Before you extract the pump, lock the crankshaft to prevent rotation. The graphite bearing (closest to vane assembly) appeared to be decidedly worn and seemed to be passing water to the water flinger below. Subsequent examination showed the graphite seal to be in quite good shape?
Note order of rotor, graphite seal, water thrower, oil seal, ball bearing.
Note corrosion on shaft & water thrower indicating that graphite seal is passing water?
Close-up of the housing which is supposedly different from a 6-vane type
Close-up showing the jack-shaft (just). Note the brass housing remained in place. I am having trouble removing this so if anyone has any suggestions, they would be gratefully received! email.
Quick update: after posting the problem on the Discussion Forum on TriumphStag.net, I got a reply within an hour or so suggesting the use of a slide hammer. Alan very kindly provided a proto-type which worked a treat and fully exposed the jack-shaft.
Any water that passes the graphite seal is thrown out via the brass housing and the slot in the crankcase to the top of the Vee:
The slot is demonstated by a 12" ruler which can be passed right through when the brass housing is removed.
In my case, I think most of my water loss was from the inlet manifold waterways, since the graphite seal appeared to be in good shape. Anti-freeze was not apparent in the vicinity of the escape slot.
Direction of water flow: the Workshop Manual indicates (page 82.00.03)that water is drawn from the rear of the LH cylinder head into the heater, and returns to the top (suction) side of the pump via the metal pipe and pump housing. The heater return mingles with water returning from the radiator via the bottom hose and water via the bypass hose when the thermostat is closed.
When the thermostat is open, the pump forces most of the hot water around the engine and through the inlet manifold to the radiator via the top hose.
This diagram is a pre-cursor to my attempt to discover if gasket hole alignment is critical for even temperature distribution. Apparently, it is not, but I have yet to pinpoint the reason for my occasional overheating.
