AVweb's engine fanatic Marc Cook has been flying a lot lately and depending, even more than usual, on engine monitors for peace of mind.
November 7, 2004
First permit me a program note. Starting this month, I'll be taking over as editor-in-chief of Kitplanes, a marvelous magazine for the homebuilt-aircraft enthusiast. In that capacity, I'll be getting more hands-on time with experimental engines than I ever thought possible, and will endeavor to report back here with the highlights of my travels. Generally, though, I'll be sticking to production-aircraft engine issues here.
It's been a busy month of flying in the Viking as well as two other airplanes in my inner circle, a Beech Bonanza A36 and a Cessna T210. I like them all for very different reasons, including their equipment. (The Bonanza has a Garmin 530 in the panel that has got me wriggling on the hook.) Most important, from my twisted perspective, is that two of the three have full engine monitoring equipment aboard. Sadly, the A36 is not one of them, proving that there's no such thing as the perfect rental airplane. Not that I'm complaining, mind ...
I'm not much of a fence sitter where engine monitors are concerned. Based on my experience -- and that of many others including a few whose opinions on matters of internal combustion mean a lot to me -- lack of the monitor is nearly a no-go item. In the last month, I've had flights in all three airplanes where the engine monitor -- or the lack of it -- was a central feature. Maybe I'm obsessing, but the health and well being of the only engine on the airplane tends to be a high priority.
Six New Jugs, Sweaty Palms
Let's start with the Cessna. (I don't know about you 210 owners, but I can't quite make myself call it Centurion. I keep thinking of Monty Pyton's "Life of Brian," but that's another story.)
The owner of the FBO -- Surface to Air Aviation at my home base of Long Beach, Calif., -- responsible for the airplane is a good friend and unashamedly fussy about the airplanes in his charge. I once got chewed out for leaving a seatbelt unfastened ... after parking the airplane. (Hey, it was dark.)
Bill had been putting off performing a second flight on the T210 after a full top overhaul. The first flight was fine, but the CHTs were high -- as expected -- and there were a few other, smaller anomalies. In the mean time, we had a set of GAMIjectors installed. It seemed a good time to do the installation, but the timing violated the primary rule of flight test: Keep the changes from test to test minimal and don't fix something that's not clearly broken. Oh, and based on recommendations from the manufacturer of the intercooler, the shop that did the GAMI installation also bumped up the Cessna's maximum fuel flow using a calibrated gauge and the proper procedure. (You're supposed to tap into the fuel system after the engine-driven pump and take a reading there. The factory gauge -- marked in fuel flow but in fact just a pressure instrument -- isn't considered accurate enough by Continental.) Make that two big changes between flights.
|JP Instruments Engine Monitor
One thing got in the way of another, and a few weeks passed before the second flight. Naturally, Bill was worried that the engine was running properly and asked me to come and fly the mission while he watched the engine monitor -- a recently installed JP Instruments' EDM-700.
We did our best to limit ground operations and soon were shooting down the runway at a good clip. I'll say this for a lightly loaded T210M, packing 310 horsepower for takeoff: It goes, man.
I can tell you there's nothing better for confidence than glancing down to see six orange bars standing even and proud -- representing individual cylinders' EGTs, there's no turbine-inlet temperature (TIT) probe on this airplane. One quick look and I know all six of those little engines are doing their thing, and I can concentrate on flying the airplane and manually controlling manifold pressure. (These T210s are prone to overboosting on the first flight of the day.)
But what's this? As the manifold pressure has come up, so has the fuel flow -- a lot. Maybe too much. In fact, the needle has gone over the redline to take up residence in the side of the gauge usually home to the MP needle. If the gauge is to be believed, this TSIO-520-R is sucking down better than 40 gallons per hour. Enough, I think briefly, to sharply cut power and, just maybe, make the engine quit. As we're coming through 200 feet and the gear just coming up, I decide it's time for action.
Believing the gauge, I tell Bill to manually lean to the indicated redline.
Mistake number one. (On this flight, anyway. Mistakes aren't cumulative, are they? Am I required to log them?) The next time I glance at the JPI -- perhaps 10 seconds later--one cylinder's EGT is at 1500 degrees and heading north. The CHT of that cylinder had already passed 400, and the digits were ratcheting up in increments of 3 or 4. I know in an instant that this is an unhappy engine with temps that will be uncontrollable -- and probably staging a detonation event waiting only for a little something extra to set it off. If I don't do something right away, it's going to be toast.
(A bit of experience helps with this assessment. I've broken-in engines before, two of which were fully instrumented. A new, tight engine does run hot but has gradual changes in temperature. You can only get wild swings of CHT and EGT by running it too lean for the conditions. Or, as the GAMI guys can prove, too rich for the conditions. That's also another story.)
