SavvyAnalysis Puzzlers

I’ll admit stealing the title from Stephen Jay Gould’s work on paleontology. I’m a big fan of his writing and he left us far too soon. So what’s the connection to engine data? It’s a stretch, and it’s not my area of study, but here goes. Until Gould – and his co-author Niles Eldredge – shared their observations, evolutionary biologists had seen the value of changes in species, but had largely ignored stasis – periods of no change in-between. Gould and Eldredge recognized that stasis – a lack of change – is in itself data. If we rename stasis to equilibrium, then the punctuations reflect changes in the data. This came to mind because lately we’ve seen charts where something changed unexpectedly, or the data suggested something changed but the data was lying, or something should have changed, but didn’t. Let’s look at some examples. First up is data from a Continental IO-550-N in a Cirrus SR 22. Fuel, Spark or Probe? There’s a series of mixture sweeps followed by mag check. The punctuation in the equilibrium that draws your attention is that drop in EGT 5 – the green trace – to zero. Dropping to precisely zero is in itself suspicious. The cylinder was making […]

Recently I was talking with fellow pilots about flight diversions – things that pop up in flight that would cause you to change your plans and make a precautionary landing. We covered the obvious examples like fire and engine failure, then added control failure, engine roughness, and deteriorating weather, but we were kind of split on equipment failures. Some said “get on the ground and sort it out” and others said “carry on cautiously.” Not surprisingly, the carry on group has more flight time than the other group, meaning they’ve had to assess more scenarios and decide if it’s wise and safe to continue as planned, and they’ve built some confidence in their ability to do so. Then we talked about flying IFR and how to work the engine data into your scan, and I was surprised to hear how many pilots – even those with lots of hours – don’t. We’re not here to criticize flight training, but engine data is at least as important when IMC as it is on a CAVU day. There’s a lot of training on how to recognize instrument failures, but almost nothing on engine data monitor anomalies. So let’s devote this month’s Puzzler to […]

Like millions of others, last week I saw Sully recently. If you’re a pilot, I don’t see how you avoid wondering what you would have done with a powerless airliner at 2800 feet and no good options. I’m not giving anything away to tell you that one of the great lines in the movie is Capt. Sullenberger realizing that despite his 42 years of experience, he’ll be judged on the three and a half minutes between when the birds hit and the water landing. I think all pilots hope they’ll do as well when tested. I also recently watched – for about the thousandth time – the Laurel & Hardy classic The Music Box. Stan & Ollie are delivering a piano, and it brought to mind a comment from film director Chris Guest about comedy. I’ll paraphrase: If they just pull the piano out of the wagon and there’s no staircase, they carry it up to the front door, ring the bell and say “here’s your piano”, there’s no story – no comedy. As I scrub charts choosing what to write about each month, I’m not looking for comedy but I try and find charts with a good story. Most of our work as analysts […]

In case there was any doubt, this Puzzler is about mag checks. Specifically, it’s to introduce a new way to display the data from mag checks. For a while now, analysts have been doing this manually and now we have a button that automates the process. And we decided to make it available to all users of Savvy Analysis. Let’s start at the beginning. Our default display puts EGTs on top and CHTs on bottom. Our GAMI button (the gold G in the column of icons under Sigmund) transforms the display into EGT-only and overlays fuel flow. That button has been there since day one of Savvy Analysis. Now, underneath the G there’s an M which will display EGTs on both top and bottom, then isolate the odd cylinders on top and the even cylinders on bottom. From here out I’ll call it the Mag layout. The goal of a LOP mag check is to identify weakness in the ignition system. A weak plug or harness might go unnoticed with a rich mixture on both mags. Going lean and down to one mag gives the weak plug nowhere to hide, and will usually result in a misfire. The instructions in our test […]

