The ability to select process parameters that provide a stable keyhole weld pool is an important requirement for a greater utilisation of plasma arc welding process as a low cost keyhole welding process.  This is because plasma arc welding has probably the largest number of variables of the fusion welding family of processes and the plasma arc keyhole weld pool is stable only within a small range of parameters.  The stable operating range is often presented in the literature as a two-dimensional parameter envelope for two of the following three primary parameters: traverse speed, current and plasma gas flow rate.   However, the size and shape of the parameter envelope is known to change with torch, gas, substrate and/or positional variables as such changes affect the transfer of heat to the substrate and the stability of the weld pool.

This paper will describe the experimental methodology underlying the construction of parameter envelopes in terms of the weld pool response, conduction in the substrate and weld imperfection generation.  Based on this analysis, results will be presented on an efficient characterisation of the parameter envelope.  Experimentally determined geometric features of a stable keyhole weld pool are correlated with the parametric envelope such that a criterion for stable weld pool behaviour is derived and advanced.   The results of the application of this method will be illustrated with respect to autogenous welds in Ti6Al4V and discussed in light of structural, material and manufacturing considerations.