This has been an expensive day. I had flown a climb performance test yesterday, and another airspeed calibration test this morning. Everything was going well, and I was hoping to get to the 25 hour point by the end of next weekend, weather permitting. This afternoon I strapped 75 lb ballast down in the aft baggage compartment to move the CG back to the middle of the envelope, and took off to fly a series of tests that I will repeat at successively more aft CGs as I work back to the aft limit.

Many of the points in my test program are pulled from the FAR 23 requirements for single-engine aircraft - they are the same tests that would be done during the type certification flight testing on a production aircraft. I added the FAR 23 test points as I was curious how the aircraft matched up.

One of the tests on the card was a pitch trim check with idle power at VNE, as called for in FAR 23.161(c)(3). I didn’t expect that there was any risk associated with this test, so I didn’t do any sort of build up test points. I pushed over to VNE, then smoothly brought the throttle towards idle. As I got back to about one-third throttle all hell broke loose. The rpm ran away, increasing rapidly through 3500 rpm. My instinctive reaction was to undo the last action I had just taken, so I pushed the throttle forward a bit. The rpm continued to increase, and I saw over 3800 rpm before it decreased back into the normal range.

I stopped the testing, and headed back towards Smiths Falls. The engine showed no signs of distress, but it was clear that the smart thing to do was to land ASAP. I looked at the recorded data, and it shows that shortly after I brought the throttle back the oil pressure started to drop sharply, indicating as low as 22 psi (but, there is almost certainly some air in the line from the engine to the oil pressure transducer, which would dampen the oil pressure indication, so the actual pressure may have been lower), and obviously the prop governor stopped getting the oil it needed to control the prop. The prop will go to low pitch if the oil supply stops, and low pitch and a bit of power at VNE led to a very high rpm. The data shows the max rpm as 3991. Ouch.

Now I need to figure out what to do next. It is clear that all rotating or reciprocating components would have experienced very high stresses. It is possible that no damage was done. It is also possible that the high stresses could have started a crack in a critical component, leading to a catastrophic engine or prop failure down the road. I looked at my Hartzell Propeller Operators Manual, and it calls for an inspection by an overhaul shop if the prop experiences a speed of more than 110% the normal limit. The normal limit is 2700 rpm, so the prop saw 148% of the limit. If the prop sheds one blade, the out of balance forces usually cause the engine mount to fail, and once the engine departs the aircraft the CG is so far aft that control is usually lost, resulting in a fatal accident. I'm not willing to risk that, so I’ll either send the prop to an overhaul shop for inspection, or buy a new one. I’ll look into the option of getting an “aerobatic” prop, which goes to high pitch if the oil pressure drops. Unfortunately, that would require a different spinner, as apparently the counterweights at the blade root won't fit inside Van's spinner. Hopefully the base diameter of the spinner isn't too much larger, or I may have to do some cowling mods too.

I’ll call Bart Lalonde at AeroSport Power to get his take on the engine aspects, but I expect I’ll pull the engine and have him open it up for inspection. I may replace things like connecting rods, pistons, etc.

I was hoping to finish off the mandatory 25 hour test program next weekend. But, now I have grounded the aircraft after 12 flights and 17.4 hours. I expect there will be a couple month hiatus in the test program. I’ll post more when I have a plan.

In retrospect, I should have paid more attention to a small drop in oil pressure that I had seen during a power-on dive on an earlier flight. That was a sign that the oil was moving towards the front of the oil sump, away from the oil pickup. After that flight I added some oil, and repeated the condition with negligible effect on the oil pressure. But, I should have recognized that pulling the power to idle could cause the oil pressure to drop, and I should have done a work-up in speed on this test. I.e. do the test at a lower speed, and repeat at successively higher speeds. Also, I had already recognized that the propeller low pitch stop was set too fine, and I had made an adjustment before this flight. I shouldn’t have even tried this test until I was satisfied that the low pitch stop was correctly set (i.e., set so that the rpm is a bit less than 2700 during the initial part of the take-off run).

I’m glad this event didn’t occur when I was on the road. If you are going into a large airport, it is not uncommon for ATC to request a high speed until about five miles on final, to facilitate approach sequencing with the airliners. I can envision that I could be coming down the ILS at 180 KIAS, then pulling the power back a long ways to slow down.