Franklin P. Jones, a Philadelphia reporter, said that experience is what allows us to recognize a mistake when we make it again. That’s a mildly cynical approach, and the true value of experience should be to recognize the pattern that led to the previous mistake, remember the outcome, and avoid making that mistake again. Applied to engine data analysis, it’s comforting to recognize a pattern and feel confident in recommending a solution – like cleaning an injector or inspecting a spark plug – then to hear back from our client that the recommendation solved the problem. It’s less comforting to see something that we haven’t seen before and I thought I would devote this Puzzler to those.  And as we’ll see later, patterns aren’t confined to squiggly lines on a computer. The sound of the engine, vibrations in the yoke or stick, and carbon or oil on the spark plugs can all be clues in recognizing a pattern.  Let’s start with data from a Piper Arrow III with a Lycoming IO-360. Data is from a JPI 830 with a 2 sec sample rate. The top display shows oil temp – the light blue trace – and oil pressure – the yellow trace. The bottom display is fuel flow (FF) – […]

I MIFfed my airplane this week. I was installing the passenger seat in my Bellanca Viking – something I’ve done many, many times over my 26 years of ownership, and broke a weld between the seat frame and the seat rail. It’s not a huge repair, and considering how many times it has been in and out of the airplane and the various derrieres that have graced that seat since it left Minnesota in 1974, it’s surprising it lasted this long.  Savvy’s founder Mike Busch has written extensively on the topic of maintenance induced failures, but I’ve seen some analysis data lately that appears to be MIFs, so I thought I would tackle them in this month’s Puzzler.  Let’s start with data from a Beech 35 Bonanza with a Continental IO-470-C. EGTs are on top with FF – the blue trace – overlaid. CHTs are on bottom.  The flight is already in progress when the data begins – FF is about 15 GPH – and EGTs are stable. They move and change rank through the leaning process which ends at about the 21 min mark, then they’re stable until FF increases at about the 50 min mark. CHTs are another story. The good news […]

Every flight is a series of decisions. Sometimes the decisions are easy — weather’s too bad, pilot’s not feeling good, airplane’s not ready or equipped for the trip. This month’s Puzzler is a series of three flights, each with its own decision tree opportunities. Let’s warm up with data from a Cirrus SR22 with a Continental/Tornado Alley IO-550-TN.  This post-maintenance flight doesn’t get into the air. The pilot noticed the low TIT and EGT 5, which is corroborated by CHT 5. FF was bouncing up and down, engine was running rough. If we looked at FF and RPM we’d see a couple of runups and mag checks. The data suggests he had decided he wasn’t going up that day, but maybe he could clear the problem – or at least get some data to troubleshoot – at different power settings. As we go through the analysis decision tree, it seems clear that cyl 5 is the problem, and its irregularity is infecting TIT 1 (the grey trace). TIT 2, on the other hand (the gold trace at the top), seems ok. Is it possible that something is wrong with the turbo on the TIT 1 side and that’s causing the problem with cyl […]

If you’ve been flying LOP in cruise for a while you might forget that not so very long ago operating piston engines LOP was, shall we say – frowned upon. POHs didn’t mention it, except to say don’t do it. Some of the resistance was dismantled with the advent of multi-cylinder EGT gauges. But Insight and JPI didn’t lead the early charge for LOP ops, nor did TCM and Lycoming. Mike Busch was sysop of the CompuServe AVSIG in the early 90s, and he deserves credit for giving online aviators a forum for sharing ideas. George Braly from GAMI deserves immense credit, and kudos for giving us the product we needed to accomplish it. But the heavy lifting on promoting Wide Open Throttle and Lean of Peak as Standard Operating Procedure (WOTLOPSOP) was done by John Deakin.  That’s Captain John Deakin of Japan Air Lines, and earlier Air America. And one of the senior pilots of the Confederate (before their name change to Commemorative) Air Force. You might know John’s Pelican’s Perch columns from AVweb. I have about 20 minutes of left seat time in CAF’s China Doll C-46 with John in the right seat. It may be the best single 20 minutes in my logbook. John found […]

