Troubleshooting the Turbo-System
Turbocharging problems seem to be among the most elusive for A&Ps to find and fix, at least judging from the feedback we get from aircraft owners. The keys to success include having a thorough understanding of the system, knowing the symptoms that often can be tip-offs to what's wrong, and using a logical troubleshooting strategy. AVweb editor Mike Busch offers all that, plus a step-by-step checklist for diagnosing those turbo gremlins.
One of the most perplexing maintenance problem areas that owners of high-performance piston aircraft have to deal with is the turbocharging system. I often hear from unhappy owners who have already incurred considerable expense in overhauling or replacing costly turbo-system components without having resolved the trouble. As always, I advise that it's almost never a good idea to "throw money at the problem" until sufficient troubleshooting has been done to identify the actual cause of the problem.
There are very good reasons that turbocharging problems tend to be difficult for mechanics to troubleshoot. They're hardly ever reproducible on the ground, often occur only at quite high altitudes, and are sometimes quite erratic or intermittent. In most cases, the mechanic has no choice but to rely entirely on a description of the symptoms provided by the owner or pilot. Unfortunately, that description is often incomplete or misleading because the owner or pilot doesn't really understand what the mechanic needs to know to diagnose the problem correctly.
Furthermore, there's precious little cockpit instrumentation that offers any direct measurement of what the turbocharging system is doing. We don't have a turbocharger spindle speed gauge or a wastegate position indicator on the panel. Turbo problems generally show up on the Manifold Pressure gauge, and that instrument provides only a very indirect indication of what's going on with the turbocharging system.
To make matters even worse, the fundamental design of the automatically-controlled turbocharging systems used on most high-performance and pressurized aircraft tends to compensate for-and therefore conceal-problems with the engine and turbo-system, often leaving pilots blissfully ignorant that mechanical problems are developing until those problems get quite serious.
Nevertheless, if the owner or pilot understand how the system works and knows that to look for, and if the mechanic employs a logical procedure for troubleshooting the system, it's usually possible to isolate turbocharging problems problem without having to resort to the "shotgun approach" of replacing or overhauling one component after another until the problem finally disappears.
The basic principles of turbocharging are quite simple. The turbocharger itself consists of exhaust-driven turbine wheel mounted on one end of a shaft, and a centrifugal compressor impeller mounted on the other end. Engine exhaust gases, which would otherwise simply be wasted energy, are used to spin the turbine at very high speed (typically 50,000 to 100,000 RPM). This drives the compressor, which is used to boost the pressure of the engine's induction air and therefore increase the engine's power output.
Turbocharging can be employed in two ways. One, known as turbonormalizing, is used to maintain sea level manifold pressure (roughly 30 in. hg.) at altitude, thereby eliminating the progressive horsepower reduction that occurs with normally-aspirated engines as the aircraft climbs. The other, known as turboboosting, boosts manifold pressure to a value significantly higher than sea level ambient (usually 35 to 45 in. hg.) to provide increased sea level horsepower. Boosted engines normally employ some means to provide adequate detonation margins, such as reduced compression ratio and intercooling.
In either case, the turbocharging system needs to include a means of controlling the turbocharger's compressor output pressure. Without such a control system, a turbocharged engine would be fundamentally unstable.
For example, a small increase in engine power would result in a small increase in exhaust volume. This would cause the turbocharger to spin faster, which would increase the compressor speed and therefore the manifold pressure. This would result in an additional increase in engine power, producing more exhaust volume, faster turbocharger speed, higher manifold pressure, etc. In other words, the system would "run away" and very possibly exceed maximum engine operating limits.
Likewise, a small decrease in engine power would cause a reduction in exhaust volume, a decrease in turbocharger speed, a reduction in manifold pressure, a further decrease in engine power, and so forth. In short, the engine would be nearly impossible to control.