At the same instant I dump the nose, I tell Bill, "Fuel. Give me more fuel." Bill, one of the smoothest and coolest pilots I know, begins screwing in on the vernier control. "No, Bill ... all of it. Full rich." As he's doing that, I reach over and pull the throttle closed ever so slightly, looking for a bit less boost.
Right away, the temperatures start down, and soon we're climbing with the hottest cylinder registering 400 degrees and the EGTs down the 1300-degree range. That's more like it. We fly for an hour and half, watching the JPI like it's about to tell us a secret. We run the engine hard and rich with the cowl flaps wide open, letting the rings get acquainted with the cylinder walls.
We land and let the airplane cool completely. On the next flight, the fuel flow does the same way-over-redline trick but now I pay more attention to EGT, which is sitting right at 1250-1270 degrees across the board. The gauge says the engine is drowning in fuel but the EGT tells another story. No doubt you've heard that absolute EGT values aren't important -- and I agree that in most cases they're not. But in this case, looking for a target EGT on the takeoff roll is one of those quick checks that tell you if the engine is, generally speaking, happy.
What you shouldn't do is fixate on a specific EGT number in relation to the other cylinders. Thanks to variances in engines, exhaust systems and probes, adjacent cylinders can have EGTs quite a bit different under normal conditions. Part of using an engine monitor effectively is understanding what is normal, and looking for values that are abnormal. (More on this mindset a bit further down.)
It took us a week to finally solve the mystery. First, a bit of background. When you order GAMIjectors for Continental engines, you're asked to provide either the existing nozzle size (they're graded into several flow ranges and given numbers, like 14D or 12A) or the code stamp on the fuel-distribution spider that lies between the single line that brings metered fuel pressure from the pump and the six individual lines that extend down to the nozzles themselves. Continental builds what's called a married system. A given divider manifold is supposed to be married to a given size-set of nozzles. Unfortunately for us, the manifold was the "one in a hundred" that GAMI says comes with no code stamp. Nothing. Just a part number and a free pass to mystery.
Somewhere along the line, this 210's nozzle and manifold had divorced. One result of this mismatch is that the in-cockpit gauge reads incorrectly. When we changed from the stock injectors to the GAMIs, we made the gauge read a higher-than-actual flow. GAMIs are designed to flow the same amount of fuel -- all injectors combined -- as the stock system, so that gauge calibration is preserved. To correct our problem, we had the manifold sent out for calibration and rebuild -- it had started to weep ever so slightly -- and it came back with the proper code stamp so that the next guys in line won't have to snoop around for answers. Now, of course, the indicated fuel flow is closer to where it should be -- although it does exceed redline on takeoff, but we're not going to change that back. The power curve out there on the rich side is quite flat, and I doubt we're giving up much thrust in return for a dramatically cooler-running engine.
This is a contorted way of arriving at the point: Without the JPI and an understanding of the benchmark EGT values for takeoff, we might have asked the shop to trim takeoff fuel flow to normalize the erroneous gauge indication, which would run this engine too lean and too hot at this very high power setting. The owner of this T210 now appreciates -- as I have for quite awhile -- that the comparatively small investment in the engine monitor may well have saved him another pre-TBO top overhaul and, quite possibly, a detonation-induced engine failure.
Now For The Vike
You can see where I'm going with this, right? Yet another anecdote about how an engine monitor can save the day. This time it's the 1969 Bellanca 31TC that I fly, owned by Troy Foster. For those unfamiliar, this is the first factory-turbocharged Viking (not yet a Super Viking), using a regular high-compression IO-540 with a pair of RaJay turbos managed by a pair of manual wastegates. Most of the time, the airplane is in non-turbo mode, with the wastegates wide open and the turbos just sitting there, adding weight.
The JPI in this airplane has for a long time helped diagnose fouled plugs. Get a rough runup, watch which EGT falls and you know which cylinder and which plug is the culprit. A normal runup will have all EGTs rising slightly when you select a single mag. (This is the result of the single spark event causing the overall combustion event to take longer.)
The instrument also has the data logging option, which I find both useful and fascinating. Troy hasn't had a computer set up to download the data for a while, so I took a Saturday and pulled the dozens of flights off the unit for examination. There's no hiding from the data. I could see every power change, every EGT drop or rise from every flight in the data pack. You can even tell who's flying the airplane from the data trace. Troy knows what works in the airplane, and uses a limited number of power settings and fuel flows to get what he wants. You can tell that he climbs to altitude, sets the mixture and then leaves well enough alone.