What gets your attention when you’re flying? Obviously traffic, terrain and weather get your attention, but let’s narrow it down to events or trends in the engine data that would cause you to consider a precautionary landing. High CHT, no or very low CHT, ditto those for EGT and TIT for sure, especially if the data is corroborated by rough running or poor performance. But there’s one that can show up without a lot of fanfare but can nonetheless change your plans — abnormal oil temperature.  When the colder ambient temps arrived for most US-based Savvy clients last Fall, we got questions about oil temp. Many were on the order of “If my oil temp never got above 180º on this flight, how can I be sure I’m burning the moisture out of the oil?”  Before we’re able to pronounce a given reading too low, it helps to know where that reading came from. If the sensor is located just ahead of the oil cooler, then we’re probably seeing the highest temperature in the circuit. If it’s just behind the cooler, we might be seeing the coolest point in the circuit. That’s where temperature probes are located on most engines we analyze, but if you’re doing your […]

I saw the phrase “fueling speculation” in a political context this week and wondered how those two words became attached in our lexicon. Why didn’t “provoking” or “prompting” or “eliciting” catch on the way “fueling” did? Then I opened my first SavvyAnalysis chart of the day and found myself doing exactly that – speculating about fuel. (I think for my headline to work you have to change from a participle to a gerund, but I think poetic license permits that.) Here’s the scenario. A 1982 Beech Baron with IO-550-N engines had just completed a prebuy inspection which converted into an annual. Our client had submitted data from a post-annual Savvy Test Profile which showed good mixture distribution on one engine and horrible distribution on the other engine. Our benchmark for good distribution is a GAMI spread of 0.5 GPH or less, because that should allow smooth operation in cruise at about 50º LOP. His three L engine sweeps averaged 0.5 GPH. His three R engine sweeps averaged 3.1 GPH. Here’s a screenshot of the first mixture sweep on the R engine, with a spread of 3.4 GPH. 3.4 is higher than we see with carbureted engines and their notoriously poor mixture distribution. It […]

The 14th century friar William of Ockham would probably cringe to know that his lofty and articulate principle “Non sunt multiplicanda entia sine necessitate” has been boiled down to “Keep it Simple.” The Latin translates to “Entities should not be multiplied without necessity”, but in the Information Age it is usually applied to shave away (using the razor) all but the essence of the issue, and look for the simplest answer. I’ll admit to a three-part bias that I bring to engine data analysis; 1- there is an explanation, 2 – with enough time and information we can find it, and 3- it’s caused by something mechanical or physical that we can fix or adjust.  If you’re a crossword puzzle fan like me, you start with the presumption that each box contains a letter, and sometimes – with little or no clue from the puzzle creator – you discover that the solution demands a number in the box, or two letters, or an emoji or some other variation on the theme. That moment, when you eventuallty abandon the Occam way and solve the puzzle, is gratifying. Since it’s the New Year, I resolve to continue to find interesting patterns that challenge our […]

First some old business. In the November Puzzler I used data that depicted intermittent and unusual patterns in EGT and CHT 5. I was happy to be able to include two pictures provided by our client — one showing the rocker arm on the engine lined up off-center to the valve stem, and the other showing the resulting wear pattern on the rocker arm foot. I thought those pictures connected the dots between the unusual nature of the EGT and CHT traces, and explained the recent dramatic rise in copper and brass/bronze in the oil analysis. But I did get some inquiries asking for details, so to clarify, the rocker arm should be centered on the stem, and pivot on the shaft to transfer the energy from the pushrod to the valve stem, working against a spring that’s trying to close the valve again. TCM’s tolerances are tight here – 0.002-0.015 inches for side clearance. We don’t know what this one measured, but to my naked eye the arm looks tweaked — the centers don’t line up — so the feeler guage tolerance may have measured within spec, but the arm wasn’t doing its job of fully opening the valve without side loads. That almost certainly happened in service; it’s hard […]

