I had noted over a few recent cross country flights that the indicated airspeeds during cruise seemed to be a few knots slower than I had noted when doing cruise performance testing shortly after the aircraft came out of the paint shop. I attributed the observation to the effect of higher temperatures now, and higher weights with passenger. But, I was still curious, so on the last leg into Smiths Falls on the way back from Wisconsin I recorded one test point of cruise performance speed vs power data, and then reduced the data to standard conditions (sea level, 15 deg C temperature, 1800 lb gross weight) so I could compare it to the results from that earlier testing. I found that the current performance was 4 to 5 kt slower than the average of all the earlier testing. Hmm.

After landing, Terry helped me clean the bugs off the leading edges, cowling and windshield. I had been religiously cleaning off the bugs after every flight, as they come off easily when they are fresh, but stick like glue if you let them dry for a few days. But, I hadn’t ever cleaned any other parts of the aircraft, as they never looked dirty. But, Terry ran the cleaning cloth over the top of the wing, and she noted that the surface was a bit sticky and covered in fine grit. This was only on the upper surfaces, so it must be stuff that settled from the air. We wiped down the upper surfaces of the wings, tail, fuselage and cowling to remove the accumulated grit.

Friday afternoon I took a half day off and did a bunch of little maintenance, including leak checks of the pitot and static system - both were completely leak-free, which means that this wasn’t an explanation for the apparent low indicated airspeed.

Saturday I changed the oil and filter, but I had to go flying first to warm the oil up. Still puzzled by the low cruise speeds, I observed the manifold pressure during takeoff (it was exactly where it should be, which rules out a dirty air filter as a cause of low engine power). I recorded speed vs power at three different power settings, did two climb performance tests (rate of climb is good measure of power) and one test point of airspeed error. After landing I changed the oil and filter, checked the cylinder compression and checked the ignition timing. The compressions are slightly lower than I would hope, but they are well within the normal range. The electronic ignition timing was perfect, and the magneto timing was slightly off, which I corrected, but this would not have affected engine power as the electronic ignition would be firing well before the mag at typical cruise conditions.

I cut open the removed oil filter, and found nothing of concern inside. There were a very few extremely tiny specks of metal, but nothing like you would see if something was wearing quickly inside the engine (like a camshaft, which would explain low power).

Saturday evening I crunched all the data. One cruise test came in at 2.0 kt slower than the tests from last year, one was 0.4 kt faster and the last was 1.6 kt faster, for an average difference from last year of 0.0 kt (these test point to test point variations are in the scatter band from the previous testing). So, now the aircraft is just as fast as it was a year ago and the only thing I did was wash off the accumulated grit. That grit is powerful stuff! I obviously need to wash the aircraft more often. Sunday afternoon, after getting back from the Arnprior Fly-In Breakfast I did a complete aircraft wash.

I also compared the climb performance against earlier data. Once I accounted for the different weight and temperature the latest results match reasonably closely to the earlier results.

The airspeed error (+0.9 kt at 139 kt IAS) fits exactly on the line of the three other airspeed error tests I did in May 2010 after the aircraft was painted. It is interesting that this point confirms the earlier indications that the static system error was slightly affected by the thickness of the paint.