Choosing a Telecentric Lens: The Four Numbers That Decide It

Four numbers, checked in the right order

A telecentric lens is described by a short list of numbers, and four of them decide whether a model fits: magnification, depth of field, working distance and distortion. Everything else is secondary for most applications.

The order matters because each check eliminates candidates cheaply. Magnification and sensor format fix the field. Depth of field against the part's height variation removes models that cannot stay in focus. Working distance against the machine layout removes models that cannot be mounted. Distortion against the tolerance removes models that cannot measure accurately enough.

Running the checks in a different order wastes effort, because a model that fails an early check cannot be rescued by passing a later one.

Across the POMEAS telecentric range magnification spans 0.061 to 12.5 X, depth of field spans 0.017 to 172 mm and working distance spans 9.5 to 468 mm, so the checks genuinely discriminate between models.

The result of the sequence is usually two or three candidates, and the final choice turns on availability and price rather than optics.

Start from the tolerance, not from the part

The instinct is to start from the part size and work outward. The better starting point is the measurement tolerance, because it sets the resolution requirement, which in turn constrains magnification and depth of field together.

A tolerance of twenty micrometres needs a lens resolving roughly seven micrometres, which can be met at moderate magnification with usable depth. A tolerance of two micrometres needs a lens resolving under one micrometre, which usually means high magnification and very shallow depth.

Starting from the tolerance makes that consequence visible immediately, before any model is examined.

It also prevents over-specification, where the finest available lens is chosen and the resulting depth of field turns out to be impractical for the fixture.

For applications that only inspect rather than measure, the tolerance is replaced by the smallest defect that must be detected, and the same logic applies.

Then check depth of field against height variation

Depth of field is the constraint that most often decides between otherwise similar models. It has to cover the part's height variation within a part, between parts, and any fixture and positioning error.

Once that sum exceeds the available depth, the options are a lower magnification, an autofocus axis, or better fixturing. Buying a different lens of the same magnification does not help, because depth of field at a given magnification is fundamentally limited.

This is why high magnification telecentric lenses are used on flat, well-presented parts and rarely on castings or stamped components without an autofocus axis.

The measured depth of field range across the POMEAS telecentric models spans 0.017 to 172 mm, which reflects the magnification spread rather than a quality difference.

Recording the height budget as a number, including every contributor, is what makes this check meaningful rather than approximate.

Field of view follows from sensor and magnification

Because a telecentric lens holds magnification substantially constant, the field of view is simply the sensor size divided by the magnification. There is no distance dependence inside the telecentric range, which removes a variable from the calculation.

That also means the field cannot be adjusted by moving the camera, so the sensor and magnification must be chosen together. Changing the field means changing one of them.

Field of view across the POMEAS telecentric range spans 0.4 to 265 mm, so most requirements can be met at more than one format, and the choice then becomes a resolution decision.

Confirming the required field with margin for positioning error, before selecting a model, avoids the most expensive kind of rework.

It also fixes the sensor format, which then feeds back into the resolution and sampling check.

Working distance and distortion close the selection

Working distance must accommodate the machine layout: clearance for the part, for the illumination and for any handling mechanism. Across the POMEAS range it spans 9.5 to 468 mm, which covers everything from close inline stations to stand-off installations.

Distortion is the last check and applies mainly to measurement applications. Across the range it spans 0.002 to 0.06 %, and for inspection tasks it can usually be ignored.

Telecentricity, specified between 0.02 to 0.04 deg for the models that publish it, is closely related and matters when the part cannot be presented perpendicular to the axis.

These two properties complete the four-number check and usually leave a shortlist that can be resolved commercially.

Checking them last is correct, because they eliminate fewer candidates than the first three.

Where telecentric is the wrong answer

Telecentric lenses solve perspective error and magnification stability, and they do it at the cost of field size, working distance and price. Where none of those errors matter, a standard fixed focal length lens does the job for less.

For presence detection, classification and coarse guidance, perspective error is irrelevant, and a standard FA lens with a wider field is usually the better choice.

For very large fields the telecentric options become large and expensive quickly, and a standard lens with software calibration is often more practical.

For parts presented at a controlled height with a tight tolerance and a modest field, telecentric is clearly correct, and the four-number check will confirm it quickly.

Recognising the boundary prevents both over-buying and the opposite error of using a standard lens where perspective error will dominate the tolerance.

Putting the check into a specification

Write four lines: required field with margin, measurement tolerance or smallest detectable defect, height variation budget, and available working distance. Add the sensor format if it is already fixed.

Each line becomes a number that eliminates candidates. The specification is then short, checkable and quotable, and a supplier can respond against it without ambiguity.

Where a line cannot be filled in, that is the real state of the project and it should be resolved before a model is ordered rather than after.

This is also the document that makes later changes visible: if the tolerance changes, its effect on the model is obvious.

Keeping it short is deliberate, because a specification with twenty requirements is harder to check and easier to ignore.

Frequently asked questions

Do I always need a telecentric lens for measurement?

No. It is needed when perspective error or magnification variation would exceed the tolerance. For flat parts presented perpendicular to the axis with a loose tolerance, a standard lens with calibration can be adequate. Telecentric costs field size and working distance.

What is the most common mistake in telecentric lens selection?

Choosing the highest resolution model available and then discovering the depth of field is too small for the part. Depth of field falls with magnification, so the tolerance and the height variation must be checked together before a model is shortlisted.

Can I change the field of view by moving a telecentric lens?

No. Magnification is substantially constant within the telecentric range, so the field is fixed by sensor size divided by magnification. Changing the field requires a different sensor or a different magnification, which is why the field must be specified first.

How much does telecentricity matter?

It matters when the part cannot be presented perpendicular to the optical axis. Telecentricity error moves the apparent edge position as the part moves in height, which a plain depth of field figure does not describe. For flat, well-presented parts it is usually secondary.

Is a larger sensor always better with a telecentric lens?

Only up to the lens image circle, and only if the extra pixels are needed for resolution. Beyond the image circle the corners lose illumination and accuracy. A larger sensor also costs readout speed and money without necessarily improving object-space resolution.

When should I use a telecentric lens instead of a microscope?

When the field is larger than a microscope covers at the required magnification, or when the part has height variation that would change magnification on a conventional objective. Telecentric holds magnification substantially constant, which a standard microscope does not.

Lighting and camera lens selection

• Machine Vision Lighting Selection Basics
• Industrial Camera Lens Selection