One of the classic lines in Casablanca comes from Claude Rains playing Captain Renault: “Round up the usual suspects.” In engine monitor analysis, when we see an EGT or CHT diverge from normalcy, it’s often the result of a clogged injector, a rogue spark plug or an induction leak. It could also be bogus data from a failing probe or a loose connection. But sometimes those suspects are examined and pronouced innocent, and we have to explore what else could be causing the problem. Let’s start with a Cirrus SR22 with a Continental IO-550-N. The oil analysis report flagged higher levels of copper and brass/bronze, noted increased blow-by, and suggested borescoping the valves and valve guides. Here’s the flight data.  Things look pretty normal except for cylinder 5 – the green trace. Can we rule out the usual suspects? A clogged injector should cause EGT and CHT to be high when ROP and low when LOP. That doesn’t happen here. A bad plug could cause the misfire patterns seen here, but a bad plug should result in slower combustion and higher EGT, and that doesn’t happen here either. An induction leak should cause a leaner mixture resulting in higher EGT and CHT across the board – also not the […]

For this Halloween Puzzler I chose three charts. The first was trick-y, and finding the source of the problem was a treat. The next one is a little scary, and the last one is downright spooky. Let’s get started. First up is the Right engine of a Baron 55, a Continental IO-550-E. Here’s the whole 40 minute flight, with FF displayed on top with EGTs. Things look pretty normal except EGTs and CHTs for cyls 4 and 6 show a problem during ground ops – before and after flight – and EGT 6 is an outlier once FF is reduced. Let’s table cyl 4 for the moment and concentrate on what’s causing cyl 6 to run hot. The usual suspects are a clogged injector or a bad spark plug. With an injector clog, we expect to see hotter EGT when ROP and cooler EGT when LOP – that didn’t happen. There’s no rank change in EGT or CHT. So the data is pointing away from fuel and toward spark.  When EGT 6 goes high, it hovers about 70 degrees above the others. There’s a clue, becasue that’s in the range we see when one plug stops firing – as in a mag check. We don’t have […]

For this back-to-school Puzzler I chose two very different scenarios. Let’s apply our analytical skills and see if we can solve them. First up is a normally aspirated SR-22 with a Continental IO-550-N. This is a test profile flight, with a series of GAMI sweeps followed by a lean-of-peak mag check. Here’s the whole flight, and the cyan trace is fuel flow. We’ll get to that giant purple spike in a bit, for now let’s start on the left. EGT 6 is high right out of the gate, stays high through the runup mag check, is lowest when FF drops into the 9 GPH range, and is high again after the tests. CHT 6 changes rank and direction of movement along with EGT 6, and that’s a characteristic of a clogged injector. Also notice that as the CHTs change cylinder 6 is always behind the pack. Slower to heat and slower to cool. If we zoom in on the GAMI sweeps, we can see that 6 is first to peak when FF leans, and last to peak when FF richens again. Also a characteristic of an injector clog. Back to the big picture, what else is going on? CHTs 1 and […]

While channel surfing recently I tuned into the opening scene of 12 Angry Men. It had been years since I had seen it, so I decided to give it a go. It’s not a flashy movie, it’s in black-and-white, and except for the final scene on the courthouse plaza, the movie takes place in one small jury room. The artistry of the movie is the story, the acting, the camera placement, and the editing. Henry Fonda leads a powerhouse cast trying to reach unanimous verdict in a murder case. On the first vote, 11 vote guilty and Henry votes not guilty. Spoiler alert: over the course of the movie Henry stands his ground, questions the evidence and one by one the other jurors re-think their positions and change their votes. In the end they all vote not guilty and go home. My uncle used to say about my aunt that she got most of her exercise by jumping to conclusions. We’ve all probably encountered a mechanic who heard a phrase or two and was immediately convinced of what was wrong with the airplane. Psychologists call it confirmation blindness — our tendency to accept evidence that supports our conclusion and reject evidence that doesn’t. […]

