Michael Oritt stated, "If you have no outgoing water flow you may not get a reading on your exhaust temperature alarm." He could be right but I fail to understand why. I monitor the surface temperature of the exhaust elbow just below the point where the raw water enters the exhaust flow. At normal operating conditions with raw water flow I can put my hand there with no real discomfort. (I estimate the temp to be 100-120 degrees F.) Without raw water this shoots up to 180-200 F pretty quickly. I have surface mounted a normally open temperature sensor at this point that closes at 160 F, and sets off a bell, the rationlization being that this should alert me about as quick as anything if the raw water flow is obstucted or shut-off for some reason. My concern with a direct flow indicator, be it before the pump or after, is that should the hose come loose at the heat exchanger, a flow will still exist from the thru hull to the loose end of the hose. True that will eventually set off the bilge high water alarm (unless all those bilge pumps we have takes it out as fast as it comes in) but that could be precious minutes after the engine temp has started to rise. Also after the raw water leaves the heat exchanger part of it is routed, thru a hose, to the tranny oil cooler and another part, thru a hose, to the engine oil cooler. If one of these hoses burst or comes off will that divert the water flow to the exhaust elbow enough to allow the exhaust stream to rise in temperature? I don't know but I hope that by monitoring the exhaust elbow temp I will be able to quickly detect the effect of any reduction in water flow at that point regardless of what caused it. Just my philosophy. Others will vary. Bill