The way turbocharger output is regulated is by means of a butterfly valve called a "wastegate" which allows a certain amount of exhaust gas to be vented overboard without going through the turbocharger. If the wastegate is fully open, almost all of the exhaust bypasses the turbocharger; if it is fully closed, virtually all of the exhaust must go through the turbocharger.
Most high-performance turbocharged aircraft (including all pressurized piston models) employ an automatic wastegate control system to regulate the turbocharger. (A few aircraft such as the Mooney 231 and Piper Seneca use fixed wastegates, while Cessna's T182/TR182 and some older aftermarket turbo conversions use manually-controlled wastegates.) The automatic system employs a hydraulic wastegate actuator and an pressure controller to maintain turbocharger output at the desired pressure.
The wastegate butterfly is normally held in the full-open position by a strong spring, allowing exhaust gas to bypass the turbocharger. Engine oil pressure applied to the wastegate actuator causes the wastegate butterfly to close, forcing exhaust gas to go through the turbocharger. The more oil pressure is applied to the wastegate actuator, the more the wastegate closes until, at about 50 PSI, the butterfly is fully closed.
The pressure controller monitors the output of the turbocharger's compressor (also known as "upper deck pressure" or UDP), and regulates the oil pressure to the wastegate actuator to hold the turbocharger output constant. The controller is a simple device that consists of an aneroid and a poppet valve. Here's how it works.
If the turbocharger output is less than the set-point of the controller, the aneroid expands and closes the poppet valve, increasing the oil pressure in the wastegate actuator and causing the wastegate to close. This causes more exhaust to pass through the turbocharger, spinning it faster, and increasing the compressor output.
If, on the other hand, turbocharger output rises above the set-point of the controller, the aneroid contracts and opens the poppet valve, decreasing the oil pressure in the wastegate actuator and allowing the wastegate to open a bit. This lets more exhaust gas bypass the turbocharger, slowing it down and decreasing compressor output. Thus, equilibrium is quickly reached whereby the turbocharger output stays right at the set-point of the controller and the system remains stable.
Most unpressurized turbos use an absolute pressure controller (APC) set to maintain turbocharger output at a few inches over engine red-line. The controller set-point is easily adjustable by screwing the controller's poppet valve seat in or out, and should be adjusted so that application of full throttle produces the proper red-line manifold pressure for takeoff.
Most pressurized piston aircraft use a variable absolute pressure controller (VAPC). The difference between an APC and VAPC is that the VAPC's set-point is varied by means of a cam connected to the throttle control. At full-throttle, the VAPC works just like an APC (and is adjusted to produce proper full-throttle MP in the same fashion). But at partial throttle settings, the VAPC set-point is reduced so that the turbocharger doesn't have to "work so hard" when the pilot throttles back to reduced manifold pressure.
To gain a better understanding of how the system works, let's follow it through an actual flight profile and see what it actually does. To make things simple, let's suppose we're flying an unpressurized airplane like my T310R that uses a simple absolute pressure controller. (The differences when flying a pressurized airplane with a VAPC are minor and not really significant for purposes of this discussion.)
You've probably noticed that when flying a normally-aspirated airplane, full-throttle manifold pressure at takeoff never quite reaches sea level ambient (around 30"), but tops out at a few inches less than that due to unavoidable pressure losses in the induction system. Likewise, for a turbocharged airplane to achieve rated red-line MP on takeoff, the APC set-point must be a few inches higher to compensate for induction system losses. My T310R's manifold pressure red-line is at 32", and the APC set-point is adjusted to about 3" higher (about 35") to produce red-line MP on takeoff.
Let's start the engines and taxi to the runup area. The APC sees that the turbocharger output is less than its set-point of 35" so it closes its poppet valve to call for wastegate to close. As soon as engine oil pressure comes up, the wastegate (which is spring-loaded to the full-open position) will close all the way. However, since the engine is at idle, there's not enough exhaust flow to spin up the turbocharger enough to produce 35" of UDP, so the wastegate remains fully closed throughout the taxi and probably even during the runup.
