How to Keep 100 Complex Metal 3D Printed Components the Same Size

The first five parts drop straight into place. By the thirtieth the fit feels tighter. By the eightieth some simply will not assemble. When you order 100 complex Metal 3D Printed Components, the real problem is rarely one outright reject-it is that every part is a little different from the last. That slow drift can stop an assembly line, force selective fitting, or wreck a production schedule.

This article explains why parts from the same build still vary, what realistic consistency looks like in practice, and the concrete steps that keep dimensions stable from piece one to piece one hundred. A production example and a supplier checklist follow.

Even on the same machine with the same file, small differences appear. Four factors account for most of the variation.

Heat and shrinkage
The laser creates steep temperature gradients. Residual stress builds as layers cool at different rates. Later, when the part is heat-treated or cut from the plate, that stress relaxes and the part can move. Industry measurements commonly show residual stress and the resulting distortion as one of the largest sources of dimensional error in powder-bed fusion-often contributing several tenths of a millimeter if left uncontrolled.

Where the part sits on the build plate
Cooling rates, gas flow and support conditions change from the center of the plate to the edges. An identical part printed in one corner can finish a few tenths of a millimeter different from the same part printed in the middle of the same build.

Powder quality changes
Fresh powder and reused powder are not identical. Particle size distribution, oxygen level and flowability shift with recycling. If the powder lot is not tightly managed, the melt pool and final density can vary enough to affect finished size.

Machine condition
Laser power, focus and recoater performance drift over time. Without regular checks, the same digital file can produce slightly different results across consecutive builds.

These effects stay hidden on a single prototype. Across 100 parts they become visible and costly.

Buyers sometimes request ±0.05 mm on every feature. That is achievable on a few critical faces after secondary machining, but it is rarely realistic for as-built surfaces across a complex part.

In well-run industrial SLM, typical results look like this:

General dimensions: ±0.1 to ±0.2 mm (or about ±0.1–0.2 % of nominal size)

Critical interfaces after light machining: ±0.05 mm or better

Larger or high-stress parts with loose process control: variation can reach ±0.3 mm

The practical approach is to mark the dimensions that actually affect assembly or function and accept functional tolerances on the rest. Over-specifying every surface raises cost and lead time without improving the finished product. Clear agreement on critical features is the foundation of reliable metal 3D printing dimensional accuracy.

Design the part for the process
Orient critical surfaces to reduce support contact and residual stress. Large flat areas parallel to the build plate often move more during cool-down; tilting or adjusting them early saves trouble later.

Lock the material and powder batch
Specify the exact alloy and keep the same powder lot-or strict recycling limits-for the entire order. Switching powder mid-job is one of the quickest ways to introduce size drift.

Freeze the printing parameters
Once a test build confirms the settings, leave laser power, scan speed, layer thickness and hatch strategy unchanged. Even small "improvements" during production can shift the whole batch.

Keep layout and orientation consistent
Print every part in the same orientation and, as far as possible, in similar locations on the plate. Random placement adds unnecessary variation.

Compensate for shrinkage with a short test series
Run a few calibration pieces, measure them, and apply a scale factor or offset to the build file before launching the full quantity. This step alone often removes the largest systematic error.

Stress-relieve while parts are still on the plate
Heat treatment before the parts are cut free lets residual stress relax while the geometry is still constrained. Cutting first and heat-treating later is a frequent cause of late distortion.

Measure and record
Use a CMM or structured-light scanner on first articles and on a defined sampling plan through the batch. Keep the data. Traceable measurements are the only reliable way to prove consistency and catch drift early.

These controls turn a 100-piece order from a gamble into a repeatable process.

A European equipment maker needed 100 stainless-steel housings with internal channels and several precision mounting faces. Early samples from a previous supplier assembled cleanly, but the first production batch showed mounting-hole positions drifting by as much as 0.25 mm toward the end of the run. Selective fitting became necessary and the schedule slipped.

The order moved to a controlled process. The team locked the powder batch, fixed every parameter after a short test series, kept identical orientation on every build, performed stress-relief heat treatment while the parts remained on the plate, and machined only the critical mounting faces. First-article and in-process CMM reports were shared with the customer.

The 100 parts arrived with critical dimensions held inside ±0.10 mm. Assembly ran without selective fitting. A follow-on order of the same quantity, run with the same process window, produced matching results. The earlier drift of 0.25 mm was eliminated mainly by on-plate heat treatment, locked parameters and consistent orientation rather than by any single machine setting.

Before committing to 100 pieces, get clear answers on these points:

Will you supply a first-article inspection report with actual measured values?

Can the same powder batch (or documented recycling limits) be maintained for the full order?

Do you stress-relieve parts while they are still attached to the build plate?

Will in-process measurement data be shared, not only a final certificate?

Can the same parameters and part orientation be held across multiple builds?

What dimensional capability do you typically achieve on parts of similar size and complexity?

A capable metal 3D printing factory or batch metal 3D printed parts supplier answers these questions directly and backs them with data. Vague replies or reluctance to share measurements are warning signs. For ongoing needs, also confirm the ability to reproduce the same process window for wholesale custom metal parts months later. Looking for a reliable SLM 3D printing manufacturer that already runs disciplined process control will save more time than chasing the lowest unit price.

Complex parts with internal features are often practical only with additive manufacturing when the quantity is 100. Simple prismatic shapes that need tight tolerances on many surfaces can still be more consistent and economical on CNC. A separate comparison of the two routes for low-to-medium volumes examines that decision in detail.

If size consistency across 100 complex metal parts is critical, send the CAD files and a drawing that highlights the important dimensions. A short process review will show whether the geometry, material and tolerances can be held reliably-and which controls are needed so the hundredth part matches the first.