| Master_Young |
Forum Moderator
 |
 |
December-08-2004
7 Posts |
|
|
|
For everyone who plays with panoramic photography, locating the right nodal point is not an easy job. I try to find out an easy way to find the nodal point based on my handy experience with different camera and lens combinations. Here's my concept: try to find out a coefficient, which can be figured out with a known nodal point. Then test the coefficient by using it to find the right nodal point for another combination. I don't know if it works for all cameras and lenses, so I'm sharing my ideas and methods with you to find out the truth.
I have a Manfrotto pano head, so this is where my example starts. As shown in the figure below, Manfrotto has two sliding plates. Sliding Plate No. 1 serves moving your camera horizontally, while Sliding Plate No. 2 helps rotating your camera vertically. These plates make sure the nodal point always falls on the central axis of the pano head. We can get two numerical values from these two sliding plates. And we will be talking more of the values in the following contents.
We can get two numerical values from sliding plate No. 1.
L1: distance between Vertical Bracket (this is also where zero marked on the plate) and Nodal Point.
L2: distance between Vertical Bracket and your camera bottom.
As you may notice, L2 is a set number that won't change. So let us see what composes the difference between L1 and L2. It's a common and rational that a camera bottom has some distance to the ring of lens. We take the distance as L4 and take the radius of camera lens as "r", then we can get an equation: L1=L2+L4+r.
In this equation, we can see that L2 is a measurable set number but varies according to different camera sets. L4 is also a measurable set number and changes with cameras. And r for radius is dependent on camera lens.
Let's have a look at the Sliding Plate 2.
We can 2 components of L3:
A length starts from nodal point to mount (we call the length "Lnp" for short)
And L5 as shown below.
So we can get another equation: L3=Lnp+L5. Lnp is dependent on camera lens, while L5 varies with your camera type.
Lnp has coloration with length of camera lens, and can be expressed this way: Lnp=Llens-z.
L1'= L1+x
L3'= L3+y
You have to notice the fact that figures on mark of sliding plate does not necessarily shows the real length of L1 or L3. As a matter of fact, they are bigger than the actual L1 or L3. We can mark them as L1' and L3' to distinguish them from L1 and L3.
L1'= L1+x
L3'= L3+y
FYI: How to figure out L3?
Fix your camera on sliding plate, with screw eye (area 1 below) sitting on the zero mark of sliding plate. Marked area 2 reads the length of L3'.
L1'= L1+x
L3'= L3+y
In the above equations, x and y are fixed measurable values.
With statements above, we make a list of all equations to make it clearer before drawing a conclusion.
L1=L2+L4+r.
LNP=Llens ¨C z
L3=LNP+ L5
L1'= L1+x
L3'= L3+y
(r, x, y, z and L2, L4, L5 and Llens are fixed measurable values)
For a same pano head, value of L1' (L1'= (L2+L4+r) +x) entirely depends on camera lens and camera type. In other words, we can easily get value of L1' with L2, L4, radius of lens and x value.
In case of L3' (L3'= (Llens-z+L5) +y), sheer depends on length of lens, values of z, y and L5.
Since we know these values and how to measure them, things are getting cleared up.
Take this set of equipments into example, Canon 10D + Sigma 8mm + Manfrotto. We have measured L1' and L3'. L1' reads 122 on the sliding plate No. 1, while L3' shows 35 on sliding plate No. 2. Sigma 8mm's size: 63mm*62.5mm, so its radius is 31.25mm. L4 measures 22mm, then according to our equation: L1=31.25+22+L2+x=122. As we know L2 and x are fixed numbers for a known pano head, we can have a shorthand (make it "a1") of L2 and x. So we get a1=122-31.25-22=68.75.
Okay, let's have a look at L3'. Sigma's length is 63mm, L5 is 33mm, so L3??=63-z+33+y=35. Be a copycat of the above shorthand, we take "-z+y" as "a2". So we get a2=-z+y=-61.
a1 and a2 depend greatly on camera lens, as we talked before.
With the above-stated information, we get these:
L1'= r+L4+a1=r+L4+68.75,
L3'=Llens+L5-61.
As it's not hard to measure r, L4 and L5, we can easily find the right spot to fix your camera with the equations above. Of course, a1 and a2 varies according to different camera lens and pano head, then we will have different equations.
"One for all and all for one." I shared my knowledge based on my experience. I'm not sure if it works for every one of you, so let's test the theory together. Please let me know your feedback. Thank you very much!
|
|
|
All Forums
Panoweaver
Topic: an easy way to find the nodal point