Now we taxi onto the runway and slowly apply full throttle for takeoff. As the engine develops more and more power, the exhaust flow increases dramatically and spins the turbocharger faster and faster, causing UDP to increase until it reaches the controller set-point of 35". At that point, the controller opens its poppet valve to relieve the oil pressure to the wastegate actuator, allowing the wastegate to open as necessary to stop the turbocharger from spinning up any faster and thereby holding UDP right at 35" (and indicated MP right at the 32" red-line).
Climbing out of 1000' AGL, we reduce to 75% cruise-climb power (which in my T310R is 29" MP and 2350 RPM). Throttling back to 29" MP reduces the exhaust flow from the engine, and reducing RPM from 2700 to 2350 reduces the exhaust flow even more. This causes the turbocharger to start slowing down, but the controller immediately notices the resulting decay of UDP and closes its poppet valve to command the wastegate to close and force more exhaust through the turbocharger, causing the turbo to spin back up to the point where UDP is steady at 35". This all happens so quickly that we're never aware that it's going on.
As we climb on up to the Flight Levels, outside ambient pressure decreases by about 1" per 1,000' of climb. This decreased pressure would normally cause a corresponding decrease in UDP (and therefore MP), but once again the controller compensates for this decay by gradually closing the wastegate more and more as we continue to climb, forcing more and more exhaust through the turbocharger and spinning it up faster and faster as required to maintain UDP at a constant 35". In the cockpit, we notice that MP is staying more-or-less right where we set it (at 29"), without the inch-per-thousand-feet drop-off that we'd expect in a normally-aspirated airplane.
Of course, this can't go on forever. If we were to keep climbing higher and higher, and the controller were to keep closing the wastegate more and more to compensate for the decreased ambient air pressure, eventually we'd reach a point where the wastegate was fully closed and the controller was no longer able to maintain 35" of UDP. In my T310R at 75% cruise-climb power, this occurs at around FL220, while in many other turbocharged aircraft, it occurs somewhat higher (FL250 or more). At this point, when the wastegate is fully closed and the automatic control system is no longer able to maintain constant UDP, the engine is said to be "bootstrapping" because the system is unregulated (and therefore unstable) and large MP variations may be observed.
But we don't want to go that high today. Let's suppose we level off at FL180 and let the airplane accelerate to cruise speed. As the airspeed increases, the ram air effect causes a small increase in induction air pressure and a corresponding (somewhat larger) increase in UDP. Again, the controller notices this happening, and commands the wastegate to open a bit in order to slow down the turbocharger and hold UDP right at 35".
Now that we're trimmed for level cruise at FL180, we slowly pull back on the prop controls to reduce RPM from 2350 (top of the green arc on the tach) to 2250 RPM. As we reduce engine RPM, exhaust volume is also reduced, causing the turbocharger to spin slower and reducing UDP. The controller reacts by commanding the wastegate to close in order to spin the turbo back up and restore 35" UDP.
Suppose we continue to reduce RPM gradually from 2250 to 2100 RPM, which is the bottom of the green arc on my T310R. As we do this, the controller closes the wastegate further and further in order to compensate for the reduced exhaust flow and maintain 35" UDP. But at some point around 2150 RPM, the wastegate will reach the fully closed position and any further reduction in RPM will cause the engine to bootstrap (indicated both by loss of MP and instability of MP readings). Upon observing the onset of bootstrapping, we increase RPM by 50 or so and see that the bootstrapping stops.
Okay, we've had enough fun, and it's time to head back to the barn. We switch off altitude hold on the autopilot, and roll in enough nose-down pitch trim to start a 1,000 FPM descent out of FL180.
As our indicated airspeed rises from its cruise value of 160 KIAS to around 200 KIAS, increased ram air tries to increase UDP above 35", but the controller sees this and opens the wastegate enough to hold UDP steady. As we descend, outside ambient increases by about 1" per 1,000', so the controller must continually open the wastegate more and more to prevent UDP from rising. In the cockpit, all we see is that MP remains rock steady at 29", right where we set it.
