I have been spending a lot of time lately thinking through the cooling system and specifically the water pump. On a liquid cooled engine the water pump is a single point of failure. A pump failure means that you are coming down and sooner than you would like.
The other thing that needs to be considered when choosing an automotive engine for an aircraft is that a pulley driven water pump is optimized for a very specific RPM range and that is usually around 1000-2200 RPM. This is where the automotive engine spends most of its operational life.
You want a liquid cooled aircraft engine to flow as much coolant as possible at its cruise RPM and for the LS1 married to a Geared Drives psru that turns out to be about 3300 RPM. This is much higher than the optimized speed of the OEM water pump. If you turn that pump faster you end up simply wasting horsepower and not getting an equivalent increase in coolant flow.
There are a couple of ways around this problem. The first has been used in racing circles for years and that is to under drive the pump by changing its pulley size. This works well when running at cruise speed but unfortunately means reduced flow at idle. This doesn’t bode well for the long taxi on 104 deg. F afternoons in Texas.
The second is to install an electric motor driven water pump in lieu of the OEM water pump. Meziere makes a number of bolt on replacement pumps that have an idler pulley built in to maintain your existing belt geometry.
The electric pump accomplishes a couple of things that are positive for automotive engines in aircraft. First, it eliminates the need to physically drive the coolant pump using a belt. This will net you 15-20 horsepower at the propeller. Second is the smoothing out of coolant flow over the entire range of engine RPM. This means that you are getting optimum cooling flow when you are taxiing down to the end of the 12,000 foot runway as well as when you are throttle to the firewall.
There have been numerous instances of automotive engines over temping on the ground while taxiing and I have to wonder whether this inefficiency of the coolant pump at idle RPMs is a contributing factor.
The last bit I am going to talk about today is in regard to redundancy and failure modes as mentioned earlier. I am going to be installing dual electric coolant pumps in my RV-10. The main pump will be mounted to the engine in place of the stock unit. The aftermarket unit from Meziere duplicates all of the hose connections of the stock unit and maintains the belt path and operation. The secondary unit will be run in series with the main unit and will be mounted onto the radiator where the bottom rad hose bung would normally be located.
This will eliminate a major failure mode in the installation of a GM LS1 engine in my RV-10 and give me the peace of mind that I need for the future flights into IMC that I plan to make after getting my instrument rating.
Have a great time building and modifying!