The web says that big data is a catch-all phrase for data sets that are so large that traditional processing applications are inadequate. Even though Savvy Analysis is just approaching its 3rd birthday, let’s call it a traditional processing application. On my last cross-country flight, my passenger was a web developer, and as we scanned for traffic, we brainstormed how the coming ADS-B rule might change engine data analysis. In the 1990s I enjoyed hearing Scott McNealy, one of the founders of Sun Microsystems, talk about smart appliances, and how a refrigerator equipped with wi-fi could monitor its systems, recognize the signs of an impending compressor failure, log onto the net, order the part, then schedule the service appointment – all without human intervention. So my friend and I tried to apply that to flying. Imagine scooting along an airway, and along with the real-time weather and real-time traffic, you have real-time engine data available. Obviously you have it already – a whole panel full – but in addition to each gauge displaying its own information, you would have a real-time graphic depiction of EGTs, CHTs, FF – all the information from your data monitor, on one screen. So when you perform […]

The comedians are fond of saying “timing is everything”. I think the quality of the material is pretty important, too, but I guess their point is that bad timing can ruin good material, and that comedy is a performance art. You might expect an article about timing from the analysis side of Savvy to be about mags, flame fronts, top dead center, and getting the most bang for the buck from your engine. This one, though, is suggestions from us about how to get the most bang for your buck out of SavvyAnalysis. When To Upload Flight Data First, let me remind you that all flight data uploaded to SavvyAnalysis is scanned for the EGT signature of a failing valve. That happens without you having to ask. If we see the pattern we will notify you. That’s true whether you’re a Savvy Maintenance client or a SavvyAnalysis Pro client. The FEVA scan has been at work for just under a year now, and we have spotted and reported the pattern to dozens of clients. So if you’re a client who only uploads flight test profiles and anomaly flights, but not normal everyday flights, hopefully the automatic FEVA analysis gives you […]

It’s wedding season. When my wife hosts a wedding shower, she organizes a game. She puts small kitchen items into brown paper lunch bags – one per bag – and ties a ribbon around each one. Each guest ties the ribbon around their neck and hangs a bag down their back. The other guests feel the bag and try to figure out what’s in it. They write down the answers, eventually all the bags are opened and someone wins. Part of me thinks this game is a remnant of a time when women were expected to know about kitchen gadgets and isn’t it a little archaic to be playing this in 2015? Maybe – but the women in the group who have earned MDs, PhDs, JDs and ATPs seem to have just as much fun with it as anyone else. I guess knowing about kitchen gadgets was just one more thing they learned on the road to liberation. I thought of this game recently because sometimes looking at engine data is like feeling the bag. You’re pretty sure of your conclusion, but you’d love to open the bag and know for sure. Fortunately this month’s Puzzler gives us two opportunities […]

Last week I heard an oldie but goodie. A couple of veteran pilots were talking to a new pilot about flying single engine planes over rough terrain at night. One of the high-timers said “If the engine quits, you pick a spot, turn on the landing light and if you don’t like what you see you turn it off again.” We all know the risks of flying, and the stats say that recurrent training and good maintenance help mitigate those risks. If we were – as the ad says, “stupid rich” – we’d have a team of mechanics check the airplane after every flight, like a Formula 1 team, and pronounce it ready for the next flight. The next best thing is to look at the engine data.  Usually in a puzzler we use an example where the data identified a problem and suggested a remedy. We’ll get to that, but before we do I want to spend a little time on questions we can’t answer, or won’t answer. Even if the team of mechanics performed their post-flight inspection and gave a thumbs-up, we’d still perform a thorough pre-flight inspection before yelling “Clear prop!” FAR 91.013 says we have to, […]

