The Savvy Aviator #26: Interpreting Your Engine Monitor
The modern probe-per-cylinder digital engine monitor is a marvelous tool for keeping tabs on your engine's health and troubleshooting its maladies. Here are some tips for figuring out what those bars and digits mean.
This is a bit embarrassing, but I might as well come clean: Up until a few years ago, I was still flying my Cessna T310R with only the primitive engine instrumentation installed by the factory in 1979. Shame on me!
I'd long since upgraded my avionics stack with conspicuous quantities of glass, including a Garmin GNS-530 navigator and a Sandel SN3308 Electronic HSI. I'd installed an XM Satellite Radio receiver to pipe stereo music to my ANR headsets. I'd added VGs to the wings and vertical tail. I'd even reupholstered my seats with the latest visco-elastic memory foam padding. But I was still relying on 30-year-old, steam-gauge engine instrumentation.
A modern, digital, engine monitor had been at or near the top of my wish list for years. Yet somehow the $5,000 I had set aside for this upgrade always seemed to get preempted by something else (usually non-aviation-related) every March when my annual comes due. In addition, I secretly dreaded what I expected to be a difficult, time-consuming and tedious task of installing such a system in my turbocharged twin, because it involves stringing a zillion wires from the cabin out through the wings to the engine nacelles. So for years I kept making excuses, procrastinating, and occasionally catching good-natured flak from my aviation friends and colleagues.
A few years ago, I finally bit the bullet. I purchased an EDM-760-6C system from J.P. Instruments (taking advantage of their Sun 'n Fun show special discount) and installed it in my airplane. The installation was indeed time-consuming -- it involved the installation of 29 sensors and 500 feet of wiring, and took me a full week -- but wasn't nearly as difficult as I anticipated and worked perfectly the first time I powered it up.
Within months, I was kicking myself for not installing the system years earlier. It's without doubt amongst the best time and money I've ever spent on my airplane.
During the past few years and more than 600 hours of flying with my engine monitor, I've gradually learned more about how to use it and how to interpret the data it displays and logs. The more I've learned, the more I'm convinced this kind of instrumentation belongs in every piston-powered aircraft. A digital engine monitor arguably offers the best cost-benefit ratio of any item of avionics you can install.
Which Monitor To Choose?
Today's most feature-rich engine monitors come from J.P. Instruments and Electronics International. These two firms are constantly trying to leapfrog one another in terms of features, and to undercut one another in cost. Both JPI and EI have enthusiastic customer followings. Both make excellent instruments and offer superb customer support. Choosing one over another is somewhat like choosing between a high-wing and low-wing airplane.
In my case, however, the choice between JPI and EI was a no-brainer. JPI makes a twin-engine model (the EDM-760) that competes with (and easily surpasses) the Insight GEMINI 1200. EI does not offer a twin-engine model, so their only solution for twin owners is to install two UBG-16 instruments, one for each engine. I had panel space for one 3-1/8-inch instrument but not for two 2-1/4-inch instruments, so my choice was pretty obvious.
Over the past two years, my engine monitor usage has evolved considerably as I've learned more about what it can show. These days, I find myself referring to the instrument during virtually every phase of flight.
At my home base airport, the taxi from my hangar to the approach end of the normally-active runway is nearly a mile, so there's plenty of time to perform various preliminary checks during taxi-out. For example, I've long used this time to check my three gyroscopic flight instruments (AI, DG, T&B) for proper operation.
Nowadays, I frequently perform a preliminary ignition check during taxi-out. This consists simply of briefly shutting off one magneto switch at a time (I have four of them in my twin) while watching the engine monitor. What I'm looking for when I turn off each mag is six EGT bars rising and none falling on the associated engine. This verifies that all six spark plugs connected to the remaining operating magneto are firing properly. (For most engines, EGT will rise 50°F or more on one mag.)
If I see five EGT bars rise and one fall, I know that a spark plug isn't firing. This isn't reason to panic -- yet -- because it may simply be an oil-fouled spark plug that will clear itself during runup. (The engines in my 310 are canted due to wing dihedral, so the inboard bottom plugs are especially vulnerable to oil fouling after extended periods of disuse.)
If the taxi-out ignition system check looks good, I may skip the usual runup altogether, particularly if this is not the first flight of the day and the engine is warm enough to fly with.
Otherwise, when I reach the runup area, I'll throttle up to normal runup RPM, which is 1,700 RPM for my direct-drive Continentals. I'll wait a moment for the EGTs to stabilize and then place the instrument into "normalize" mode (which sets all EGT bars to mid-scale and increases their sensitivity from 40 °F to 10 °F per division). Then I'll once again briefly shut off each mag switch one at a time while watching the engine monitor and looking for six bars rising and none falling.
