Changes to ISO 10110-6

Status

· I have made some progress on resolving

issues and writing proposed text.

· Reinhold Litscher (sp?) has agreed to

accept assistance on rewrite.

· His group has already begun work on it.
· It will be interesting to see if we pick the

same issues, and similar solutions

My views so far:

PowerPoint doesn't do strikethrough

So...

Black is current text in document

Strikethrough is Blue

My additions are in Red underline

My comments are in Green

3.1

optical system
optical element, subassembly, or assembly

This is vague and allows for circular

definitions later, as in 3.11.
Needs more work.

3.2

optical axis (of an optical system)
theoretical axis about which the

ideal

optical system is nominally rotationally

symmetric

Note: Deflecting elements and systems,

such as plane mirrors, prisms, etc. are

exceptions.

Exceptions to what exactly? Are they to be

defined elsewhere?

3.4

datum feature
Real feature of a part (such as

the outer

edge of the lens cylinder or

a spherical

surface) that is used to establish the

location of a datum.

The datum is the true

geometric counterpart of the datum

feature.

A datum feature should be

accessible and of sufficient size to permit

its use.

Current 3.5

cylindrical datum feature

datum of a cylindrical surface (such as the

outer diameter of a lens) is the axis of the

cylindrical surface datum feature, which is

the axis of the smallest circumscribed

cylinder that contacts the diameter.

rim

edge of a lens or inner cylinder of a lens

barrel used to establish the cylindrical

datum.

3.x

cylindrical datum
axis of the smallest circumscribed cylinder

that touches an exterior cylindrical datum
feature; alternately, axis of the largest
inscribed cylinder that touches an interior
cylindrical datum feature.

3.5

Sorry, not done yet

· The definition is now clear, but it won't be

helpful in the real world because we have
imperfect cylinders and interacting datums

Two "smallest circumscribed
diameters"

More 3.5:

Geometric impossibility

No intersection, so
no datum axis exists

Yet in practice, lens is fine

4/1'

Reference Figure 5

Why? Datum system is defined by
intersection of center
of curvature of A with axis of B.

3.5 a proposal

Define cylinder datum to be the center point of a circular

target line at the axial midpoint of barrel edge, unless

another position is specified (as in Figure 4.)

Revise Figure 4 with correct symbol for circular target line

as shown below

Needs more work and discussion

As indicated

Where applied

3.5 alternate proposal

· Chinese comment on figures 5, 6, 7, 9, 12:

"In the DIN 3140-6 (withdrawn) the datum axis was

defined by the centre of curvature and the datum

point on the cylinder axis established by a plane of

the curvature at the cut-section with the outer edge

of the lens cylinder. If the datum point is not in this

plane, define the point (distance a) as in figure 4."

3.7

datum point
specified point on the datum axis.

The

location of a datum point must be specified
adequately by dimension lines to other
datum features.

3.9, 3.10

· Both need a figure

1 Rotational axis of aspheric surface
2 Aspheric surface
3 Datum axis

Tilt angle of aspheric surface

L Lateral displacement of aspheric surface

3.10

lateral displacement of an aspheric surface
distance from the

point of rotational symmetry

vertex

of the aspheric surface to the datum axis.

The vertex is the intersection of the asphere's

axis of rotational symmetry with its surface (or

the projection of its surface in the case of off-

axis segments.)

reason: the "point of rotational symmetry" is not

defined, and the intention is different from the

datum of a spherical surface (which is displaced

from the surface by one radius.)

3.11

Tilt angle of an optical element or subsystem

angle between the datum axis of the element or

subsystem and the system datum axis of which

the element or subsystem is a part.

Problem 1: 3.1 defines an optical system as an

optical element, subassembly, or assembly. So

the whole thing is circular and ambiguous.

Problem 2: Now that we're dealing with elements

or subsystems in relation to each other, the

orientation of their tilts and displacements

matter

3.11 ctd

We need to better define what is desired
and intended:
1. Tilt between optical axis and datum axis.
2. Tilt between one datum axis and another.
3. A note explaining that the tilt is in the plane

containing the two axes, i.e. the orientation of

maximum tilt value, unless otherwise specified.

