| Home | Articles | CP reprints |

SOME THOUGHTS ON COOLING
(From CP42, Page 3, October, 1984)

     The following observations are based entirely on my own experiences over 900 hours of flying in my Long-EZ, N26MS.
     Since first flight in January of 1981, my engine, a Lycoming O-235-L2C, baffled exactly per section IIL of the Long-EZ plans, has run with cylinder head temperatures that were not even, to say the least. Cylinder #4 (forward right side) had always run the hottest. Cylinder #1 (aft left side) had always run the coolest. At normal cruise power in level flight there was normally a disparity of up to 100F between these two cylinders. Cylinders #2 and #3 ran within 5 of each other at all times (#2 is aft right side and #3 is forward left side). This problem was not entirely as bad as it sounds in that even the hottest cylinder #4 never did exceed or in fact even come close to the red line temperature as called out by Avco Lycoming. (500F as measured at the bayonet fitting under the cylinder head, not at the base of a spark plug).
     On an average cross country trip, in stabilized level flight, my cylinders would run as follows: #1 - 280F, #2 - 345F, #3 - 341F, #4 - 413F. We flew the airplane for over 700 hours with this condition. Over this period I tried many small ideas in baffling, tightening the baffles, loosening the baffles, sealing every single little gap in the baffles. None of these changes made a really significant improvement.
     Sure I picked up 5 and 10 increments from time to time, but I was trying to cure a 100 difference! There is not enough space here to list all the different ideas I tried, but they included removing the intake duct completely, and installing a throttle body fuel injector instead of a carburetor.
     Finally I tried a small deflector plate, consisting of a piece of .032 aluminum approximately 4" x 5. I bent it so I could rivet it to the inside of the lower cowling lip. I mounted it off center on the right side, under the #4 cylinder, hoping perhaps it might aim the incoming, high velocity cooling air directly at the #4 cylinder. See sketch.
wpe1B.gif (13066 bytes)
     I test flew it, with little enthusiasm and was amazed to say the least. #4 was now the coldest cylinder. Unfortunately #2 was not too hot, however I now knew I was on to something. To make a long story short, I tried 6 different iterations of various width and height of deflectors in several positions on the lower cowling. Currently I have 3 deflectors, one on the left side, one on the right side, and one on the centerline, aft of the carburetor. See sketch.wpe1C.gif (3234 bytes)
     These deflectors are made of foam and glass (1 ply of BID over pour—in—place). My results at this time, with 150 hours of flight time on the above, maximum power climb to 12,000 feet, #3 cylinder is the hottest at 435F, #4 is the coolest at 395F. In level flight at 8500 feet, at an economy cruise power setting (approximately 60%) of 17" manifold pressure, 2500 RPM, indicating 121 knots for a true airspeed of 136 kt/157 mph COAT +1C), fuel flow was 4.4 gph. Cylinder #1 -345F, #2 - 31F, #3 - 338 F, #4 - 343F.
     On another trip at 8500 feet at maximum available power (approximately 75%) 21.4" manifold pressure, 2850 RPM, indicating 146 knots for a true airspeed of 163 knots (OAT +4C) fuel flow was 6.4 gph. Cylinder #1 - 363F, #2 - 361F, #3 - 364F, #4 - 363F.
     So it can be seen that with a little trial and error, the cooling air deflectors can be made to work rather well. I don’t claim that they will work for everyone, but I believe that if you are having similar cylinder head temperature disparities, it may be worth a try. I must emphasize that this test was done on a Long—EZ with a flush NACA cooling inlet and a Lycoming 0—235 engine.
     Whether or not this idea would work on an airplane with a standard rain inlet, I can’t say, but again, it may be worth a try. Remember that I have all four cylinders instrumented with CHT as well as EGT with a calibrated digital gauge, so I knew at all times what was happening with each change.