After talking through our difficulties with high oil temps, we determined there were likely three contributing factors. First, the nose wheel was not retracted for first flight. We learned this disrupts the airflow to the naca scoop and thus can cause issues with cooling. Second, the prop was left at high rpm, thus causing the engine to generate more heat from the faster rotation speed. Third, there were a number of gaps that needed to be sealed between the baffles and the engine. We took care of the gaps, but would have to return to the air to test it out. On September 5th, 2011 we tried to do just that. Unfortunately when the boost fuel pump was turned on before take-off the engine coughed and sputtered a bit, even though fuel pressure only went up by about 3psi. After playing with idle fuel mixture settings, and installing fresh spark plugs, we still couldn’t get it to run smoothly with the boost pump on. Eventually it wouldn’t even idle correctly without the boost pump on. So we took the fuel injection servo off and sent it in for rebuild, accepting we’d be down for about 6 weeks.
We got the servo back about Thanksgiving, and promptly reinstalled it. Much to our disappointment though, the engine would still not idle below about 1100rpm. The only good news was that it didn’t get any worse with the boost pump on or off. We ran through the Lycoming troubleshooting manual, checking for induction leaks, proper idle fuel flow, etc… Nothing seemed to indicate a problem. The issue was the same no matter if we ran on the left ignition, right ignition, or both. After many calls to the ignition vendor (ElectroAir), we finally determined that the system was having trouble reading the crank angle sensors. We replaced one of the sensors as it seemed to indicate that one of the was completely failed, but that didn’t really solve the problem either.
After an exhaustive look at the wiring to the ignition control units we discovered that the “Key Switch Upgrade” we opted for to allow the systems to function with a normal aircraft (P-Lead style) ignition switch, was not an ideal design. The keyswitch lead that is provided by ElectroAir is a tap off the crank sensor signal wire. The wire then gets ground and the system fails to see that the engine is turning, and thus doesn’t fire. The problem is that this make the signal wire very long in a pusher airplane, and very susceptible to interference. We redesigned the circuit to instead trigger relays that would kill power to the control modules, as the main power feeds are not as sensitive to noise.
After this change, the engine was once again running extremely smoothly. It hadn’t run THAT smooth since it was on the test stand at Bolduc. So by this time it’s now February 2012, and 7 months have passed since 4TF was last in the air. Not wanting to jump right back into a canard without having recent experience, John opted to wait until he could get some additional transition training in another Cozy. After spending a couple of hours with Bob Bittner in his Cozy, John was ready once again fly 4TF, and on June 12th, 2012 he got his chance. John’s Narrative to the Cozy Mailing list is below:
I arrived at the airport about 5:30pm and began inspecting the plane. Winds were getting calmer and calmer. When I finished the preflight, fueling, and W&B, winds were down to about 5 kts. My friend Rob Jahnke arrived to assist my wife with ground crew / spotter duties, and I went through my ground crew emergency procedures with him (electrical bus switches, battery locations, fuel valve, latch operation, fire extinguisher locations, etc.), then saddled up.
In my nervousness I failed to turn the fuel selector back on after showing Rob how to turn it off. I started the engine, and about 8 seconds later it promptly died. This made me pause to think if I was really in a good frame of mind for conducting the flight. My nervousness was just as bad as when I did first flight. Ultimately I took a deep breath, and resolved to working my checklist and flight test card letter for letter. Plan the flight….Fly the plan.
I taxied out to 31, the runup was good, so I advised the controller of my intent to circle the field at 5000′. Winds had dropped to 4kts, and I was cleared for take off.
4TF pushed strongly down the runway, but I resisted the urge to rotate too early, as I fear that may have contributed to my PIO from first flight. I wait to bring the nose up till about 80, but only lifted just enough to get the wheel off. I held there for a brief moment, but the mains didn’t want to get left behind, so up we went. In no time she added 4000′ underneath me and I leveled at 5000′. Oil temps had hit about 220 at this point, and I pulled the prop control to bring the rpm down to 2300. I had to reduce throttle a lot to maintain speed below 140kts. 13-14″mp and 2300rpm yielded 125-130kts at 5000′. Oil temps stabilized at 200, I was hoping to see it below that, but it’s far better than the 240 peak I saw on first flight. CHT peaked at about 395 during climb out, but dropped to 350-360 in cruise (though this was pretty slow cruising). The engine ran strong the whole time, and I was able to work through my entire flight test card. When I got to the end of the card (about 35 minutes into the flight), I really didn’t want to come down yet. The engine was running perfectly, the plane was flying so nice, so I just circled around a couple more times and played with the throttle and flight controls a little more. Throughout the flight I was active on tower frequency, but monitoring my ground crew on 123.125. They reported all looked well from their angle. No smoke or flames or dangling objects.
I informed the tower I would be descending back into the pattern and began my descent. My first approach to the runway I was high and fast, and I could see there was no way I’d be able to bleed it off fast enough, so I just went around. The second approached was looking really good, but as I got close to the runway the speed was decaying faster than I wanted. In hindsight it was probably just fine, but I just got a little nervous. At the last second I mashed the throttle and went around….again. The third approach looked good as well, and this time I didn’t panic at the end. According to the datalog the wheels touched at 80.1Kts. My butt-o-meter said they touched ‘nicely’. Another 57 minutes in the log book.
