3D Scanner vs Fixed CMM vs Portable CMM: Which Measurement System Should You Use?

A gearbox housing coming off the line this afternoon needs three different questions answered. Does the cast surface match the CAD model across its whole profile? Do the twenty bolt holes sit inside position tolerance? And can the housing even be carried to a measurement lab, or does the equipment have to come to it? No single instrument answers all three equally well, and picking one based on a single accuracy number on a spec sheet is how shops end up with the wrong tool for the job.

A 3D scanner, a fixed CMM, and a portable CMM (PCMM) are not three accuracy tiers of the same device. They are three different ways of capturing measurement data, and each one is built around a different tradeoff between coverage, discrete-point accuracy, and where the measurement actually happens. This guide compares them across the variables that actually decide the choice: tolerance, freeform surface capability, hole and feature access, speed, environment, part mobility, and reporting output.
 

1. Three Different Approaches, Not Three Tiers of the Same Thing

A fixed CMM is a stationary, contact-based machine, typically built on a granite bed with a bridge, gantry, or cantilever frame, that moves a touch probe to specific XYZ coordinates inside a controlled lab environment. It measures one point at a time, and the measurement result can be reported with traceability and an uncertainty statement for the measurement process.

A portable CMM (PCMM), such as SCANOLOGY's AccuArm, is also a contact-based probe, but it is built as an articulated arm that travels to the part instead of requiring the part to travel to it. It also acquires discrete tactile points, but its kinematics, environment, setup, and uncertainty differ from those of a fixed CMM.

A 3D scanner is a non-contact optical system that captures full-surface data as a dense point cloud or mesh, rather than a handful of discrete points. It answers a different kind of question: not just "is this one hole in tolerance," but "how does this entire surface compare to the reference model."

2. Tolerance: What Each System Can Actually Hold

Fixed CMM performance is commonly evaluated using acceptance and reverification tests in ISO 10360-2 or the related ASME B89.4.10360.2 standard, which define acceptance and reverification tests for linear dimensional measurement on stationary Cartesian CMMs. Under these standards, a well-maintained fixed CMM in a controlled lab can support low measurement uncertainty when the machine, environment, task, and uncertainty budget are suitable, which is why it remains the reference instrument for the tightest tolerance callouts on a drawing.

ISO 10360-12 specifies acceptance and reverification tests for articulated-arm coordinate measuring systems; it does not certify the arm, which governs acceptance and reverification testing specifically for articulated-arm measuring machines. AccuArm's published maximum permissible error figures for its 6-axis, 2 m S-grade configuration list a SPAT (single-point articulation performance) of 0.016 mm, an EUni (point-to-point length measurement error) of 0.022 mm, and a PFORM (probing form error) of 0.015 mm, according to SCANOLOGY's current AccuArm specifications. An EUni figure in that range may be suitable for some hole-position and dimensional measurements, subject to a task-specific capability and uncertainty assessment, though the final conformity call on any specific feature still depends on feature strategy, environment, and the shop's own uncertainty budget, not the MPE figure alone.

3D scanners report accuracy differently, and the distinction matters when comparing across categories. Point accuracy describes how precisely the scanner resolves a single feature directly in front of it, while volumetric accuracy describes how that precision holds up across the full scan volume. SCANOLOGY's SIMSCAN-S Gen2 is rated at 0.015 mm accuracy with 0.015 mm plus 0.03 mm per meter volumetric accuracy, reported alongside references to ISO 10360-13, VDI/VDE 2634 Part 3, and JJF 1951 in the product documentation; each characteristic should be read with its stated method and conditions, per the current product specifications. Comparing a scanner's point accuracy figure against a fixed CMM's linear measurement uncertainty is comparing two different measurement classes, not two numbers on the same scale.

The practical takeaway is to select a fixed or portable CMM from the feature, access, environment, required uncertainty, and validated process—not from a headline specification or tolerance threshold alone. For full-surface deviation and freeform comparison, tolerance is not the deciding factor at all, coverage is.
 

3. Freeform Surfaces vs Discrete Features

This is where the three categories split most clearly by job, not by accuracy. Fixed CMMs and portable CMM arms are built to hit specific, pre-defined points: a bore center, a datum plane, a bolt circle. They are not efficient at capturing a sculpted surface, a cast fillet, or an organic curve, because doing that point by point with a touch probe would take an impractical number of probe hits to approximate what a scanner captures in one pass.

