Engine Monitors

okay very good well any rate I'm I'm just back about a week ago from from my annual pilgrimage to AirVenture and this is a tonight it's going to be a slightly modified version of one of the eight talks that I gave it air venture - happily - standing-room-only crowds on the forums Plaza this is a fairly long presentation so I'm gonna try to get through it as quickly as I can so that we have a little time at the end for for Q&A and yeah I'm just advancing the slide just to make sure that that everything is working and it looks to me like it's it's coming up okay Mike I see it it's the next slide titled lots to discuss good we had a little problem with the with the June webinar a little technical problem so I want to make sure it didn't repeat anyway there's there's lots of stuff to talk about tonight and I'm going to try to get through it as quickly as I can we'll be talking starting off talking a little bit about the history of engine monitors the different generations of engine monitors and the key features that distinguish those generations I'll then talk a little bit about routine monitoring techniques that that what I recommend you do on every flight there using an engine monitor beginning with engine start and and concluding with the descendant landing will take go through the flight phase by phase by phase I'll talk a little bit about how I recommend that you set your alarms on your engine monitor alarms are one of the important features of most engine monitors - to make sure that that it gets your attention when something is is out of Tolerance and needs you to look at it we'll talk a little bit about the flight test profiles that we recommend flying in order to capture the most useful data on engine monitors we asked our managed maintenance clients to fly these profiles prior to you can you inspecting and anytime they suspect that there might be something wrong with the engine or the electrical system or whatever so that we can gather as much data as possible in the most efficient fashion and then I'll spend probably half the time tonight talking to you about how we use engine monitors to troubleshoot problems and just give you an idea of what the with the process we use to go through diagnosis by eliminating various causes that aren't consistent with the data until hopefully we've narrowed the problem down to just one or two possibilities so let's talk just for a little bit about the history of these things the very first things that I suppose you could call engine monitors although they didn't really don't rise to that level is when Alcor which is the original company that that introduced egt instrumentation for piston aircraft engines back in the in the 60s came out with an enhancement to their original analog egt gauge consisting of a rotary switch that would allow you to look at the eg T's on individual cylinders the originally GT gauges only had a single EGT probe located somewhere in the exhaust cluster but with the advent of this switch you could install individual egt probes near the exhaust port of each individual cylinder and then with the rotary switch select which of the four or six cylinders you wanted to look at pretty crude but it was the first baby step towards the engine monitors that we have today a little later on Alcor and another firm called the KS a introduced the first multi cylinder eg team instruments where you could actually watch all of the cylinders at once and again these were purely analog instruments they consisted of some vertical readout analog gauges that were cleverly stacked up and packaged in a three and an eighth inch instrument package and in fact 420 engine airplanes they even put a switch on there so that you could select whether you wanted to monitor the left engine of the right engine again these instruments only measured egt provided no CHT information and nowadays we know that CHT is really the most important thing to monitor they had no microprocessors in them they didn't provide any alarm capability they didn't log any data they were just a bunch of analog egt gauges all stacked up into a single instrument the very first things that we can properly call engine monitors although they still a little bit crude came with the introduction of two instruments one by insight the original jam or graphic engine monitor which provided a a pseudo analog bar graph using using an LED display and actually provided some microprocessor logic to do things like help you find peak egt when leaning and so on and another instrument from a company called Electronics international the u.s. eight ultimate scanner eight which provided the ability to scan up to sixteen channels of data typically all the eg T's and C HTS and a few other inputs but with a a digital readout were the absolute value of the of the temperature measurements were displayed on the instrument both of these instruments had microprocessors in them they and they had some processing capability they had some alarming capability very primitive but they did not again have any ability to log data into memory and then and then dump it for subsequent analysis so I call these the first generation engine monitors and they were a pretty big breakthrough and started us down the road to where we are today then in the early 70s a company called JP Instruments JP I introduced a device called the EDM 700 which turns out to be the best-selling engine monitor of all time and is in more airplanes than I think probably all other engine monitors combined today DDM 700 combined the the bar graph display from the insight Tim with the digital readout of the electronics instrument ultimate scanner and put those all into a single instrument and for the first time