Airworthiness Directives. Ask an aircraft owner what’s their biggest worry and it’s probably an A.D.. It’s right up there with structural icing and a fire in flight. Even there you’ve got an edge. Good flight planning probably keeps you away from structural icing and great maintenance probably lessens your chance of a fire in flight. But an A.D. is kind of the luck of the draw. Some airplane somewhere had a problem, then someone noticed it was more than one, then it became a trend, then the whole fleet is presumed guilty until proven innocent. But the FAA gave us something that most government agencies don’t have – Alternate Methods of Compliance.
Quick review – An AD contains the required method for resolving an unsafe condition in an aircraft, aircraft engine, propeller, or appliance. An AMOC provides an acceptable level of safety for a different way, other than the one specified in the AD, to address the unsafe condition. What’s all this talk about A.D.s? Did the FAA just issue a new one for my Bellanca Viking? No, knock wood. It came up in a recent hangar flying session and we wished that some of life’s least appealing scenarios came with AMOCs. For us it was AMOCs for your neighbor’s daughter’s piano recital, the tasting party for your brother-in-law’s new blueberry jasmine homebrew (it was as bad as it sounds), and your company’s endless series of mandatory HR reviews. And what’s the connection to engine data? None that I can think of. So let’s get to it and look at data from a Cirrus SR22, a Diamond DA40, and a Cessna 172XP.
This Cirrus SR22 is powered by a Continental IO-550 with data from a Garmin Perspective with a one second sample rate. Here are EGT’s, CHT’s and FF from a recent flight. Cursor is off to the left to keep it out of the way.
FF looks ok. CHTs look ok except 3 is low during taxi-out, and that’s where EGT 3 is also low and jumpy. Then both are low again after landing. Bottom line – there are a few adjustments to be made but the data appears reliable. On the next flight we got this.
All 6 EGTs bug out for about an hour, beginning after the big mixture pull for cruise – at about 20 minutes. They had done this earlier during taxi-out at about 3 minutes. And EGT 5 has its own issues, posting down spikes at 7 minutes and 70 minutes later. Both of those are timed with a FF reduction. EGTs dropout but CHTs and FF don’t – is anything else showing a dropout like this? Here are oil pressure, RPMs and MAP for the same flight.
So no, it’s confined to EGTs. I reported “We know that this can’t be all six independent sensors failing at the same time. If it’s a connection issue, the logical spots are the connectors at the firewall and the connector on the Garmin unit in the panel. I didn’t check every single parameter, but the ones I looked at – oil temp, oil, pressure, voltage, RPMs – don’t show any anomaly in their signal. To me that argues for a firewall problem versus a panel problem.” The owner replied “I have had that problem in the past years and done troubleshooting for it. It clears up for a while and then comes back.” Which probably explains why he didn’t report missing EGTs for an hour in his first ticket post. He had gotten used to it.
Scott Wasmer is our avionics specialist and knew right away what this was. He posted “There is a single plug on your firewall which has only the EGT wires, and all of the EGT wires. This is your most likely point of failure. I would suspect the plug not completely tight, and comes loose with vibration, or something internal to it but that seems highly unlikely. It could also have dirty contacts – also highly unlikely. The plug is J1016/P1016 and is located on the firewall, near top center.”
Not pointing fingers here but it does give the rest of us a chance to review our policies on taking off with known anomalies. If all of your EGTs dropped out during taxi-out would you still launch? Maybe it depends on flight conditions – how blue is the sky? – or the length of the flight or who’s onboard or the importance of getting there. All valid concerns.
Next is a Diamond DA40 powered by a Lycoming IO-360 with data from an EI CGR-30p with a one second sample rate. Here’s a typical recent flight.
Two clusters of tiny FF spikes at 01:25 and 01:55 but otherwise ok. Here’s the next flight.
EGT 3 is. hot mess. If it had dropped and stayed there we’d be more likely to suspect a failed probe or a severed connection. This looks like something’s loose – or maybe the wire is straining against a tie-wrap or an Adel clamp. It always makes sense to check the connection first, but especially when we see a pattern like this.
Here’s data from a Cessna 172XP powered by a Continental IO-360 with data from a JPI 830 with a one second sample rate. Here’s the beginning of the flight.
Strange – it just ends at the 9 minute mark. Here’s the rest.
Why did the JPI stop logging then start logging again? One of the many questions we had. Usually that only happens when power is shut off and restored. But the owner’s concern was the big drop in EGT and CHT 4. Let’s put a pin in the big drop and start at the left edge. At the 15 minute mark FF is reduced from about 12 to just over 9 GPH. CHT 4 drops from highest to lowest and stays there. Ditto for EGT 4, except CHT 4 drops while the others rise and EGT 4t barely moves while the others rise. When EGT doesn’t respond to changes in FF it suggests an induction problem, versus a FF or spark problem.
FF doesn’t change at the 3 hour mark where cylinder 4 has its excursion. I wondered if any other parameter changed. Here’s altitude and EGTs zoomed into the problem area.
Something’s fishy. The pilot reported a descent from 6500 to 4500 msl to dodge weather but this altitude trace can’t be accurate. We were pretty sure that had one cylinder essentially stopped making power that the pilot would’ve reported roughness on the ticket and probably would have not continued on for 40 minutes like nothing had happened. The pilot and copilot noticed the drop in real time, but there was no roughness. FF increases a little bit at 03:37 but the cylinder doesn’t respond. FF increases a little more at 03:45 and the cylinder acts normal again. Normal as in earlier in the flight – still lower than the others.
This still had the odor of an induction leak about it. Not making normal power, but still making some argues against a FF problem or a spark problem. Changing with altitude. So we overlaid MAP onto EGTs and found the recovery to “normal” was timed with a slight change in MAP.
