Abstract <p>In the industrial introduction of additive technology, it is important to identify internal defects of the added layer (real time online monitoring) when producing large and expensive components whose reliability is essential. To that end, an additive system for the production of metal parts with X-ray tomographic monitoring of the added layers was developed in [1]. The tomographic images are analyzed by means of artificial intelligence (neural networks), with feedback of the results so as to adjust the production conditions. The development of specialized X-ray tomographic systems for additive technology and specialized algorithms for reconstructing the tomograms in the case of a conical X-ray beam and spiral scanning was considered in [2–7]. In the present work, the production of additive layers (with and without defects) of 08G2S alloy steel by WAAM (wire arc additive machining) technology is studied experimentally [8]. This is the fastest additive technology, associated with the least production defects, and so is in great demand for large and expensive components where reliability is essential and hence defects are impermissible. The structure of slots in the added layers is investigated by optical microscopy. The tomograms of layers with and without defects obtained by means of a high-resolution industrial X-ray tomographic system are also studied. The tomograms of the added layers are compared with the observed slots, so as to determine whether X-ray tomography may be used for online monitoring of the quality of the added layers, with neural network analysis of the tomograms to detect defects.</p>