One recent day a series of engine data from a Piper Turbo Saratoga showed up for analysis as part of a prebuy examination. With regular (non-prebuy) analysis we’re looking for things that need fixing and in the case of a pre-annual test profile we’re looking for things that need adjusting to get the engine close to spec. With a prebuy of course we’re still looking for anomalies but also for trends and showstoppers – things that would materially impact the purchase price. Three years ago I wrote a Puzzler about analysis for prebuy exams.
The data from this Saratoga caused quite a ruckus among the analysis staff. It has some of the highest CHTs we’ve ever seen in 14 years of looking at engine data. The test pilot references were obvious – from Clark Gable’s Test Pilot from 1938 to The Right Stuff from 1983. Don’t think for a second that by showing this Savvy is reconsidering its recommended CHT redlines of 400º for Continentals and 420º for Lycomings. We’re not. We know ours are lower than manufacturers recommendations and we remain convinced that lower temps accrue to increased longevity. But we’re happy to report that this engine didn’t explode, notwithstanding some exotic powerplant management.
This month we’ll also look at data from a Cessna 172, a Cessna 340. a Cirrus SR22 and a Piper Lance.
We’ll start with the Saratoga, powered by a Lycoming TIO-540 and data from a Gamin G3X. Our data went back 52 flights. Setting the HI CHT alert at 400º flagged almost every flight. Setting it at 430º still flagged almost every flight. Setting it at 450º still flagged a lot of flights. Here’s a typical flight.
EGTs in the 1500s are about right. FF over 18 GPH isn’t skimpy, but not enough for this engine with its presumably poor cooling. It’s not just one jug or one side. Even the coolest cylinders are around 430º. This is pretty typical of the data we have. But there are a couple of noteworthy flights.
Yep that’s 670º for CHT 3. And not just a spike. It stayed above Lycoming’s 500º redline for about 45 minutes. EGT 3 is lower there which argues against a failing CHT 3 probe. There’s a detonation event on this flight.
CHT 3 doesn’t spike and it “only” reaches 480º, but along with the wobbly trace of EGT 3 is the classic signature of a detonation event.
Maybe the buyer isn’t fazed by any of this and has plans for how to deal with this engine. But it certainly impacts the value of the airplane and ought to be part of the negotiation.
And kudos to those CHT probes for hanging in there way past what they signed up for.
Here’s a Cessna 172 powered by a Lycoming O-360 with data from a Garmin G3X.
CHT spread on the ground is amazingly low, indicating uniform cooling. The EGT and CHT spreads widen at max power but still ok. The reported issue was oil pressure so let’s isolate that.
Is this a sensor failure or a loose connection? It happens around the time the plane would be leveling off after takeoff, so that could be either. It’s hugging the zero line at the 22 min mark, but dips below the zero line twice. We know there can’t be negative oil pressure. It always makes sense to check the connection first but in this case, it looks like a sensor failure because of the negative numbers.
Here’s a Cessna 340 powered by a RAM Continental TSIO-520 with data from a Garmin 600Txi.
Lots of interesting patterns in the CHTs while the EGTs are stable and uniform. That indicates some uneven cooling. But that’s not what got the pilot’s attention. Here’s R engine oil pressure from the same flight.
This is deja vu all over again. Just like we saw in the 172. Big spikes and negative numbers. Same conclusion.
Here’s a Cirrus SR22 powered by a Continental IO-550 with data from an Avidyne.
CHT 6 is the problem. It starts with the first sample at engine start and just kind of runs parallel except for 1) a brief moment in initial climb where it joins the others, and 2) a series of spikes to Avidyne’s limit of 500 and to zero. Avidyne probes fail high. Again, it makes sense to check the connection first, but this also looks like a sensor failure.
Not that you’ll get a vote when it really happens, but if you have a sensor failure or a loose connection would you prefer it dropped to zero and stayed there, or tracked high like this one — just enough to distract you.
Here’s a Piper Lance II powered by a Lycoming IO-540 with data from a JPI 830.
The concern is those peaks in CHT 3. Let’s zoom in to where the action is.
There’s a strong inverse correlation between CHT 3 and EGT 3 that rules out a problem with either sensor. This is combustion. It’s a series of light pre-ignition events that goes on for almost an hour. And when it’s over at 01:30 the flight continues for another two and a half hours.
We recommended borescoping the cylinder and our client reported all was good. We would not have been surprised to hear about a cracked electrode. He said no, but he did change the top plugs from massive to fine wire plugs.
Our engines were designed to be strong and sturdy and sometimes the data shows us that they can be stronger and studier than they’ve been tested for.