By the time we get down to pattern altitude, the wastegate is most of the way open. It stays there until we throttle way back for our final descent and landing. When we do that, the reduction in engine power causes exhaust volume to fall, and the controller has to close the wastegate to make up for it and maintain 35" UDP. Eventually, as we close the throttle all the way prior to touchdown, even full-closed wastegate is not enough to maintain 35" UDP because the idling engine is hardly putting out any exhaust volume at all. The wastegate remains fully closed as we turn off the runway and taxi in to the ramp. It remains fully closed until we pull the mixtures to idle cutoff, at which point engine oil pressure goes away and the wastegate returns to its spring-loaded full-open position.
The preceding discussion is all predicated on an engine and turbo-system that is working properly, of course. A wide variety of mechanical ailments can interfere with the proper operation of the system. These include:
Internal engine problems
All of these problems can result in improper operation of the turbocharging system, but each one tends to produce symptoms that are subtly different in character. Therefore, a careful analysis of the symptoms can often help pinpoint the cause of the problem, or at least rule out some of the possibilities and help narrow the search. But since-as we noted earlier-most turbo problems show up only at high altitudes and seldom on the ground, it's often up to the pilot (rather than the mechanic) to make critical observations of the symptoms and decipher what they mean.
One of the most common causes of turbo-system problems are leaks in the induction system. I recall, for example, helping a member troubleshoot his Cessna T310 in which it turned out that an engine control cable had been chafing against the engine's induction manifold in a hard-to-see location. Eventually the steel cable wore a slot all the way through the wall of the cast aluminum induction pipe, creating a fairly significant induction system leak.
The pilot squawked the problem when he notice a significant manifold pressure "split" between the two engines while cruising at the Flight Levels. The engines appeared to be operating normally on the ground, during takeoff, and when operating at low and middle altitudes. The problem only showed up when the airplane was flying up high. If you think about the consequences of an induction leak, that's not surprising.
Consider what happens during a full-power takeoff at or near sea level. (Once again, let's assume the airplane in question is my Cessna T310R so we can use the same numbers as we did before.) Manifold pressure inside the induction manifold is at red-line (32"), but that's only a trifle greater than outside ambient pressure (around 30"). So relatively little induction air escapes through the leak. What little loss there is will be sensed by the turbo controller as a loss of UDP, and the controller will cause the wastegate to close just a trifle, compensating for the small loss and effectively concealing the problem. From the cockpit point of view, both MP needles are right where they should be and everything appears nominal.
At 1,000' we throttle back to 29" MP for cruise-climb. What's the outside ambient pressure at 1,000'? About 29"! So now, the induction leak becomes a total irrelevancy, since the pressure inside and outside the induction manifold are virtually identical, and so there's no loss of pressure through the leak at all.
As we continue to climb at 29" MP, outside ambient decreases by about 1" per 1,000' so the pressure differential between inside and outside the induction manifold increases steadily. More and more induction air escapes through the leak. However, the turbo controller senses this loss and keeps closing the wastegate and cranking up the turbo output to compensate for it. In the cockpit, the MP needle never wavers from 29" and so the pilot remains blissfully unaware of the problem. The wastegate on the leaky engine is closed more than it should be, and the turbocharger on that engine is spinning faster than it should be, but the engine is running just fine and there's no cockpit instrumentation to provide the slightest clue that something's awry.
The pilot's first indication of a problem comes as the airplane climbs through 15,000' and the MP on the troubled engine starts to fall and become erratic, while the MP on the other engine remains rock-solid at 29". What's happened, of course, is that the pressure differential across induction leak has become so great (about 15" now) that even the maximum output of the turbocharger can no longer keep up with the loss. The controller, it its now-futile attempt to compensate for the leak, has commanded the wastegate to go fully closed, and the engine has started to bootstrap something that should normally not happen until the airplane climbs well into the Flight Levels.