A 3D scanner does the opposite well. It captures the entire surface as a mesh, so a full-panel deviation map, a reverse-engineering scan, or a cast-part surface comparison against nominal CAD comes from a single scan pass rather than a manually planned point program. KSCAN-E, for example, covers a scanning area up to 1440 by 1000 mm per frame at up to 0.020 mm accuracy, which is built for exactly this kind of full-coverage capture on mid-size mechanical parts.

Neither approach replaces the other on a part that has both freeform surfaces and precision datums, which describes most real parts. That is the case for combining scan data with probed points rather than picking one system to do both jobs.

https://www.3d-scantech.com/product/accuarm/
 

4. Holes, Bores, and Other Discrete Features a Laser Can't See Directly

Deep bores, blind holes, internal bosses, and datum features hidden behind a wall of material are a specific weak point for optical scanning. A laser or structured-light scanner needs line of sight to a surface, and a hole deep enough that the scanner cannot see its far wall or floor will not return reliable data for that feature, no matter how good the scanner's accuracy rating is on open surfaces.

This is precisely the gap contact probing is built to close. A fixed CMM's probe stylus can reach into a bore and touch its wall directly, and so can a portable CMM's probe tip. AccuArm's 360-degree joint rotation lets the probe tip swing into a bore or boss from a new angle without unclamping and repositioning the arm's base, which matters on housings and castings with several internal bosses and bores clustered close together.

The practical pattern most shops land on: scan the accessible freeform surfaces, then probe the handful of features the scanner cannot see. AccuArm and SCANOLOGY's scanners share the same DefinSight metrology software, so scan data and probed points end up in one aligned dataset instead of two files that have to be manually reconciled afterward.
 

5. Measurement Speed and Throughput

Speed favors the 3D scanner by a wide margin when the job is full-surface coverage. SIMSCAN-S Gen2 captures up to 8.1 million measurements per second, and KSCAN-E reaches 8.29 million measurements per second, according to current product specifications, which means a full-part scan that would take a CMM program dozens or hundreds of individual probe touches to approximate finishes in seconds.

Contact probing on a fixed CMM or a portable CMM is inherently point by point, and each point takes a probe cycle: approach, touch, retract, move to the next coordinate. That is slower per feature, but each acquired point can contribute to a documented measurement result under a defined procedure, which matters when a report has to report a specific dimensional result rather than characterize an overall surface.

The speed comparison only matters in context. A shop running hourly go/no-go checks on a stamped bracket benefits from scanner throughput. A first-article report certifying twelve specific datum dimensions benefits from CMM or PCMM point-by-point traceability, even if it takes longer per feature.
 

6. Environment: Lab Floor, Shop Floor, or Field

A fixed CMM's accuracy figures assume a controlled environment, typically a temperature-stable metrology lab with vibration isolation, because thermal expansion and vibration both introduce measurement error at the micron level the machine is built to resolve. Moving a fixed CMM out of that environment, or bringing a large or awkward part into it, is usually not practical.

AccuArm is built for the opposite assumption. Its carbon fiber arm construction, hot-swappable batteries, and USB Wi-Fi connectivity are designed so the arm goes to the part on a shop floor or in the field rather than requiring the part to come to a lab. SCANOLOGY's handheld and tracking scanners follow the same logic: KSCAN-E and NimbleTrack Gen2 both operate wirelessly, which removes cable routing as a constraint when working around fixtures, tooling, or an assembly line.

For any part that is too large, too heavy, or too fixed in place to move, the environment question effectively pre-selects between a fixed CMM and the two portable options before tolerance or surface type even enter the decision.

https://www.3d-scantech.com/scanology-unveils-nimbletrack-gen2-further-faster-mighter/
 

7. Does the Part Move, or Does the Equipment Move?

This is the practical flip side of the environment question, and it is worth separating out because it changes cycle time even when accuracy requirements are identical. Every part sent to a fixed CMM has to be removed from its fixture or line, transported, mounted, and eventually returned, and that handling time adds up fast on high-mix or high-volume production, independent of how long the actual measurement takes.