the EDM 700 also offered the capability of data logging originally it was an option later it became standard equipment but the instrument was actually able to capture data log it into memory and provided a way to to download that data into a laptop computer and then and then process it and analyze it subsequently electronics international quickly responded with a with a comparable instrument called the ubg 16 JP I actually in sight introduced a the first engine monitor for twins the insight Gemini JPI quickly responded to that by introducing their own twin-engine instrument called the EDM 760 which is basically two EDM 700s packaged into a single three 1/8 inch instrument and all of these all of these various competitive instruments would call second-generation monitors had the ability to log data to memory and and download it for subsequent analysis in recent years starting about five years ago we started to see the introduction of much fancier engine monitors that I call third-generation distinguished by having high-resolution flat panel color displays that allow them to display a whole lot more data it with much higher resolution than the old bar graph type LED displays many of these were certified for primary instrument replacement so that instead of just being a supplemental instrument they could actually replace the factory installed engine instruments which was a good thing because some of them are quite large a needle on a panel space and the only way to get that panel space is to rip some stuff out these elaborate flat panel displays these third-generation monitors tend to be quite a bit more expensive than the second generation supplemental instruments an EDM 700 can typically be installed for about 3,000 bucks whereas many of these third generation monitors can can run $10,000 or above but they're very very fancy and very very capable and a lot of people have been installing them and then finally in most new manufacturer aircraft we're seeing what I call fourth generation engine monitors where the engine monitor capability is no longer a separate instrument but is is packaged into the aircraft's multifunction display so this is something that we find in glass cockpit airplanes where the where the engine monitor capability is is part of the MFD rather than being a discrete separate instrument and this slide is showing both the avid I'm offering on the left and the Garmin g1000 on the right Garmin for some reason when they introduced the g1000 created a quite elaborate engine monitor page on the MFD but did not provide any capability to log or download the data it was some kind of lawyer thing and recently they've started changed being that they started offering the ability to to extract the data for analysis originally in the cirrus sr20 and sr22 then recently in all of the single-engine cessnas that are g1000 equipped and my understanding is that that on a model by model basis they're going to be providing enhancements to all of the other g1000 equipped airplanes so the data can be extracted that change turned out to be a software change only and required no hardware modification which is a good thing okay so much for the history of engine monitors let's talk a little bit about how we how we use them or how I recommend using them during the various phases of a routine flight and we'll start with wood engine start once you start the engine and power up your avionics including the engine monitor you want to make sure that all of the cylinders in the engine are are making egt immediately after engine start if you discover that that one or more cylinders don't start firing immediately which is will be very clear on the engine monitor because there'll be one or two egt bars that are missing and normally is also a company with with the noticeable engine roughness that is a that's the symptom of what what mechanics call morning sickness which indicates that a valve is sticking when the engine is cold morning sickness is quite common in Lycoming engines and some of the earlier design continental engines we don't often see it in the later design continental engines three sixty four seventy five twenty five fifty series engines because they use a different valve guide material but lycoming z' and some of the early Continentals we do see sticking valve problems on a fairly regular basis and the first clue that you have a sticking valve problem is normally this morning sickness if if you see that if you see that it takes a while for one or more cylinders to actually start combusting and generating a GT it's a condition that you're going to want to follow up on with your mechanic fairly quickly because if you ignore it and allow it to get worse eventually you can wind up having a stuck valve in flight and that can do some serious damage to the engine so you want to catch it and deal with it early and the engine monitor will make it easy to catch that early and and determine exactly which cylinder is that is involved and then the mechanic can can can check out the valve valves do something called a wobble check and if you have a sticking valve that generally can be resolved without removing the cylinder from the engine okay once we have the engine started and were taxiing out to the run-up area I like to perform a preliminary ignition check while I'm taxing it's basically the the same the same kind of thing that you normally do during run-up except except that that it's simply done it at taxi rpm rather than at a higher rpm so as I'm taxing out to the run-up area I'll normalize the egt display which means that all of the a GT bars get centered on the screen at the same height and the and the sensitivity of the e GT display is increased and then I'll just go