Replaced the primary SMPS IC and same poor DC supply . Determined it was not due to excessive load, shorted turns in the inverter Tx unlikely. Tried the secondary SMPS powered from bench ps and that brought up the internal Vref , requiring a supply of 10V , not 8.6V which is just below the undervoltage lock out. The Vref supplies the "thermal" LED opamps etc. Went back to the first SMPS and placed 4.7uF across the little 2.2uF SMD aluminium electrolytic that smoothed the sense voltage and the DC supply came up to 13.2V. So bad ESR cap again it looks like, falsely telling the IC that the Tx was overloading, rectified sense voltage was too high. The opto coupler was not for the primary SMPS, coupling is the other way round, probably to pass mains AC-ON confirmation, to lock out the 24V DC option, not used , as only mains use . Will tidy up cut traces and should hopefully be back into operation. About 2,500 GBP amp and as a not current model and long wire HV ducted cabling and transformered speakers etc, perhaps that again if a replacement 5 ch distribution amplifier was not compatible and all the other kit had to be replaced also