rear node typically moves forward, often far in front of the front element.

- Leigh

 

[frobozz]

Hi Duncan,

Of course that's not impossible, but I've never encountered it.

On long FL lenses, the rear node typically moves forward, often far in front of the front element.

- Leigh

I totally get why that's the case with, say, a long FL lens for a 35mm camera. But you're saying that's the case for large format lenses too?

(If it matters, most of the lenses I'm playing with are 100 year old triple-convertible lenses.)

Duncan

 

[Leigh B]

I totally get why that's the case with, say, a long FL lens for a 35mm camera.
But you're saying that's the case for large format lenses too?

Hi Duncan,

(I was editing the earlier post while you were typing, so I'll put it here instead.)

Yes, telephoto designs can be used for any focal length regardless of the intended film format.

Note for example the H and H' locations on this drawing of a Nikon 1200mm lens for LF cameras:
The front node H is 560mm in front of the front element. The rear node H' is 222 mm behind H.

- Leigh

 

[frobozz]

OK, cool! These old lenses must not be of the telephoto design type then? I'll have to see if I can find both nodes for one of these old guys, just for grins.

Seems like it would be easy enough to get the wrong node when doing tests, when they're that close together compared to the lens focal length.

Duncan

 

[Leigh B]

Hi Duncan,

I can't know whether or not any particular lens is a telephoto, but you can make an educated guess, thus:
Compare the Flange Focal Length (FFL) and the optical focal length (FL).
If the FFL is much shorter than the FL, the lens is likely a telephoto.

In the drawing I posted above, the FFL is 755.7mm, and the FL is 1200mm.

-----

Theoretically it would be easy to confuse the nodes, but practically there's no good way to locate the front node experimentally.

The rear node is easy if you have a reasonably rigid support mechanism.

Focus the image of an infinite subject* on the film plane.
Then measure forward from the film plane by a distance equal to the lens' true (not nominal) focal length.
That's the point corresponding to the rear node, although the true location may be +/- a bit from that point.

- Leigh

*NB: Practically, any subject more than 20 focal lengths in front of the lens is at "infinity", close enough.
Farther than that is desirable, of course. The subject need not be a mile or more distant.

 

[frobozz]

Leigh,

The test above is a bit impossible when I don't know the true focal length (in fact that is what I am trying to determine!)

From the tests I have done on my lenses, where the nodal point is behind the physical lens, that would put the FL shorter than the FFL and thus not a telephoto design I guess.

Isn't it true that the nodal slide test (wiggle the lens about some point, the GG image does not wiggle) would also show success at the front nodal point? That is what I was thinking of when saying it might be easy to confuse them.

Thanks,
Duncan

 

[c.d.ewen]

Dear Dunc:

I've been holding out on you. Not intentionally, of course. I just didn't remember the way I used to determine FL's, until I came across my sextant the other day. Try this:

Find a scene where you can photograph two distant, i.e., as close to infinity as possible, objects that are roughly equidistant from you, and point your camera at a center point between them and take a picture. I used to shoot a mile across the Hudson River at the telegraph poles underneath the NJ Palisades. Take your sextant (you've got one, right?) and measure the angle between the two points. Scan the resulting negative and measure the distance between the two points. PhotoShop has an easy tool for this. Divide the distance and angle by two. Now you've got a right triangle and know the base and one angle. Do the math; the length of the perpendicular is the FL.

Back in the late 1990's, I used to shoot immersive 360 degree digital panoramas - before I discovered the fun of Cirkuts. I usually determined nodal points just by swinging the camera side to side while observing a near and far point, and moving an adjustable mounting plate until the parallax disappeared. You might try the laser pointer method on this page, which would allow you to mark the nodal point on the lens. I miss my old CP990.

Charley

 

[Euphus]

Leigh,

The test above is a bit impossible when I don't know the true focal length (in fact that is what I am trying to determine!)

From the tests I have done on my lenses, where the nodal point is behind the physical lens, that would put the FL shorter than the FFL and thus not a telephoto design I guess.

Isn't it true that the nodal slide test (wiggle the lens about some point, the GG image does not wiggle) would also show success at the front nodal point? That is what I was thinking of when saying it might be easy to confuse them.

Thanks,
Duncan

I've been reading this old thread with many of the questions you had and discussed with Mark on determining the correct FL for lenses without using a lot of film for testing. Which method did you settle on and why?
Thanks,
Euphus