Keyence Measurement Microscope Alternatives: What One Machine Can Actually Measure

An integrated measurement microscope is sold on one number: the accuracy it holds. The difficulty appears when it has to be second-sourced or replaced after discontinuation, because the number on the datasheet is almost never the number your part experiences — it is taken on a standard artefact, at one magnification, over one part of the field, under a lighting setting you may not be running.

In the POMEAS global company intelligence set of 2,602 companies, 54 have publicly documented Keyence equipment in use on their own lines: a station was built around one integrated instrument, that instrument is now a single source, and someone must decide what a second source has to prove. Six parameters decide it, listed below with the values POMEAS publishes. Where a value is not published, the table records a dash.

The six parameters to verify before substituting a measurement microscope

1. The field captured in one exposure, and the envelope around it

An instrument that reads a whole part at once has two size limits, and confusing them is the most common specification error: the field captured in one exposure, and the stage travel, which is larger but costs cycle time each time it is used. IMAGE 3 Series publishes both — Ø100 mm as the largest single-shot field on IMAGE 3, 25 × 25 mm as the maximum at high magnification on MAX2, and 300 × 200 × 75 mm as the envelope depending on model. A part that fits the envelope does not automatically fit the field; one measured across several acquisitions falls under section 4.

2. Pixel equivalent inside that field

The floor on what one frame can resolve is set by one division:

pixel equivalent (µm/pixel) = field width in µm ÷ pixel count across that width

IMAGE 3 gives one input but not the other: a 1-inch 20-megapixel monochrome CMOS, two on some models, with single-shot fields of 6.4 × 4.8 mm / 18.8 × 14.1 mm and 6 × 6 mm / 25 × 25 mm. The horizontal pixel count is not published, so the result is not published either. MICRO IMAGE3 inverts the problem: the image is 4024 × 3036, but the field depends on the objective fitted — objective magnification 1X–150X, digital display magnification 1X–8000X, and only the first is optical. Sub-pixel extraction stabilises an edge already resolved; it does not create detail.

3. The accuracy figure, and the area it was quoted over

An accuracy number without an area is not a specification. The IMAGE 3 pages attach area qualifiers to the published figures — the set includes 5 × 5 mm, 20 × 20 mm, 113 × 105 mm, 185 × 95 mm, 200 × 105 mm, 180 × 180 mm, 200 × 200 mm, 270 × 180 mm and Ø80 mm — because a machine holding a tight figure over 5 × 5 mm will not hold it over 270 × 180 mm. Which qualifier belongs to which model is not stated on the cached page. The headline value is ±0.25 µm, model dependent, described on MAX and MAX2 as high-precision repeatability: the spread of repeated readings, not closeness to truth.

4. What carries the seam error when the part exceeds one field

Once a part exceeds one field, the result depends on stitching, and the seam error is carried by the positioning system, the residual distortion in each field, and the compensation model. MetX Series is built for this: grating scales at 0.1 µm or 0.5 µm, multi-region capture and stitching, and compensation for linear error, squareness error, multi-layer spatial plane error and combined error. Where a stage is used instead of a fixed field, ask for stage repeatability — MICRO IMAGE3 publishes 1 µm across a 100 × 100 mm motorised platform, with 200 × 200, 300 × 200 and 300 × 300 mm platforms available.

5. How the Z value is produced

Height is the parameter most often skipped, because an integrated microscope produces it four ways:

- Autofocus travel — MICRO IMAGE3 publishes 30 mm of Z autofocus travel, Z resolution to 0.1 µm, up to 75 mm of sample height. - A separate height sensor — on IMAGE 3, an 8055-point spectral confocal with 55 mm working distance, ±3 mm range, ±2 µm repeatability, 17 µm spot and 0.02 %F.S. linearity, optional on MAX2, with the Z travel term L limited to 75 mm from the spectral focus zero. - A scanning vertical axis — WLI 1000 scans 0–10 mm with vertical resolution below 0.2 nm, on a 50 × 50 mm motorised stage, with 20 mm motorised plus 75 mm manual Z and ±12° manual tilt. - A compound stack — MetX accepts white-light confocal, 3D line laser and probe modules alongside the image channel.

6. Which illumination channels exist, and which one your edge comes from

An edge is produced by a lighting configuration, not by a part. IMAGE 3 is configured with surface light, coaxial light and contour backlight, adjustable for metal, plastic, dark materials, reflective surfaces and low-contrast workpieces. MetX uses a five-ring, eight-zone high-brightness white cold source for surface work and a high-brightness parallel telecentric cold source for contour. WLI 1000 adds brightfield to its PSI, VSI and extended-depth modes. Record which channel your current recipe uses before anything changes.

