The 0.3 mm rule: how to tell if your parts need an optical CMM or a bridge CMM 

Six questions to answer before you look at a single spec sheet

Most CMM decisions start in the wrong place. 

They start with a brochure. A spec sheet. A side-by-side of accuracy figures. 

The right place to start is your inspection bench, with the one part that keeps holding up the line. 

Here’s a number worth knowing: 0.3 mm. That’s the diameter of the smallest stylus tip. If the gap you need to measure is narrower than that, no amount of probing will get you there. And that single fact settles more “bridge or optical” debates than any datasheet. 

So which machine do your parts actually need? Grab that problem part. These six questions will tell you. 

First, the 30-second version 

Both are coordinate measuring machines. The difference is which sensor does most of the work. 

• Optical CMM: a camera leads. Variable zoom optics and independently controlled lighting (top light, back light, segmented ring light) create the contrast for automated edge measurement. Contact and other non-contact sensors back it up. Best for tiny, densely packed, 2D or delicate features. 

• Bridge CMM: a contact probe leads. Best for 3D form and geometry on surfaces a stylus can reach without disturbing them, and for parts bigger than a compact optical system can take. 

One trap to avoid: “bridge” describes the structure, not the sensor. OPTIV S is itself a fixed-bridge, moving-table optical CMM. In this article, “bridge CMM” means a machine where touch probing does most of the measuring. 

Question 1: Can a stylus physically get there? 

Look at pin cavities, narrow slots, closely spaced edges and features hidden inside bores. Then look at the path to them, because the shank behind the tip needs room too. 

If the stylus can’t get in, tactile alone can’t do the job. Vision measures visible edges without contact, and a chromatic white light sensor can reach into small grooves, cavities and holes. If every feature is probe-friendly and you mainly need 3D form, a bridge CMM is the natural fit. 

Question 2: What happens when you touch it? 

Thin walls. Flexible sections. Polished cosmetic faces. Mirror-finish metal. Transparent plastics. 

If the part bends, marks or won’t give you the same reading twice, probing force is your problem. A vision sensor measures delicate materials without deforming them, and a chromatic white light sensor handles reflective and transparent surfaces, including the thickness of transparent parts. Rigid parts with tough surfaces? Contact is fine. 

Reality check: going non-contact removes probing force. It doesn’t remove clamping distortion. If your results change with how the part is held, fix the fixture before you change the sensor. 

Question 3: How many machines does one part visit? 

Count every stop: comparator, vision system, CMM, surface instrument, the lab you outsource to. Every handover adds time, handling and another chance for error. 

If vision is central and you also need contact or other non-contact checks, an optical CMM with multi-sensor capabilities can bring them into one coordinate system, so you measure relationships between features in one go. If contact already covers almost everything, a bridge CMM may be all you need, and bridge machines can take extra sensors too. 

Question 4: How many faces need measuring? 

Datums on one face, the features they control on another. Profile or flatness on several sides. 

An optical CMM with a CNC rotary table turns the part under program control, and PC-DMIS carries the coordinate system through the rotation. No re-clamping, no re-measuring datums. A well-configured tactile setup can also reach multiple faces, so this one comes down to sensor access in the configuration you’re quoted. 

One catch: a rotary table can bring a face to the sensor, but it can’t reveal a feature your fixture is covering. Plan clamps and supports around what needs measuring. 

Question 5: Will it fit, in its fixture? 

This is the clearest dividing line. Optical CMMs are built for small, feature-dense parts, so their measuring volumes are smaller. Bridge CMMs go much bigger. Beyond roughly two metres, you’re into gantry or horizontal arm territory. 

Measure the part in its fixture, not bare. If it rotates, allow for the volume it sweeps and the clearance each sensor needs. For reference, OPTIV S 4.4.3 measures within 400 × 400 × 300 mm with a 30 kg table payload, and larger volumes exist in the range. SoSo, one configuration being too small doesn’t rule optical out. 

Question 6: Where does the job really stall? 

This is the question most suppliers skip. Be honest about the answer. 

• Small or delicate features, or shuttling between optical and contact systems? An optical CMM with multi-sensor capabilities  could combine those checks, cut handling and give you one report. 

• Slow contact measurement on large, rigid parts? A suitably configured bridge CMM may help, if a probe can reach every feature. Compare inspection cycles before you commit. 

• Programming, queues, repeat checks or reporting? That’s not a machine-type problem. Look at software, skills, fixturing and capacity first. 

Sometimes the right answer is “you don’t need a new type of machine.” Better to hear that before you buy than after. 

Your cheat sheet 

• Explore optical when fine features or touch-sensitive surfaces make non-contact measurement necessary, especially when vision has to work alongside other sensors. 

• Explore a tactile bridge CMM when parts are rigid, every feature is probe-accessible, and 3D form is the main job. Check volume and payload with the fixture included. 

• Fix the process first when your current methods already cover every feature and the delay is programming, queues or reporting. 

“But which one is more accurate?” 

Wrong question. Measurement error is declared per sensor, against the relevant part of ISO 10360, for a specific machine and configuration. That compares configurations, not categories. 

The better question: can this method reach and resolve your feature at all? A sub-micron spec is worthless if the sensor can’t see or touch the thing you need to measure. 

Where OPTIV S fits 

OPTIV S is our compact optical CMM for parts that need more than one measurement method. Vision, contact scanning and optional chromatic white light scanning work from one fixture, in one automated cycle. An optional CNC rotary table reaches other faces without re-clamping. PC-DMIS programs every sensor and pulls all results into one report, with every sensor’s measurement error declared against ISO 10360. 

Bring us your most challenging part 

Pick the part that causes the most headaches. Bring it, and its drawing, to a Hexagon measurement specialist. We’ll tell you straight whether an optical CMM, a bridge CMM or a change to your current process would fix the bottleneck.