With a digital engine monitor, I no longer even look at the "mag drop" on the tachometer. RPM drop is a very crude and indirect indication of ignition performance -- EGT rise of each cylinder is much better. If the engine runs smoothly on each mag individually, and if all of the EGT bars rise together as each mag is switched off, you're good to go.
If it's the first flight of the day or the OAT is unusually cold, I'll cycle the props once -- otherwise I won't bother. Finally, I'll take the engine monitor out of normalize mode, throttle back to idle, run my before-takeoff checklist, and I'm good to go.
Ever since I started flying twins 18 years ago, my takeoff procedure has been to taxi into position on the runway, set the brakes, slowly throttle the engines up to about 50% power, and then scan the engine gauges to ensure that there are no big "splits" between the two engines and that everything is in the green. If everything looks good, I release the brakes and slowly throttle the engines up the rest of the way to full takeoff power while rolling down the runway. When the engines reach full power, I make one more scan of the engine gauges to ensure that both fuel flows are where they should be, no splits, and everything still in the green. By that time, I'm usually close to liftoff speed.
To keep things simple and consistent, I now use precisely the same procedure when flying singles (except that there aren't any splits to look for).
Now that I have the engine monitor installed, my takeoff procedure is essentially unchanged except that I include the engine monitor in my two scans of the engine instruments. During my 50%-power scan, I check to make sure that all EGT bars are coming up evenly. During my full-power scan, I check again to make sure that all EGT bars are about where they should be for takeoff -- around 1,300 °F for my turbocharged engines, typically somewhere around 1,200 °F for normally-aspirated.
If any EGT is substantially hotter or colder than the others, that constitutes grounds for rejecting the takeoff and returning to the runup area to sort things out. Ditto if either fuel flow is less than normal for takeoff.
Climbing out of 1,000 AGL, I typically accelerate to cruise-climb airspeed. Then I make an initial power reduction to approximately 75% power, and lean to something close to best-power fuel flow (around 100 °F to 150 °F rich of peak EGT -- ROP).
At this point, my engine monitor focus shifts from EGT to CHT: I endeavor to make sure that all CHTs remain at or below 380 °F or so. If I notice any CHT edging up higher than 380 °F, I take prompt corrective action -- either by enrichening the mixture, increasing the airspeed, or both.
Upon reaching cruise altitude, I level off (usually by pressing George's altitude-hold button) and allow the airspeed to accelerate and stabilize. I then perform a "big mixture pull" to reduce fuel flow comfortably into lean of peak (LOP) territory, and close the cowl flaps.
Next, I fine-tune the mixture. If I want to go fast, I increase fuel flow until the hottest cylinder(s) reach 380 °F. If I want to go far, I reduce fuel flow to the edge of perceptible engine roughness. Often, I set the fuel flow to something in between these two extremes.
Once the mixture is set and everything is stable, I switch the engine analyzer to "normalize" and remain in that mode until it's time to descend. The normalize mode equalizes all the EGT bars at mid-scale, so that any variation becomes very obvious. Equally important, the normalize mode increases bar-graph display sensitivity four-fold (to 10°F per division), so that even small variations can be seen clearly. This is quite important, because small EGT excursions can be evidence of big engine problems.
|Badly warped and burned exhaust valve removed from cylinder #3 of a 1981 Cessna T210N (click for larger view). The cylinder measured 0/80 in a compression test.|
I recently received a perfect example of this from a friend who flies a 1981 Cessna T210N. When his airplane went in for annual inspection, the #3 cylinder had 0/80 compression and a borescope inspection revealed a badly warped and burned exhaust valve.
The owner downloaded the data from the memory of his JPI EDM-700 engine monitor, and found that the #3 cylinder had been exhibiting jittery EGT readings for some time -- the classic signature of a failing exhaust valve (see graphic below).
|Engine monitor download from 1981 Cessna T210N showing jittery EGT on cylinder #3, the classic signature of a failing exhaust valve. (Click for larger view.)|
It is interesting to note that the aircraft owner never noticed the jittery EGTs on cylinder #3 until he downloaded the engine monitor data. That's because he had not developed the habit to switching his JPI monitor to normalize mode during cruise. The jittery EGT was not apparent with the instrument display in its normal mode (40 °F per division), but would have been quite obvious had he switched the instrument to normalize mode (10 °F per division). Needless to say, the owner changed his cockpit procedures.
Descent and Landing
Prior to commencing descent from cruise altitude, I switch my engine monitor from normalize mode to normal mode. About five to 10 minutes from my destination, I start making incremental power reductions to cool the engine down prior to entering the traffic pattern or commencing the instrument approach.