Portable CMM arms and 3D scanners both go to the part instead. That matters most for parts on an assembly line, weld fixture, or vehicle chassis that cannot be pulled without disrupting production, and for parts too large to fit a CMM's working envelope in the first place. NimbleTrack Gen2 extends this further for large-volume work, using an optical tracker to hold one stable coordinate frame across a tracking distance of up to 4.2 meters, so a full vehicle body panel or large tooling fixture can be measured in place rather than stitched together from multiple repositioned setups.
 

8. Reporting and Data Deliverables

Fixed CMM and portable CMM reports are built around discrete coordinate data: a list of measured points compared against nominal values, with pass or fail status per dimension. That format is exactly what a formal first-article inspection report or a customer PPAP submission typically requires for specific datum callouts.

3D scanner output is built around full-field comparison: a color deviation map showing where the entire surface sits relative to CAD, not just a sampled set of points. DefinSight handles scanning, meshing, alignment, and inspection in one interface, generating both GD&T call-outs on discrete features and full-surface deviation maps from the same dataset, so a report can include both without exporting data between separate programs. For lighter deliverables, such as a quick visual comparison or a printable mesh, SCANOLOGY's free ScanViewer covers distance checks and color deviation maps without a full metrology license.

The two report styles are not competing formats, they answer different questions on the same part, which is why a combined scan-plus-probe workflow often produces a single report covering both.
 

9. Matching the System to the Job

Freeform surfaces, full-panel deviation, or reverse engineering work point toward a 3D scanner. SIMSCAN-S Gen2 handles small, hard-to-reach parts and fine features, KSCAN-E covers mid-size mechanical components with mixed surface density, and NimbleTrack Gen2 extends coverage to multi-meter parts that need one stable coordinate frame.

Discrete critical dimensions on a part that cannot leave the shop floor point toward a portable CMM. AccuArm probes bore centers, datum planes, and bolt patterns directly on the line, and its reach options from 2 m to 4.5 m across the S, E, and C grades let buyers match arm length and accuracy grade to the part's envelope and tolerance instead of over-buying reach they don't need.

Measurements requiring very low uncertainty on a limited number of tight-tolerance features still point toward a fixed CMM, which SCANOLOGY does not manufacture. The practical path most shops use is to cover full-surface inspection and field-level discrete features with a scanner and AccuArm, then reserve a lab's existing fixed CMM for the handful of dimensions that genuinely need that level of measurement capability and documented uncertainty, rather than routing every part through the CMM queue.

Most real parts need more than one of these three approaches, which is exactly why AccuArm and SCANOLOGY's scanners share the same DefinSight dataset instead of operating as separate, disconnected tools.
 

Frequently Asked Questions

Is a portable CMM the same thing as a 3D scanner?
No. A portable CMM like AccuArm is a contact probe on an articulated arm that measures discrete points, the same measurement principle as a fixed CMM, just outside a lab. A 3D scanner is a non-contact optical system that captures full-surface data. They solve different problems and are evaluated using different performance standards, ISO 10360-12 for AccuArm versus ISO 10360-13 or VDI/VDE 2634 for SCANOLOGY's optical scanners.

Can AccuArm replace a fixed CMM entirely?
AccuArm's published EUni of 0.022 mm for the 2 m, S-grade configuration should be assessed with the feature strategy, environment, and complete uncertainty budget before making a conformity decision. For the small number of features that require very low measurement uncertainty, a fixed CMM in a controlled lab remains the reference instrument, and AccuArm is built to complement that workflow rather than replace it.

Which system is faster for shift-level quality checks?
A 3D scanner is faster by a wide margin when the check is a full-surface pass/fail against a tolerance band. SIMSCAN-S Gen2 and KSCAN-E both scan in the millions of measurements per second, which suits hourly or shift-level sampling far better than a point-by-point CMM program.

Do I need both a scanner and a portable CMM?
If the part has both freeform surfaces and a handful of critical discrete features a scanner cannot see directly, such as deep bores or internal bosses, yes. Scan the surfaces, probe the features with AccuArm, and both data sets land in the same DefinSight file rather than requiring manual reconciliation.

What does a fixed CMM report look like compared to a scanner report?
A fixed CMM report lists discrete measured points against nominal values with pass or fail status per dimension. A scanner report from DefinSight includes a full-surface color deviation map alongside GD&T call-outs on specific features, covering both a complete-surface view and discrete tolerance checks from one dataset.