4. A note stating that if the orientation of the tilt is

important then the tilt should be specified in

terms of rotations about X, Y, and Z axes in a

separate text note.

3.12

Lateral displacement of an optical

element or subsystem

Distance between the datum axis of the

element or subsystem and the datum axis
of the system which the element or
subsystem is a part, measured at the
datum point of the subsystem.

My head hurts.

Figure 2

I don't see any hint of how the datum point's

position is defined.

Should we have something in the text?
See 3.7 suggestion:

datum point
specified point on the datum axis.

The

location of a datum point must be specified
adequately by dimension lines to other
datum features.

4.2

If the aspherical effect of the surface is small

compared to its spherical power, the
centring tolerance may be specified in
accordance with 4.1, as if it were a
spherical surface.

3.8, 4.1, and Figure 1

vs. 3.9 and 4.2

Rotational axis of aspheric surface

Datum axis

Surface normal of aspheric surface

When does the aspherical deviation become
"small enough" to switch from one to the other?
We should add an informative note cautioning
about the difference.

Surface normal of spherical surface
(Figure 1.)

5.3.4

...If more than one datum axis is indicated in

the drawing, the reference letters of the
appropriate datum

system

axis

shall be

appended to the tolerance values.

Once around the Maypole

· 3.7 datum point is "specified point on

datum axis."

· 5.2 datum point coincides with point of

symmetry of aspheric surface.

· 3.10 lateral displacement of aspheric

surface is distance from point of rotational
symmetry to datum axis!

Ideas?

Figure 3

· Impractical for thin lenses: Cylinder edge not "of

sufficient size to permit use." (see 3.4)

· Irrelevant for thin lenses: Few lenses are press-

fit into barrels, and diametral tolerance between

lens and cell makes it meaningless.

· Suggestion: Guide designers to use this form

only for lenses with barrel length >

· Per Chinese comment, delete 3b because it is a

simplification fraught with opportunities for error

4/3'

4/2'

4/3'

4/2'

Figure 4

· Legend refers to circular target line B1 but

drawing shows datum B. Indicate B1.

· Remove "C1" on left side of drawing .

4/0,5'

Figure 5

· Per Chinese comment, delete 5b. WAY

too easy to misinterpret even with the
standard in front of you.

4/1'

4/2'

Figure 10 -

A bad example

Don't select two nearly concentric

radii to create a datum axis

Not even tangent!

Better

4/3'

Better still

4/3'

Figure 14

· Legend: "centring tolerance indication for

a surface without optical function"

· Really serves as a good figure for TIR

discussion

0,1

TIR What is it?

Karow: True Indicator Reading = edge thickness

variation

Willey and Parks: Total Indicator Runout = edge

thickness variation

TIR ctd.

Photonics Dictionary: Total Image Runout =

circular orbit diameter of image from a lens
that is rotated about its rim. (Also total
internal reflection.)

TIR ctd.

Yoder, Strong, Brown, DeVany, Laikin,

Horne, Kinglsake, Shannon, Zschommler,
MIL-HDBK-14: No mention

Smith, Zemax manual: Total Internal

Reflection (no other mention)

Malacara: No mention (although "decenter"

is defined in terms of the system axis)

Meadows, Geometrical Dimensioning

and Tolerancing:

· Circular Runout: "provides control of circular

elements of a surface. The tolerance is applied

independently at any circular line element as the

part is rotated 360

. Where applied to surfaces

constructed around a datum axis, circular runout

may be used to control the cumulative variations

of circularity and coaxiality. Where applied to

surfaces constructed at right angles to the

datum axis, circular runout controls circular

elements of a planar surface (wobble.)"

Meadows, Geometrical Dimensioning

and Tolerancing:

· Total Runout: "provides composite control of all surface

elements. The tolerance is applied simultaneously to all
circular and profile measuring positions as the part is
rotated 360

. Where applied to nominally cylindrical

surfaces constructed around a datum axis, total runout is
used to control cumulative variations of circularity,
straightness, taper and coaxiality of a surface. Where
applied to surfaces at right angles to a datum axis, total
runout controls variations of perpendicualrity (to detect
wobble) and flatness (to detect concavity and convexity.)"