I'm the tinkerer. I'll routinely do a full lean-cycle test in a flight to see if the mixture distribution has changed. I'll try running the turbos even on a flight that doesn't strictly require them. I'm still building my mental database of the most efficient power modes in the airplane. I recently flew with Steve Ells, who took my old job at AOPA Pilot. About an hour into the flight, he looked over as I was fiddling with the mixture and said, "Can't you just leave the engine alone? It doesn't need your help." That's Steve, always cutting to the chase.
Still, even with those variables, the data are intriguing. And, in one instance, a literal lifesaver.
Troy called me after a flight from Phoenix. "Let's meet at the hangar," he said. "I think we may have a problem with number two." Cylinder two on this high-time Lycoming IO-540 had always shown a below-the-pack EGT reading, regardless of the mode of flight. Quietly, I suspected a probe or installation issue, but wanted to do some testing to eliminate that possibility. For the previous 30 hours or so of operation, number two lagged behind the pack. But it would respond in kind with the others during the lean cycle, peaking quite near everyone else, even in turbo mode. We vowed to keep an eye on it.
But Troy noticed that on this flight, the number-two EGT had trailed the pack slightly more than is normal. Not by a lot -- indeed, the data suggest that the difference would hardly have shown up on an analog gauge and probably wouldn't be noticeable on the JPI in the standard mode. (There's a Normalize mode that makes each segment on the bar graph represent fewer degrees.) Troy reported that the engine sounded and performed fine. However, he noticed that in addition to the raw EGT for number two being slightly low, the differential alarm sounded during the taxi-in back at Long Beach. It had never done that before, and, what's more, the alarm is set fairly high, at 400 degrees. That means the difference between the highest and lowest EGT has to be 400 degrees or more before the JPI hollers. That's a big spread, even at low power.
A few days later, we uncowled the airplane and got a big surprise. On this Viking, the exhaust system is split in two for each bank, but there's a small crossover tube, maybe a half-inch in diameter, that joins the system between the number one and number two cylinders. And that pipe had come free of the fitting right at number two, not three inches from the exhaust port. Judging by the soot on the pipe and the surrounding area, a good amount of exhaust gas came shooting out of the breech. Right onto the fiberglass cowling.
Without the JPI signaling a problem, how long would it have been before we discovered the problem? Another flight? A week? A month? Until the next annual? And would the cowling have survived? I'm convinced that the JPI, allied to Troy's naturally suspicious nature -- maybe I should say curious, as Troy's a genuinely good-mooder -- saved the day.
And let's have a moment of praise for the data logging. We pulled the files from the JPI and found the exact moment when the pipe came adrift, about six minutes after departure from Phoenix. It's right there in the Excel file. One data point and number two is right where it always was, the next it's dropped some 50 degrees further from the pack.
Two issues here: Have the data available and do something when it suggests a big change from the norm.
There's yet another chapter to the Viking story. Troy and I know the high-time engine is likely to signal the need for an overhaul any day now. It's been doing its thing for a long time, probably deserves a vacation. Although we were pretty sure the exhaust leak was responsible for the drop in EGT, Troy thought it was a good idea to do a quick compression test while the engine was laid bare. That, too, was a good thought, as number two (our nemesis!) retuned a resounding 12/80 compression score.
Well, there you go. Ironically, it had just passed the annual inspection with an acceptable number. Because we could hear air coming from the crankcase, we suspected either a broken ring or excessive wear at the top of the bore. The resulting very low compression ratio might be seen in the EGT, but the CHT for that cylinder was very close to where it normally runs. In any event, the cylinder shop said the exhaust valve and seat "were shot."
Because the oil analysis is good and the other cylinders are strong, the major overhaul has been deferred and we'll have the cylinder overhauled. Soon, we'll have the airplane back.
And I'll get to do more break-in flights. Something I don't really mind with an engine monitor.
The Bonanza: A Last Comment
|Six-Position EGT Gauge
Which brings us to the A36 Bonanza I fly, also run by Surface to Air. It's a fine airplane, and runs well. But the best thing in the panel is the Garmin 530. For the engine instrumentation, it's just about bone stock: tiny CHT gauge of indeterminate accuracy and an analog EGT gauge with a big six-position switch. (Better than the single-jug version, I suppose, but not by much when compared with the JPI.) No GAMIs, either.
Every time I fly this airplane, I feel a bit naked. I have but the barest of tools to read the engine. The single-point CHT doesn't tell me a whole lot -- except to confirm why the number-two cylinder on Bonanzas is most often overhauled because it runs hot -- and a quick flip through the EGT switch doesn't tell me a lot more than all six are running. It's just not feasible to switch through on the takeoff roll, which is where I need the most information presented as quickly and efficiently as possible.
I suppose it depends on your disposition. Some pilots can't fly without GPS or an active-noise-reduction headset. For me, it's an engine monitor. Every flight without one is a big guessing game.