Flight 2 was humbling. I’d walked away from the first flight feeling cautiously optimistic, and I went into Flight 2 expecting to build on that. Instead, it turned into a shorter, more stressful flight than I’d hoped—and it taught me some things I hadn’t anticipated. I’m writing this partly to document what happened and partly because I suspect some of you have run into similar situations. If you have, I’d genuinely love to hear how you handled it.
The flight took place on April 17, about a week after the first flight. Duration was roughly 28 minutes, and I burned 9.5 gallons out of the left tank—a number confirmed when the fuel truck put exactly 9.5 gallons back in afterward, which was a nice validation of the fuel totalizer. The right tank needed just 0.2 gallons, which appeared to be thermal expansion loss. So the totalizer seems to be tracking accurately, which is reassuring.
East Side, Lower Altitude
The first curveball came before I even took off. On Flight 1, we’d been given the west side of the field at 1,800 feet MSL—a comfortable racetrack that kept me away from most of the traffic flow. For Flight 2, the controllers asked me to stay on the east side of the field and would only approve 1,400 feet MSL instead of 1,800.
I get it—they have traffic to manage and can’t always accommodate an experimental doing racetrack patterns wherever it wants. But from a second-flight test perspective, the east side at 1,400 feet gave me a noticeably tighter box to work in. Less altitude, less maneuvering room, and a busier environment. Not ideal, but workable.
There was also a practical cost to working out the airspace coordination on the ground: time. The back-and-forth with the tower while sitting at the hold-short added up, and so did the waits for arriving and departing traffic before I could get onto the runway. By the time I was actually airborne, the engine had been running and sitting far longer than I would have liked—giving the CHTs a head start on warming up before I even had the cooling airflow of flight.
The flight map above shows the racetrack pattern east of KHEF. You can see from the altitude trace at the bottom that I stayed in the 1,200–1,600 foot MSL range throughout—tighter than I would have liked, but manageable once the tower and I were on the same page.
CHTs: Elevated at Rotation, Then Trending Down
Cylinder head temperatures were something I watched closely throughout. The chart below shows the full picture.
The data-logging system flagged this flight with “CHT > 435°F on takeoff for 3 min,” which captures it well. Starting from the ground roll, all six cylinders climbed steeply. By the time I was rotating and in the initial climb, every cylinder had crossed the 435°F red line. The peak cluster—clearly visible in the circled area on the chart—shows the hottest cylinders reaching somewhere in the 460–475°F range, with the leading cylinders at the upper end of that band and CHT4 running a bit cooler than the rest. The elevated temperatures persisted for roughly three minutes before beginning to come back down as the engine found its footing at cruise power.
It was stressful in the moment. Seeing most of your cylinders above redline on a brand new airplane, in a tight pattern at 1,400 feet, is not a comfortable place to be. That said, the temperatures did trend downward on their own, and by the time I reached the cruise portion of the flight, the CHTs had settled back into a more reasonable range—roughly 350–400°F depending on the cylinder.
I’m honestly not certain how much of this is expected break-in behavior versus something worth worrying about. My read from the community is that high CHTs during the initial climb on a new engine—before the rings have seated—are fairly normal. But I’d genuinely welcome input from others who have data from their own early flights. What did your CHT peaks look like at this stage, and how quickly did they improve?
Oil Pressure: Stable and Reassuring
Oil pressure was one of the few systems that gave me no reason for concern. The chart below shows what it looked like across the full flight.
During taxi, pressure fluctuated between roughly 50 and 80 psi—normal variation at low RPM. Once I was airborne and at cruise power, it settled into a steady 75–80 psi and stayed there. That was the green band, and it held throughout the flight. Amid everything else that was happening, seeing that line stay flat and calm was genuinely helpful.
The Oil Temperature Problem
This is the main story of Flight 2, and I want to tell it honestly because the judgment calls involved are the kind of thing that’s hard to describe from the outside.
Before either Flight 1 or Flight 2, I had noticed some finickiness in the wiring near the oil temperature probe in the engine compartment. When I moved those wires during preflight work, the oil temp gauge in the cockpit would give erratic readings. I couldn’t reproduce it consistently before either flight, so I logged it as a known issue to investigate and flew anyway. In hindsight, I should have thought harder about what my abort criteria would be if that problem appeared in the air—because it did.
