Metal FFF sintering shrinkage rate measurements by X-ray computed tomography

Fabien Léonard, Samuel Tammas-Williams

Research output: Contribution to journalArticlepeer-review


Fused filament fabrication (FFF), also known under the brand name fused deposition modelling, has recently emerged as a more straightforward and economic alternative to established 3D printing or additive manufacturing technologies such as powder bed fusion for metal part manufacturing. In this work, we report on possible approaches to the measurements of FFF sintering shrinkage rates from X-ray computed tomography (XCT) data, with particular focus on those that use greyscale data, and are thus independent of any errors introduced by segmentation. Metallic parts were produced using FFF technology, which relies on an isotropic scaling factor applied to the STL model of the part to be produced to compensate for the shrinkage happening during the sintering of the printed (green) part. We report that for the stainless-steel parts produced, the shrinkage rates were not isotropic and the one along the build direction was the largest, the other two being similar to one another. Overall, the shrinkage rate measured by the different methodologies all fell between 15 % and 16 %. XCT data also clearly showed the typical defects encountered with this technology, such as voids resulting from insufficient overlap or caused by print head blockages, are not mitigated by the sintering process. Autocorrelation was introduced as a way of characterising the 3D structure of the sample before and after sintering, and thus the sintering shrinkage rates.

Original languageEnglish
Pages (from-to)631-644
Number of pages14
JournalNondestructive testing and evaluation
Issue number5
Early online date11 Aug 2022
Publication statusPublished - 3 Sep 2022


  • Fused filament fabrication
  • SSR
  • autocorrelation
  • fused deposition modelling


Dive into the research topics of 'Metal FFF sintering shrinkage rate measurements by X-ray computed tomography'. Together they form a unique fingerprint.

Cite this