During this phase, my engine-monitor focus is primarily on the displayed cooldown rate. Lycoming has published a recommendation that cylinders be cooled no faster than 50 °F/minute. I'm somewhat anal about cooldown, so I endeavor to hold my rate to 30°F/minute or less until all CHTs are down to 250 °F or less.
Any time I've installed a nifty new item of electronics on my instrument panel, I've found that I tend to get preoccupied with the new black box and let my scan go to hell. That certainly happened when I first put in the Garmin GNS530 and again when I added the Sandel SN3308. Sure enough, when first I installed my engine monitor two years ago, I found myself spending way too much time fixated on the new instrument.
Part of the solution to this problem lies in getting familiar and comfortable enough with the instrument that you can absorb its data display or change modes in just a few seconds. The other part is achieved by programming the unit's alarm limits so that it will demand your attention anytime something unusual is going on with your engine(s).
My JPI EDM-760 is mounted way over on the copilot's side of the panel along with the other engine instruments -- not the ideal location to be sure, but the best I could do without making massive changes to my panel layout. Fortunately, the JPI is equipped with a remote alarm output that can be connected to a remotely-mounted light or horn. I connected mine to a large amber annunciator mounted next to my airspeed indicator, putting it right in the heart of my flight instrument scan. When that light starts flashing, there's no way I can miss it.
As delivered, my JPI monitor came with its various alarm limits set so that the unit would not alarm except in relatively extreme conditions. I soon learned that it's better to reprogram the alarm limits to much more conservative values so that the monitor will get my attention whenever any engine parameter is even mildly out of the ordinary. Here's a bit more detail about how I've set up my instrument:
CHT: JPI's default high-CHT alarm limit is 450 °F. Because I endeavor always to limit my CHTs to 380 °F or less, I've programmed my monitor to alarm at 400°F. Any time the CHT alarm goes off, I take immediate corrective action -- generally enrichening the mixture if I'm ROP or leaning it further if I'm LOP.
Oil Temperature: JPI's default high oil temperature alarm limit is 230 °F. I know from experience that my oil temperatures normally run between 185 °F and 195°F, so I've programmed my monitor to alarm at 210 °F. There's also a low oil-temperature alarm limit that defaults to 90 °F; I've left mine at that value.
TIT: JPI's default high-TIT alarm limit is 1650 °F, which corresponds to the manufacturer's red-line limit for my turbochargers. I've reset my TIT alarm limit to an admittedly conservative 1600 °F.
EGT Difference: An extraordinarily valuable feature of the JPI monitor is that it calculates the difference between the highest and lowest EGT on each engine, and alarms if that difference gets too high. This is a good tool for detecting if a cylinder has gone cold or into heavy detonation or pre-ignition. JPI's default setting for the EGT difference alarm is 500 °F. Because I know from experience that my EGTs normally remain within about 60 °F of one another in flight, I've reprogrammed this alarm to a much more conservative 120 °F. The downside of this is that I will almost always get a "DIFF" alarm during ground operations; the upside is that I am confident of catching any significant in-flight cylinder anomaly very quickly.
Cooldown: As mentioned earlier, the JPI calculates CHT cooldown rate, and triggers an alarm if that rate becomes excessive. JPI's default cooldown alarm limit is 60 °F/minute. Lycoming's recommended limit is 50 °F/minute. I've set my alarm at a very conservative 30°F/minute.
Bus Voltage: The JPI has both low and high alarm-limits for bus voltage. For 28 V airplanes, JPI's default limits are 24 V and 32 V (or 12 V and 16 V for 14 V airplanes). My regulators are adjusted to maintain a bus voltage of 27.8 V plus or minus 0.2 V, so I've set my voltage alarm limits to 25.5 V (low) and 29.5 V (high).
Now, I am not suggesting that you use the same alarm limits I use in my Cessna T310R -- they may not be appropriate for your aircraft. What I am suggesting is that for each of these parameters, you should determine the normal operating range for your aircraft, and then program your engine monitor alarm limits just a little bit beyond that range. This will ensure that you get an early alert for any unusual or untoward event.
In addition to routine monitoring, a digital engine monitor permits you to perform various diagnostic tests that can be extraordinarily useful in accurately diagnosing engine problems:
Mixture Distribution Test
Also known as the "GAMI lean test," this procedure enables you to evaluate how much mixture variation exists among the cylinders of your engine. It is usually performed at about 65% cruise power at an altitude of 6,000 to 10,000 feet; if you have cowl flaps, they should be open.
Starting with a full-rich mixture, write down the EGT of each cylinder. Now lean very slowly until the first cylinder reaches peak EGT, and note peak EGT value for that cylinder and the fuel flow at which that peak was achieved. Continue leaning very, very slowly until each cylinder reaches peak EGT, and again write down the peak EGT value for each cylinder and the fuel flow at which each peak was achieved.