These are the classic symptoms of an induction leak problem: normal operation at takeoff and low altitude, and the premature onset of bootstrapping (i.e., loss of MP and MP regulation) at higher altitudes. Unfortunately, there are other kinds of problems (e.g., exhaust leaks) that can produce the same symptoms.
So how can you be sure?
Good question! It turns out that there's another symptom-one that the owner of this aircraft missed-that can often be used to distinguish an induction leak problem like this one from various other kinds of turbo-related problems. Best of all, this symptom is one that can be checked without having to take the airplane up to high altitude, or even leaving the ground at all! The tip-off is higher-than-normal MP when the engine is throttled back to idle.
Consider an engine idling on the ground. The engine is "trying to breathe" but the throttle is retarded to idle, closing the throttle butterfly and choking off most of the available induction air. (It's called a throttle because it chokes off the engine's airway!) The result is a significant vacuum in the induction manifold, as the engine consumes the air in the induction manifold but the closed throttle butterfly blocks the inflow of air to take its place. In the cockpit, this shows up as a very low MP reading (typically, something on the order of 12" to 15") far below outside ambient (around 30" at sea level).
But suppose there's a substantial leak in the induction plumbing somewhere between the throttle and the cylinders. What happens? Ambient air rushes in through the leak because of the vacuum in the induction manifold. In the cockpit, this shows up as a higher-than-normal MP indication at idle perhaps 17" instead of 14". The engine will also be idling leaner than usual-since the leak lets in more air but not more fuel-so the engine may tend to stumble a bit when you throttle-up for taxi (at least if the leak is big enough).
High MP at idle isn't a perfect tool for diagnosing induction leaks. Some induction leaks won't produce this symptom (e.g., leaks in the upper deck portion of the system prior to the throttle body). Also, the symptom can be produced by other things besides an induction leak (e.g., a non-firing cylinder or a badly misadjusted idle mixture).
But certainly if you see both abnormal bootstrapping at altitude and high MP at idle, certainly the odds favor an induction leak, and that's probably the first place you should look for trouble.
If you suspect an induction leak (based on the observed symptoms), the first step should be to confirm the diagnosis by performing a critical altitude check. The procedure is described in detail in your aircraft service manual, and consists of a test flight at altitude in which certain power settings are established at certain altitudes, and the MP readings are recorded. The service manual has tables that establish how much MP you should be able to obtain at these benchmark altitudes under specified conditions of RPM, fuel flow and temperature. If your engine falls significantly short, then you can be sure you have a problem and odds are that it's an induction leak (although there are other possibilities that we will discuss later on).
If you can't find anything obviously wrong after careful visual inspection of the induction system, a simple pressure check may be in order. All that's required is to pressurize the induction system with a few PSI of air-one good way is simply to pump air into a cylinder as if you were doing a compression check, but rotate the prop so that the cylinder's intake valve is open-then close the throttle and go over the entire induction system with a soapy water spray, looking for leaks that reveal themselves by blowing bubbles. Some leaks are expected at the induction system drains and, to a lesser extent, around the throttle shaft, but the rest of the induction system should be completely airtight.
An exhaust leak can produce similar symptoms to an induction leak-the onset of bootstrapping at a lower-than-normal altitude-because any exhaust that escapes through a leak bypasses the turbocharger just as if it escaped through an open wastegate. Just like with an induction leak, the turbo controller will try to compensate for (and thereby cover up) the problem by commanding the wastegate to close, so the symptoms generally won't show up until the airplane is at high altitude. (Unlike a lower-deck induction system leak, an exhaust leak will not affect MP at idle.)
Exhaust leaks are inherently much more dangerous than induction leaks, because of the very serious threat of in-flight fire. Fortunately, exhaust leaks are usually a lot easier to detect because they typically leave brightly-colored exhaust stains (and sometimes also obvious heat damage) that can be detected visually during an engine-compartment inspection. All turbocharged aircraft should have their exhaust systems meticulously inspected for leaks every 50 hours, and this is required by Airworthiness Directive for certain aircraft such as the Cessna T210 and all turbocharged twin Cessnas.
