3D Scanning Shiny Metal Parts Without Spray: A Field Test with Ultra-Black Fabric
ANSOKEN Customer Visit, Case 004: A precision sheet-metal shop in Tokyo
Tackling a 3D scanner that cannot capture shiny, reflective metal parts
Where is our "black" actually being used? In this series, we visit customers who use our materials and report what we learn on site.
What you will learn
- Why metal parts often fail to scan on a 3D scanner
- Field results of reducing reflections without matting spray
- The finding that a key part of the fix was not the part itself, but the surface it sits on
- The open issues on accuracy and repeatability, since this test is still in progress
This time we visited a precision sheet-metal shop in Tokyo that makes parts for industries including semiconductor manufacturing equipment. They had introduced a 3D scanner to speed up dimensional measurement, but it could not properly scan metal parts because of their gloss and reflections.
A 3D scanner projects laser or structured light onto an object and captures its shape as a set of points (a point cloud) from the light that returns. It is used for dimensional measurement and automated inspection, but highly reflective surfaces such as bare metal do not return the light correctly, which leaves gaps in the data. This is a common problem across the industry.
Note: this test is still in progress. We hope it is useful as a hint for choosing materials and setting up your scans.
Who this article is for
- Sheet-metal and metalworking shops considering a 3D scanner
- Anyone who has a 3D scanner but struggles to scan metal parts
- Anyone looking for an alternative to matting spray for reflection control
- Manufacturers working to automate measurement and inspection
- Anyone having trouble 3D scanning reflective or glossy materials
About the shop
The shop specializes in precision and prototype sheet metal, handling prototypes and small lots for fields such as medical robots and semiconductors, where tight punching and bending tolerances are required.
Why do shiny metal parts reflect so badly in a 3D scanner?
The shop introduced a Keyence VL-800 3D scanner for dimensional measurement, replacing much of their manual caliper work.
The demo before purchase used only plastic parts, which scanned without problems. After installation, however, metal parts would not scan properly because of reflections. Highly reflective surfaces like metal are technically difficult for any 3D scanner that uses laser or structured light.
On the right is the scan data. The black areas were not captured because of reflections, and the surface, which should be flat, came out wavy. Metal reflects, of course, but can such a reflective part be scanned at all?
On-site testing: changing every condition we could
The shop had already tried a 3D scanning spray. The scanner maker recommended matting spray, but the people doing the measurements had practical concerns:
- The coating thickness varies from spot to spot
- Results depend on who applies it
- Scanning spray is expensive
- It is hard to hit exactly the right area
Could they avoid relying on spray, with its running cost? That was the starting point of this test.
The material we brought was our ultra-black fabric, Musou Black Fabric KIWAMI, which reflects only about 0.1% of visible light (at 550 nm). In an earlier test, we had confirmed that laying this fabric under the object absorbs light reflected from the floor and reduces noise in the scan. This time we tested what it could do when scanning the metal part itself.
Together with the measurement staff, we cut the fabric into a 50 cm circle and laid it on the stage of the VL-800. We then changed the following conditions one at a time. We tested two materials, stainless steel and aluminum, both among the hardest metals to scan.
- Scan brightness
- Stage height
- Raising only the metal part on a block
- How the part is placed (its angle)
- Lining the inside wall of the scanner with the fabric
- Scan mode (high definition / standard)
Lining the inside wall of the scanner did not help much in this test. The problem was reflection from the metal part itself, and treating the reflections inside the scanner did not reach it.
Why the breakthrough was on the stage surface
As the test went on, it became clear that both the reflection from the surface the part sits on and the reflection from the metal part itself were disturbing the scan.
First, we covered the whole stage with Musou Black Fabric KIWAMI and placed the metal part on it. With the part lying flat (left), reflections still prevented a good scan. That suggested a new hypothesis: the bottom face of the part itself was reflecting light.
Changing the angle of the part
The result, from a single scan so far: no part of the data appears to be missing because of reflections. Reflection from the stage was disturbing the scan of the metal part. Based on this result, the countermeasures are:
- Lay Musou Black Fabric KIWAMI on the stage
- Place the metal part so that no flat face lies parallel to the stage
Note also that the color of the part in the scan data is now closer to the real part.
Remaining issues with dimensional accuracy
We made progress, but the problem is not fully solved. Here are the issues that remain:
- The scanned dimensions tend to be about 0.08 mm smaller than the actual size. Ideally they should match caliper measurements
- We have too few tests yet to say whether the results are within tolerance
- Depending on part geometry, reflection from the stage cannot always be avoided
- Some areas are still hard to scan, such as the inside corner of an L-shape and the sides of holes (reflections occur easily, or the light cannot reach)
- Finding the right angle for each part takes time
Summary: reflection control is not only about the part, but also about how and where it sits
The biggest lesson from this test is that reflections can be reduced not only by treating the object itself, but also by controlling reflections from the surrounding environment, especially the surface the part sits on. Treating the inside of the scanner had little effect, while treating the stage surface led to real progress.
Many companies that have introduced similar scanners seem to struggle with reflections from metal parts. If you are measuring metal parts with a 3D scanner, or facing similar reflection problems, please feel free to contact us.
We plan to continue our visits and tests with the shop and will report on the next steps. Many thanks to the team for their time and cooperation on site. (Test date: July 30, 2026)
FAQ
- Q. Why do shiny metal parts fail to scan on a 3D scanner?
- A. On glossy, highly reflective metal surfaces, the laser or structured light does not return to the sensor correctly, so data is lost. Compared with plastic parts, metal parts are among the hardest objects to 3D scan.
- Q. Is there a way to reduce reflections in 3D scanning without scanning spray?
- A. In this test, laying Musou Black Fabric KIWAMI on the stage and placing the part so that no flat face lies parallel to the stage reduced the reflection problems. The test is still ongoing, and the effect may be limited depending on part geometry.
- Q. Do 3D scan dimensions match caliper measurements?
- A. So far, the scanned dimensions tend to be about 0.08 mm smaller than the actual size. Whether this is within tolerance needs further testing.
Ultra-black fabric with about 0.1% reflectance at 550 nm. Available from A4 size. VL Flock Sheet
A lower-cost black flocked sheet for backgrounds and stage covers.
Originally published in Japanese on August 25, 2026. Translated into English in October 2026.
Related: What Is Reflectance? The Science Behind True Blackness