Where POMEAS instruments land on those six

How to read this table

Values in different rows were not measured under the same conditions. The ±0.25 µm on IMAGE 3 is repeatability on a bench instrument; the ±2 µm on LM-1030 is measuring accuracy on a moving cylindrical part. The dashes that matter most are the single-shot fields: neither MetX nor MICRO IMAGE3 is sold on one fixed field.

What those numbers mean on the line

Scenario 1 — Batch of small flat parts, many 2D features

IMAGE 3 Series fits. The part is placed in any orientation, the system recognises position and direction, one press runs the program, and under the right model and recipe the series measures up to 300 locations and 100 workpieces in one program. With Ø100 mm inside one acquisition, cycle time is nearly independent of feature count; the limit is resolution across that field.

Scenario 2 — Small features that need magnification and a moving stage

MICRO IMAGE3 fits when the feature is smaller than the field can resolve: objectives 1X–150X on 4024 × 3036, a 100 × 100 mm motorised stage at 1 µm repeatability, and 30 mm of Z autofocus at 0.1 µm resolution for parts with height differences. The trade is cycle time — forty features cost forty positions.

Scenario 3 — A part larger than one field, with height in the specification

MetX Series covers it: a selectable range from 300 × 200 × 200 mm to 600 × 800 × 200 mm, grating scales at 0.1 µm or 0.5 µm, compensation across the travel, and multi-region capture and stitching. Where the requirement is thickness, step, flatness or profile, the confocal, line-laser or probe module produces it, not the image channel.

Scenario 4 — Nanometre surface work, or continuous on-line dimension

Two instruments, not interchangeable. WLI 1000 handles the surface question: vertical resolution below 0.2 nm over a 0–10 mm scan, in PSI, VSI, extended-depth or brightfield mode. LM-1030 handles the line question: 0.3–30 mm non-contact at ±2 µm and 2,400 samples per second, with a GaN green LED, telecentric optics and HL-CCD continuous exposure across 160 ± 40 mm, at 0 to +50 °C, 35 % to 85 % RH non-condensing, IP64.

Writing the replacement specification when the original is discontinued

The failure mode is copying the headline number off the old datasheet into the new specification. That produces a document nobody can test, because the old number carries conditions never written down. Write six lines: features and tolerances per feature; largest envelope and whether one field is required; material, surface condition and the illumination producing the edge; datum and fixturing; environment, if line-side — LM-1030 publishes 0 to +50 °C, 35 % to 85 % RH non-condensing, IP64; and output format. Then require a quote against that document, feature by feature, with the area qualifier beside each accuracy figure.

What this comparison does not tell you

This is a parameter reference, not a compatibility statement. No claim is made that any POMEAS instrument is a drop-in replacement for a specific third-party measurement microscope, and none should be inferred. Whether a substitute works depends on the part, fixturing, datum, lighting and recipe, none of which a datasheet settles.

Third-party brand names appear only as recorded in customer-supplied intelligence: companies describing their own installed equipment in public material. No third-party specification is quoted, because none was verified. The count of 54 is evidence about installed base, not an assessment of any product.

Optical resolution is not measurement accuracy. Resolution follows from optics, sensor pitch and wavelength; accuracy additionally depends on edge contrast, illumination stability, datum definition, thermal state and algorithm. This article does not repeat the method comparison already published on this site between one-press image measurement and a conventional VMM.

FAQ

Is a quoted ±0.25 µm the accuracy I will get on my part?

No. It is a model-dependent figure for the IMAGE 3 Series, described on MAX and MAX2 as high-precision repeatability under stated conditions over a stated area.

How do I work out the pixel equivalent myself?

Divide the field width in micrometres by the pixel count across that width. POMEAS publishes the field and the 1-inch 20 MP sensor, but not the horizontal pixel count.

Can one measurement microscope do both dimensions and height?

Only with a height channel — IMAGE 3 offers an 8055-point spectral confocal at ±3 mm; MICRO IMAGE3 gives 30 mm of Z autofocus at 0.1 µm; MetX accepts confocal, line-laser and probe modules.

What decides the error when my part is bigger than one field?

Positioning, residual distortion in each field, and the compensation model. MetX publishes 0.1 µm or 0.5 µm grating scales with linear, squareness and spatial compensation.

Why is the field of view a dash for some models?

Because those instruments are not sold on one fixed field — MetX is built around stitching, MICRO IMAGE3 around interchangeable 1X–150X objectives.

What should I send a vendor when asking for a second source?

Four numbers: largest part envelope in millimetres, smallest tolerance, smallest feature, and material with surface condition.

---

Next Step

If you are specifying a replacement and need measured values for your own conditions — your sensor size, your working distance, your feature size — send us those four numbers and we will run them against the current catalogue and tell you what actually fits, including when nothing does.