Because the exhaust system operates under extreme heat and pressure, and because exhaust gas is so very corrosive, exhaust leaks can sometimes develop suddenly (a "blowout") rather than gradually. The pilot of a turbocharged aircraft who experiences a sudden unexplained loss of manifold pressure in-flight should assume that an exhaust failure may have occurred, and should put the airplane on the ground at the earliest possible moment. If the aircraft is a twin, the pilot should consider the possibility of shutting down and securing the engine to minimize the threat of in-flight fire.
I don't mean to frighten you with this statement. I have done extensive investigation of exhaust failures in turbocharged aircraft, and has concluded that the risk of an in-flight exhaust failure (particularly one of the "blowout" variety) is extremely remote on aircraft whose exhaust systems have been properly maintained and inspected. The vast majority of in-flight blowouts and exhaust-related fires involved exhaust components with very high time and usually ones with poor-quality weld repairs that failed.
Internal engine problems
A third possible cause of bootstrapping at a lower-than-normal altitude is an internal engine problem that prevents one or more cylinders from firing. This would most likely be something that reduces the compression of a cylinder to near-zero (such as a valve that's badly burned or stuck open), or something that prevents both spark plugs in the cylinder from firing (such as severe lead fouling of both plugs). A non-firing cylinder reduces the exhaust output of the engine by one-sixth (assuming a six-cylinder engine), and this means less flow through the turbocharger. As usual, the controller will try to compensate (and cover the problem up) by closing the wastegate, but this means that the wastegate will go full-closed at a lower-than-normal altitude.
You'd think that a six-cylinder engine that was firing on only five cylinders would be very obvious to the pilot, wouldn't you? Well, I can tell you from firsthand experience that unless you have a probe-per-cylinder EGT system that shows one cylinder running ice cold, you're very likely not even to notice the loss of one cylinder particularly in a twin where any roughness or loss of power is masked by the other engine. Trust me on this one!
So if you notice bootstrapping at unusually low altitudes but can't seem to find any leaks in the induction and exhaust, it's definitely worth doing a compression check and having a look at the plugs to make see whether one cylinder is not firing and/or operating at near-zero compression. If so, you've probably found your culprit.
By the way, a zero-compression cylinder will generally cause the same abnormally high MP at idle as an induction leak. This is a symptom you should watch for before every flight. If you see it, it's usually a tip-off that something significant is wrong.
The other most common cause of turbo-system problems, besides induction leaks, are problems with the wastegate and wastegate actuator. It makes perfect sense that the wastegate would be one of the most problematic parts of the turbocharging system, because it performs such an unenviable job: regulating the flow of incredibly hot and corrosive exhaust gases.
Most wastegate problems are of the "sticky wastegate" variety in which the shaft on which the wastegate butterfly pivots gets "coked up" with byproducts of combustion (a nasty concoction of lead, carbon and sulfur) to the point that it no longer opens and closes smoothly when commanded to do so by the wastegate actuator. Another somewhat less common cause of "sticky wastegate syndrome" occurs when the wastegate actuator itself starts to bind as a result of the accumulation of oil-borne deposits, O-ring deterioration, and/or scoring of the actuator cylinder.
Whatever the exact cause of the sticky wastegate, the result is that the constant series adjustments commanded by the turbo controller-which are normally executed so rapidly and smoothly that they are unnoticeable to the pilot-become jerky and erratic. The result shows up as abnormal MP fluctuations, especially during periods of constant wastegate movement such as climb, descent, and flight in turbulent air.
It's easy to confuse the erratic MP fluctuations caused by a sticky wastegate with the unregulated MP fluctuations caused by bootstrapping, but they're really quite different if you know what you're looking for. Bootstrapping (due to a fully-closed wastegate) is a condition that predictably occurs at high altitude and low engine RPM, and which can be made to disappear at will by increasing RPM a bit or descending a bit. On the other hand, erratic MP fluctuations due to a sticky wastegate generally occur at various altitudes and RPM settings, and are most obvious during changes in altitude, power settings, and airspeed (all times when wastegate adjustments are most likely to be commanded by the controller).