The oil temperature chart tells the story. During taxi, oil temperature climbed gradually from a cold start of around 80°F, rising through the 150s, 170s, and into the 200s by the time I reached the runway—a normal warm-up curve. At takeoff, the real underlying oil temperature was somewhere around 210–215°F, which is in the upper end of the green band and nudging into the yellow caution zone. That in itself was worth noting—warmer than I’d seen in the previous flight, probably a reflection of taking off later in the day.
But the story isn’t the underlying temperature. The story is those spikes.
Three times during the flight, the oil temperature gauge went erratic and shot off-scale high—the near-vertical lines you can see reaching 325–350°F and beyond on the chart. Those aren’t real readings. Oil temperature cannot physically rise at that rate; a genuine spike that steep would mean something catastrophic was happening, not an intermittent wiring fault. I knew intellectually that these were false readings. But there’s a difference between knowing that and feeling okay about it when you’re flying a new experimental airplane in a tight pattern at 1,400 feet.
After the first spike, I kept flying and watched. After the second, I started thinking seriously about returning. After the third—when the gauge went off-scale high and then simply went dark, leaving me with no oil temperature indication at all for a minute or two—I made the call to land early and get it sorted out on the ground.
It’s worth being honest about why that decision felt as significant as it did. In isolation, an erratic gauge on an otherwise healthy engine is a manageable problem. But Flight 2 had been accumulating stressors from the start: a tighter-than-preferred practice area, a lower altitude than I’d wanted, CHTs above redline on takeoff, and now a primary engine parameter going intermittently dark. None of those things individually would necessarily have ended the flight. Together, they made landing early feel like the obvious right call. I wasn’t going to keep flying in a compressed pattern at 1,400 feet with unreliable oil temperature data on a low-time new engine.
I think it was the right decision. But I’d genuinely welcome input from others on how they think through these accumulating-stressor situations. What’s your personal threshold? I don’t always have a clean answer to that.
What Was Actually Wrong
After landing, I went looking for the root cause. In the firewall-forward area, the oil temperature probe wiring is spliced using crimp connectors—four of them joining two wires together. One of those four crimps had not adequately grabbed the wire. I was able to pull it free by hand once I’d unwrapped the bundle.
That was the whole problem. One poorly seated crimp. We re-did it, and the oil temperature gauge worked without issue on every subsequent flight.
The lesson I took from it: a visual inspection isn’t always enough to catch a bad crimp, and an intermittent behavior that shows up on the bench needs to be resolved before flight—not logged and hoped away. I knew about this issue before I flew and didn’t define clear abort criteria for it. I won’t make that mistake again.
A Few Other Notes
PFD-1 (the AHRS-1 attitude indicator) tumbled again on the takeoff roll, consistent with what happened on Flight 1. It seems to correlate with the application of forward acceleration—likely some vibratory effect on the AHRS unit. It’s on the list to sort out, but it’s lower priority than flight-safety-critical items for now. PFD-2 and the G5 standby remained solid throughout.
The Insta360 X5 camera I’ve been using for documentation ran out of battery before the takeoff roll on this flight, so there’s no video to share. I’ll be more deliberate about charging it before future sessions.
What I Took Away
Flight 2 was short and more stressful than I’d planned. But the outcome was fine—nothing broke, I made a conservative decision to land when my instrumentation became unreliable, and we found and fixed the actual problem before the next flight. That’s the process working as it should.
A few things I’m carrying forward:
- Known issues need abort criteria before departure. If something is behaving oddly on the ground, decide in advance what you’ll do if it shows up in the air. Don’t leave that decision for the moment.
- Airspace coordination is worth doing ahead of time—and worth holding firm on. Getting assigned the east side at 1,400 feet added unnecessary pressure to an already-demanding flight. For subsequent flights, I’ve made a point to coordinate specifically for the west side of the field. Until I have full confidence in the aircraft and it’s ready to venture further outside the Class Delta for the remaining flight test program, having the more open, higher-altitude practice area on the west side is genuinely important—not just a preference. I’d encourage any experimental builder doing early Phase 1 testing at a busy Class D airport to have that conversation with the tower in advance, and be clear about what you need and why.
- The fuel totalizer appears accurate. Having the refueled quantity match the totalizer reading exactly was a genuinely useful data point—I’m more confident in that system now.
As always, if you’ve been through something similar—erratic instrumentation on an early test flight, a wiring issue that surfaced at an inconvenient time, or a tricky judgment call about when to land—I’d really like to hear about it in the comments. I don’t have all the answers on this airplane yet, and the conversations here have been more useful than I expected.
Flight 3 is up next.
jamessclarkiv.com 