Once this data has been gathered, you can derive two valuable pieces of information. The first is the difference between full-rich EGT and peak EGT for each cylinder (referred to as the "lean range" for that cylinder), and the second is the difference in fuel flow between the first cylinder and the last cylinder to reach peak EGT (referred to as the "GAMI spread").
For most engines, the lean range of each cylinder -- the EGT rise from full-rich to peak -- should be around 250 °F to 300 °F. If any cylinder has a substantially lower lean range than the others, it may be operating too lean at takeoff power and might be vulnerable to overheating or detonation. (Suspect a clogged injector nozzle or an induction leak.)
The "GAMI spread" is a measure of uneven mixture distribution. The smaller the spread, the better. A fuel-injected engine with properly tuned fuel nozzles will exhibit a GAMI spread in the vicinity of 0.5 gallons per hour (GPH). Using stock nozzles, injected Lycoming and crossflow Continental engines typically have a spread around 1.0 GPH, and injected bottom-induction Continentals often have a spread of 1.5 GPH or more. Some carbureted engines (e.g., Continental O-470-series) can have spreads in the 2–3 GPH range. If your engine has a GAMI spread above 1.0 GPH, it probably won't be able to run smoothly at LOP mixtures.
In-Flight Mag Check
Every student pilot is taught to perform a mag check during the preflight engine runup, but many pilots have never performed an in-flight runup and are uncomfortable with the idea. That's unfortunate, because an in-flight mag check is an excellent diagnostic procedure, and a far more exacting test of the ignition system than the usual runup mag check.
In fact, the best and most revealing ignition system test you can perform is an in-flight mag check with the engine leaned aggressively LOP. The leaner the mixture, the more difficult it is to ignite. Therefore, if your ignition system performance is marginal, it will show up during a LOP in-flight mag check long before it becomes apparent in any other phase of operation.
If you have a digital engine monitor, the in-flight mag check should be done using the same procedure as previously described for a runup mag check. With the engine at normal cruise power and mixture (the leaner the better), switch the engine monitor to normalize mode, switch off one magneto at a time, and watch for all EGT bars to rise at least 50 °F. You should feel a perceptible loss of power with the engine running on one magneto, but it should continue to run smoothly.
Induction Leak Test
This in-flight test is an effective method for detecting leaks in the engine's induction system. It is best accomplished in level cruise flight at about 5,000 feet MSL. Start with a relatively high power setting -- full throttle for normally aspirated engines, or about 28" MP for turbocharged engines -- and full-rich mixture. Write down the EGT for each cylinder. Now throttle back the engine to reduce the MP to about 20" and again write down the EGT for each cylinder.
Disregard the absolute EGT values. Instead, calculate the change in EGT ("delta") for each cylinder between the high-power and low-power settings. Ideally, the amount of EGT change should be similar for all cylinders. If one cylinder (or two adjacent cylinders) exhibit(s) significantly less change than the others, suspect an induction system leak affecting that cylinder (or those adjacent cylinders).
Here's the principle behind this test: At the high-power setting, the induction manifold pressure is very close to outside ambient pressure, so any induction leak will have little or no effect on engine operation. At the lower-power setting, the manifold pressure is considerably lower than outside ambient, so any induction leak will cause the affected cylinder (or cylinders) to run substantially leaner than the others.
Modern digital engine monitors provide the capability for logging engine data in-flight, and then dumping the logged data for post-flight analysis. The amount of data that can be logged is a function of both instrument memory capacity and data sampling rate. For example, my JPI EDM-760 memory capacity can capture up to 17 hours of engine operation if data is sampled every six seconds (the default setting). It can also be configured to sample as often as every two seconds, but doing so reduces the memory capacity to less than 6 hours.
JPI and EI monitors are normally installed with a hardwired serial data jack. This permits logged data to be downloaded into any laptop computer or PDA via a standard RS232 serial port.
Insight's GEM and GEMINI monitors utilize an infrared (IrDA) link for downloading data. They require the use of an HP 200LX palmtop computer to receive the download. (The HP 200LX is obsolete and no longer in production, but can still be found on eBay.) Once downloaded into the palmtop, the data can then be transferred to a laptop or desktop computer by means of either a serial link or a flash memory card.
The downloaded data can be analyzed using general-purpose software such as Microsoft Excel, but there are also two excellent software packages especially designed for graphing and analyzing engine monitor data. EGTrends publishes a superb $150 package called "EGView" that can be used with all engine monitor makes and models.
J.P. Instruments offers its own graphing package called "EZTrends" that can be used with JPI's EDM-series engine monitors as well as other makes. While less sophisticated than EGView, the price is hard to beat (it's free).
See you next month.
Want to read more from Mike Busch? Check out the rest of his Savvy Aviator columns.