Spring training is about to wrap up, and opening day is almost here. Last week a friend and I made our annual pilgrimage to Glendale, AZ to immerse ourselves in baseball. Four games in 45 hours. At least in Arizona, spring training and GA are good neighbors. The White Sox/Dodger park is next to Glendale, the Reds/Indians park is next to Goodyear, Padres/Mariners is not far from Deer Valley, Giants/Rockies is across town near Scottsdale and Falcon. So not only do you get a baseball game, you get to watch the traffic pattern, and a little higher and further away are the constant arrivals and departures at Sky Harbor.  I know a little bit about baseball, my friend is a walking almanac and former baseball writer. We sat behind a lady keeping score – on paper in a spiral notebook – and I wondered if I could explain engine data analysis to her faster than she could explain the arcane symbols of scorekeeping to me. Her language has been around a lot longer than mine, but it’s interesting to think that in both cases, if you know the language, you can look at the data and recreate the event in […]

Flying is full of acronyms. GUMP, ADS-B, BFR, NTSB, ASRS, AOPA, EAA, NBAA, GA, WOXOF, NORDO, GPS, ILS, VOR, FARs, the 6 T’s – the list goes on. Add online chatting and texting and it’s almost TMI. In engine data analysis we have our own acronyms, and the last page of the Savvy analysis report form explains what they mean.  In this puzzler I want to cover GAMI, GIGO and FOD. Not actually GAMI the company – whose research and innovation gave a much-needed boost to GA when it needed one. GAMi as in spread – the calculation of mixture distribution determined by comparing the EGT peaks as fuel flow changes. GIGO stands for garbage in garbage out, and FOD is foreign object damage.  With GAMI sweeps, we get the most reliable results by beginning the test at 100º F rich of peak, then leaning to the onset of engine roughness, then richening back to the setting for 100º F ROP. We like to have a series of three such sweeps for reliable analysis. Here’s a set of sweeps with excellent pace, and consistent rich start points and lean end points. (The blue trace is fuel flow.) Now let’s zoom […]

 I’ve been looking at stock charts for a long time. I mean a REALLY long time – like before the Internet when they came in big bound books. I’ve also been reading music for a long time. So when the concept of engine data analysis came along, I found the concept familiar. In music, you find the prettiest notes in the middle of the instrument’s range. High notes, low notes and giant skips in pitch tend to stick out. The veteran stock chart watchers look for things like double bottoms, triple tops, breakouts, volume surges, price gaps, and compare that with everyday ordinary behavior. So what does any of this have to do with this month’s puzzler? Patterns and expectations. With engine data analysis, there are patterns we expect to see and when they’re not there we have to figure out why. What matters to a stock trader is not what the fundamentals say is supposed to happen to the price, but what buyers and sellers do with their money – and that’s what you see in the chart. In studying an orchestral score you look for melody, harmony, counterpoint and dynamics to determine what the sound will be. With engine data, we […]

Ronald Reagan used to say “Trust but Verify”. He was talking about Intercontinental Ballistic Missiles, but the philosophy works for data from engine data monitors, too. While we sit comfortably in the cockpit scanning the gauges, our data-gathering probes and cables are out there trying to work next to some hot metal in wind that qualifies as an F5 tornado. We give the data the benefit of the doubt, but sometimes the data makes us scratch our chins and look for a way to verify.  Normal, Bold, Hide and Isolate Just like snake oil salesmen don’t start their pitch with “Everything I’m about to say is !@#$%^&”, a failing probe or a loose connection will probably keep sending data and it’s up to us to separate truth from fiction. Let’s look at some examples. Take a look at the data from this IO-550-N. No, it’s not the new 3-cylinder version – we’ve set up the chart to display only the traces for EGT and CHT 1, 2, and 4. When you notice something unusual, it helps to isolate that trace, but it also helps to include a “normal” trace for comparison. If you haven’t used the chart tools, this might […]