Plasma Transferred Arc deposition of powdered high performances alloys: process parameters optimisation as a function of alloy and geometrical configuration

https://doi.org/10.1016/j.surfcoat.2004.02.013Get rights and content

Abstract

The deposition of high wear- and corrosion-resistant alloys through Plasma Transferred Arc (PTA) is an easily automated process combining the possibility to obtain very thick coatings with high deposition speeds, low thermal distortion of the part and negligible dilution levels, thanks to a very high energy-concentration [S. Kalpakjian, S.R. Schmid, Manufacturing Engineering and Technology—4th edition, Addison-Wesley Publishing, 2001, p. 789–790, Scripta Materialia 37 (1997) 721, Applied Surface Science 201 (2002) 154, Surface and Coatings Technology 71 (1995) 196, Journal of Materials Processing Technology 128 (2002) 169, M. Bonacini, Plasma ad arco trasferito per riporti saldati con superleghe in polvere, “Saldatura e taglio termico verso il 2000” IIS Meeting, October 3rd 1998, Milan-Italy, Wear 250 (2001) 611, Wear 249 (2002) 846, Composites Science and Technology 58 (1998) 299]. Literature studies regard mostly wear or high temperature behaviour of the deposited alloys and plasma hardening treatment without powder [Applied Surface Science 201 (2002) 154, Surface and Coatings Technology 71 (1995) 196, Journal of Materials Processing Technology 128 (2002) 169, M. Bonacini, Plasma ad arco trasferito per riporti saldati con superleghe in polvere, “Saldatura e taglio termico verso il 2000” IIS Meeting, October 3rd 1998, Milan-Italy, Wear 250 (2001) 611, Surface and Coatings Technology 106 (1998) 156, Wear 249 (2002) 846, Surface and Coatings Technology 92 (1997) 157, Wear 181–183 (1995) 810, Wear 225–229 (1999) 1114], rather than an optimisation of process parameters, also in critical geometric configurations. An experimental campaign has been carried out on the deposition of two nickel- and a cobalt-base superalloys: Hastelloy 276, Inconel 625 and Stellite 6. Benchmarks of C-Mn steel have been chosen to test geometrical configurations that are critical for the application (corners and grooves). Specimen characterisation through liquid penetration inspection, optical- and scanning electron microscopy and microhardness tests proved that process parameters optimisation depends only on the geometrical configuration and not on the deposited alloy. This result suggests the importance of an accurate design of the reciprocal positioning and movement between the torch and the part to be coated, as a function of the deposition geometry.

Introduction

In Plasma Transferred Arc (PTA) deposition of metal powders plasma is set up within the torch between the tungsten electrode and the nozzle, through a low-current pilot arc; then the arc transfers from the orifice to the work piece, becoming part of the electrical circuit, as shown in Fig. 1. Metal powder is fed into the collimated plasma beam, ensuring high energy-concentration: thus narrow and deep welds can be obtained [1], [2], [3], [4], [5], [6], [7], [8]. The process is characterised by extremely high temperatures (up to 20,000–30,000 °C [1]), excellent arc stability, low thermal distortion of the part, high welding speeds (torch/part relative speed: 2–16 mm/s) [1], [2], [3], [4], [5], [6], [7], low environmental impact (low oxides emissions, low percentage of ultra-fine powder) [6]. Plasma high thermal intensity allows high deposition rates and single layer deposits, with evident time savings.

Among the wide choice of metal powders for PTA deposition, Ni- and Co-base alloys are commonly used for high wear- and corrosion-resistant coatings, able to increase a component service life and ensure high temperature performances [3], [4], [7], [13], [14], [15]. Powders are obtained through protective atmosphere atomisation, ensuring a narrow grain size distribution and absence of oxidation, which could cause porosity in the deposited layer [6]. The spherical geometry obtained allows regularity of alimentation, whereas the negligible ultra-fine percentage of the powders, with grains dimensions between 50 and 150 μm [10], ensures easier manipulation with respect to other deposition techniques (oxyfuel gas welding, plasma spray, etc.) as shown in Table 1.

PTA high energetic efficiency, combined with very low dilution levels, yields to single-layer thick coatings that exhibit microstructural properties characteristic of a bulk material, independent of the substrate properties [2], [3], [9], [12]. Dilution, that is the percentage of the base material melted into the coating, is within 5–10% [3]. Additional characteristics of the deposits are a fine microstructure due to high cooling rates [2], [11], excellent shape constancy of the weld, and high adhesion between coating and support.

The industrial applications of these high-performance coatings are quite wide: moulds for glass and ceramics; automotive valves; chemical and petrol-chemical valves; lamination cylinders; plastics extrusion screws and dies. The deposits can be obtained with completely automated unmanned machines. The strategic potential of the PTA process is very little sustained by literature data useful to outline its technological limits. In particular, most of reference studies regard wear- and corrosion performances of the coatings, but do not consider the optimisation of the involved parameters, as a function of the deposition geometry. So, in accordance with the belief that process optimisation is as determinant to the coating properties as the alloys selection [3], a critical benchmark geometry has been chosen to determine process parameters influence on the quality of the deposited coating.

Section snippets

Experimental plan

To achieve PTA process parameters optimisation, as a function of alloy and geometrical configuration, a benchmark has been chosen of laminated ASTM A105/97 (Table 2), low C steel commonly used to be coated and welded. The geometry of the two deposition areas: a 90° corner and a narrow low-spoiled groove shown in Fig. 2, is critical for the application. Process parameters must ensure melting and good adhesion even in the vertex, avoiding at the same time high dilution and cracks or flaws due to

Results and discussion

Inspection and non-destructive tests were used to carry out process parameters optimisation. In the groove deposition of Stellite 6, for the groups 1 and 2 of tests, an immediate visual and liquid penetration inspection showed lack of adhesion between coating and support, so a higher plasma gas flux was adopted with respect to the Ni-base alloys, to obtain higher penetration. Poor adhesion could also be noted in the corner, for the first group of samples, as shown in Fig. 3. The nozzle diameter

Conclusions

A procedure was determined for deposition experimental analysis, which lead to process parameters identification and optimisation.

The influence of PTA as a thermal treatment on the base material was analysed. It was proved that structural modifications, with the generation of three different zones (melted zone, HAZ, unaffected zone), depend most on the physical effect of heat transfer and almost not on the alloys. In effect, the same microstructure could be observed in the base material

Acknowledgements

The authors would like to acknowledge Commersald S.P.A., in Modena, for materials and machines to carry out the depositions, and Prof. Ercole Soragni, Dept. of Materials Engineering, Modena, for help with the electro-chemical etching.

References (15)

  • R.L. Deuis et al.

    Scripta Materialia

    (1997 (15 September))
  • A.S.C.M. d'Oliveira et al.

    Applied Surface Science

    (2002 (30 November))
  • L.J. Yang et al.

    Surface and Coatings Technology

    (1995 (March))
  • E. Bourithis et al.

    Journal of Materials Processing Technology

    (2002)
  • J.N. Aoh et al.

    Wear

    (2001)
  • W. Xibao et al.

    Surface and Coatings Technology

    (1998)
  • R.L. Deuis et al.

    Composites Science and Technology

    (1998)
There are more references available in the full text version of this article.

Cited by (72)

View all citing articles on Scopus
View full text