If you suspect you might have a sticky wastegate, it's easy to check in the shop. Simply remove the oil line that runs from the engine oil pump to the wastegate actuator. Hook a source of adjustable air pressure to the oil inlet port of the actuator-an ordinary cylinder compression tester is ideal for this purpose. Now simply watch the wastegate assembly as you slowly and repeatedly vary the air pressure from zero to 50 PSI and back. As air pressure reaches 15 PSI or so, the wastegate should start to close smoothly, reaching its fully-closed position when the pressure reaches around 50 PSI. As you back the pressure down towards zero, the wastegate should open smoothly. Watch for any signs of jerkiness or binding as you exercise the wastegate in this fashion. Any tendency to stick should be obvious during this test. Also make sure the wastegate butterfly opens and closes fully, a total movement of approximately 90 degrees of shaft rotation.
If the wastegate appears to be sticky, it's possible that you might be able to "rescue" it by giving it a good soak overnight in a strong penetrant like Mouse Milk or AeroKroil. But don't count on it. If a penetrant soak doesn't result in silky-smooth action, it's time to yank the wastegate and send it out for overhaul. Expect to pay around $400 for the overhaul, plus a couple of hours of labor to remove and reinstall. Make sure that the overhauled wastegate is installed with the proper high-temp attaching hardware and new gaskets.
When the turbocharging system starts acting up, pilots and mechanics alike have a tendency to name the turbo controller as prime suspect. In fact, the controller is hardly ever the culprit, and it's an unfortunate fact that lots of perfectly good controllers are sent out for overhaul in the course of "shotgunning" turbo system problems.
The turbo controller is seldom the culprit for two reasons: it has a terribly easy job, and there's not a whole lot that can go wrong with it. Think about it for a minute: the controller never sees hot exhaust gases or high engine temperatures like other turbo-system components. It spends its life in almost air-conditioned luxury, watching upper-deck pressure at one end, and regulating oil flow to the wastegate actuator at the other. If I had to come back as a turbo-system component in my next life, I'd almost surely apply for the job of turbo controller!
That's not to say that the controller cannot cause turbo problems, only that it's one of the system last components you should suspect. Before you send the controller out for overhaul (which doesn't come cheap), there are a few things you should try first:
If you're flying a twin, try swapping the left and right controllers and see if the problem changes sides or stays put (or goes away altogether). The swap generally takes less than an hour and could save you many hundreds of bucks and a week or two of downtime.
If the controller is at fault, it might just be that the poppet valve is sludged up. Disconnecting the oil lines and shooting a few shots of shop air into the oil return port might dislodge the gunk and fix the controller problem. (At least it worked for me last time I tried it.)
The controller's upper deck air reference line and inlet port should contain nothing but air. Disconnect the line and inspect for any signs of liquid contamination (fuel or oil). If you see any, purge the line with solvent and shop air, and disassemble the controller's aneroid chamber and clean it out, too. (That's a whole lot less scary than it sounds.)
If the problem follows the controller when you swap sides (twins only) and a simple cleaning doesn't resolve the problem (singles or twins), only then should you consider sending the thing out for overhaul.
Of course, turbocharging problems can also be caused by-ta da!-the turbocharger itself. I'm listing this component last, not because it seldom fails, but because when it does fail, the failure is more or less obvious. While a turbocharger can last a full engine TBO if the engine is operated with sufficient TLC, it's certainly not unusual for a turbo to need a mid-term overhaul.
