Notes. Here 𝒰(a, b) is a uniform prior between a and b. 𝒩(μ, σ) is a normal distribution centred on μ with width σ. 𝒯𝒩(μ, σ, a, b) is a truncated normal distribution with the peak at μ, with width σ, truncated below a or above b. 𝒰^s(a, b) is a uniform distribution with softened edges, where the log prior is 1 − e^{(a − x)/0.01} if x < a, 1 − e^{(x − b)/0.01} if x > b, and zero otherwise. In this case, during model evaluation, the parameter was forced to the range of a to b. ^( * )Only the main cluster position was allowed to vary from the source detection position during the MCMC run. Other cluster positions were fitted for once, then fixed to these positions. If fitted using the peak position, these parameters were frozen. ^( * *)Normalisations defined at 1 keV assuming a power-law photon index of 1.7. ^( † )The best fitting value of θ₀ was used as θ_fit to reduce boundary issues when sampling. In the output catalogue, the value of θ₀ was reduced to the range about θ_fit = 0.