My understanding and educated guesswork is as follows...mikeonb4c wrote:Out of interest, what warning pattern would you expect to see as coolant level drops due to a pinhole leak in a hose, or maybe poor circulation due to blocked rad, failing water pump etc? Would it be easy for the average owner to understand and act on?Driver+Passengers wrote:Totally off topic for this thread, but +1 for monitoring pressure first (and level/temp, if so desired).
As coolant level drops due to a pinhole leak in a hose (over time, typically many days/weeks) there will be less coolant to expand and the typical running pressure when the engine is up to temp will be lower. I'm currently trying to quantify this somewhat by varying the initial/cold coolant level in my tank.
Air settled in the head/matrices would thermally expand, causing pressure rise, high-level coolant alarm to trigger and temperature increase.
Poor flow will result in increased temperatures and an increase in temp/pressure will be observed. If localised boiling occurs around the exhaust ports, you would observe a spike in pressure, and a high coolant alarm would likely react, too. Only then after coolant expulsion would the low level alarm trigger.
Catastrophic failure of a hose would dump all pressure, trigger low level alarm and shortly give rise to temp alarm.
With a leaky gasket or cracked head, pressure and temperature may increase unduly with load, or even no load. Only after coolant expulsion will a low level alarm trigger. High level alarm would help in this case. Alternatively, level may drop and you're back to the same symptoms as a pin-hole leak.
Alarming on coolant pressure conditions requires rise/fall alarms, not just absolute thresholds, implying R&D, untested kit, etc... Monitoring pressure is not "install and forget". Low level (and high level) coolant alarms are much easier for the average owner to understand and act on. Also, pressure is unlikely to tell you on start up if the header tank is empty, this is where low level alarm definitely wins out.
I believe there is going to be a delay in almost every situation before a low coolant alarm triggers. Similarly, as has been well thrashed over the years, the position of a temperature sensor on the head is going to give readings that mean different slightly things. Typically, all rely on heat transfer through metal and so there will again be a delay.
That's why, with the exception of a cold start with an empty header, I rely on pressure to tell me what's what. I have not yet installed a low (or high) coolant alarm. I haven't even replaced my digital temp gauge(s), though I have modified the dash gauge. My alarm conditions for pressure are still immature (<2psi, >15psi, >3psi/s, <-3psi/s).
I'd be happy to stand corrected on any of this.