Turbochargers have several failure modes, most of which are more-or-less self-diagnosing. For example, turbos sometimes fail catastrophically-the engine suddenly goes normally-aspirated (or quits from over-rich mixture) and the aircraft starts trailing black smoke (actually, oil being pumped into the hot exhaust). You put the airplane on the ground fast, the tower rolls the equipment, and your mechanic doesn't have much difficulty figuring out what to do next. A catastrophic failure like this generally stems from a turbocharger that suddenly develops a serious out-of-balance condition while spinning at 50,000 RPM or so, often because the turbo ingested some foreign object (like a nut, bolt, alternate air door hinge, or chunk of exhaust valve) that damaged the compressor or turbine wheel.
A less spectacular failure mode occurs when the turbocharger center section wears out, generally resulting in engine oil winding up where it doesn't belong: in the compressor and/or turbine portions of the turbo. If oil leaks into the compressor, it will result in oil-soaked induction plumbing (you'll likely see oil dripping from an induction system drain) and oil-fouled spark plugs. If it leaks into the turbine, it will result in an abnormal accumulation of oily deposits on the tailpipe and belly, and sometimes oil dripping from the tailpipe after shutdown. A visual inspection of the turbocharger will confirm oil where it doesn't belong, and often excessive play when the shaft is wiggled. Time to yank and overhaul the beast.
A third turbocharger failure mode is a bit more subtle, and stems from the fact that the turbine wheel operates under tremendous centrifugal forces as it spins at 50,000 to 80,000 RPM, while the metal loses much of its strength at the white-hot 1600°F temperatures at which the turbine operates. The result is that the turbine blades develop a very gradual "stretch" over the life of the turbocharger-particularly if it's run hot and hard-and ultimately they can stretch enough that they actually start to scrape on the turbine housing. The turbo should be inspected for signs of blade scrape at each annual (and any other time the tailpipe is removed). If it is noted, it's overhaul time.
If you have a turbo-system problem, the best way to avoid falling victim to the expensive, time-wasting "shotgun" approach is to devise a logical troubleshooting strategy. Your strategy should seek to find or rule out potential causes in a sequence that starts with the most common failure areas and the quickest, easiest and cheapest troubleshooting steps, and then gradually works towards rarer failures and more difficult and costly actions. It's impossible for me to suggest a one-size-fits-all strategy, but here's a starting point:
The first step in any turbo troubleshooting strategy should be a thorough test flight to document the exact symptoms. By all means do a "critical altitude check" as documented in the service manual for your airplane, to determine if you have a premature bootstrapping problem and to quantify just how serious it is. Be sure to note whether any erratic or abnormal MP indications occur only at high altitudes and/or low RPMs (indicating bootstrapping), or if they occur primarily during climbs, descents, and/or airspeed changes (indicative of a sticky wastegate). Also note whether MP at idle is higher than normal (indicative of an induction system leak).
Do a quick compression check and spark plug inspection to determine whether you have a cylinder not firing or with near-zero compression.
Visually inspect the induction system looking for any sign of a leak (chafed-through manifold, cracked induction coupling, loose hose clamp, etc.). If nothing suspicious is found, pressurize the induction system and use soapy water to search for leaks.
Visually inspect the exhaust system looking for any sign of a leak (tip-off is usually brightly colored exhaust stains). If nothing suspicious is found, pressurize the exhaust system and use soapy water to search for leaks.
Connect a source of variable regulated shop air (such as a cylinder compression tester) to the oil inlet port of the wastegate actuator, and exercise the actuator by repeatedly varying the air pressure between 0 and 50 PSI, watching for any sign of sticky operations. If the wastegate is sticky, try an overnight penetrant soak. If that doesn't free it up, pull the wastegate and have it overhauled.
Remove the oil lines from the turbo controller and blow shop air through the poppet valve to dislodge any sludge. Remove the air reference line and inspect for any signs of fluid contamination. In a twin, try exchanging the two controllers to see whether the problem moves with the controller. If it does, and if cleaning the controller doesn't help, send it out for overhaul.
Check the turbocharger for signs of FOD, oil in the compressor or turbine, excessive center section play, and turbine blade scrape. If any of those problems are noted, pull the turbo for overhaul.