through and and and cycle the mags the normal both left both right both for singles or just cycle each of the four mags which is in turn four twins and verify that when you operate when you turn off each magneto you see all of the egt T's rise all for all six depending on how many cylinders you have on your engine and that no EG T's fall or become erratic if you switch off a mag and you have one or more EPP bars fall instead of rise that's an early warning that you've got some non firing spark plugs and you may just want to skip the run-up and and and turn around and go back to the shop and have them check it out or you may want to to continue to the run-up area and investigate further if the preliminary ignition check is normal and I'm in a hurry sometimes I'll just go ahead and skip the normal run up and and just go ahead and take off if you have time it's always good to do a standard run-up but the taxi out jack gives you a way of objecting the ignition system early then once we get to the run-up area we will do a standard ignition check this time it will be at the normal run-up rpm for the Poh for Continentals mostly it's mostly there the run-up is done at 1700 rpm and for lycoming typically somewhere between 1,800 2,000 RPM whatever the the european lycoming guidance on on doing run ups has a change about 18 months ago and while your Poh probably tells you to do the run-up with the mixture full rich it's much better to do it lean and that's that's what like homing current guidance is in their most recent service bulletins I always do my my run ups lean to peak rpm the procedure for doing the ignition check is the same as we discussed for the preliminary taxi check except that it's done at a higher rpm again we normalize the egt display with by pushing a button on the on the engine monitor to to equalize all the bars at mid scale and crank up the sensitivity and then we go ahead and and and run the engine on on in single mag mode on one mag and then the other and when we operate in single magneto mode we want to verify that all the EG T's rise that none of them fall or become erratic and that the engine runs smoothly on each mag individually the Poh calls for for doing an RPM drop check to see how much the RPMs drop on a single mag but that's really an archaic test that was that that was written into the Poh back in the days when we didn't have engine monitors so if you do have an engine monitor I would I would ignore the tachometer completely and focus on the egt display again the acid test of four for an ignition check during run-up is that all the egt bars rise when you switch off a magneto none of them fall all of them remain stable and the engine runs smoothly if it meets all those criteria we really don't care what the RPM drop is then on on takeoff and this is not really that much of an engine monitor stuff but it just drives me nuts to see how a lot of pilots handle their takeoff because they'll they'll Jam the throttle all the way forward and and and do a major insult to the engine we want it we want to do the take-off the way the professionals do taxi in a position I normally set the brakes and throttle up slowly to about 50% power while holding the brakes scanning the gauges making sure everything is in the green in a twin-engine airplane you want to make sure that both engines come up together and there were no major splits on any of the any of the dual gauges ensure that the ETS are all coming up together and are all somewhere in the general ballpark of one another and then if everything looks good release the brakes start rolling down the runway and very slowly throttle up to a hundred percent power but move the throttle slowly warm up the cylinder slowly don't jam in the throttle it's very very tough on the equipment and it is not the way to to get good engine longevity as we're rolling down the runway we want to and and bringing up the throttle slowly to the throttle to to 100% power we want to scan the engine gauges again make sure everything's the grain if it's a twin make sure that there are no splits all EE GT bars should be reasonably even and in the right general ballpark for high compression engines normally aspirated engines designed to run on 100 low lead we normally expect eg T's on takeoff at takeoff power to be somewhere in the 1200s degree fahrenheit range the exact value is critical but it needs to be somewhere in the ballpark for lower compression engine certified for 480 octane fuel and for turbocharged engines which have a lower compression ratio we expect the eg T's to be higher than that typically in the 1300s or so at full takeoff power but that's just kind of a sanity check to make sure that that everything is working if if things don't look right abort the takeoff we don't want to take problems into the air once we break ground and start climbing we want to shift our focus from eg T's 2ch TS and for the balance of the flight CH T's will be the main thing that that we worry about we want to make sure that CH TS are relatively cool for legacy aircraft like the 1979 Cessna 310 that I fly with fairly inefficient cooling systems we want to keep all of our cylinder head temperatures below 400 degrees Fahrenheit at all times and I prefer to keep them below 380 if at all possible is to give a little bit of extra cushion for recent design aircraft like the the the Cessna corvallis or the Cirrus sr22 or the diamond da40 or other recently designed airplanes that have very very efficient cooling systems we want to limit our ch TS even more for for that generation of aircraft I can only recommend keeping the CH T's no higher than 380 degrees Fahrenheit and preferably keep them down at 360 or or less those the those targets that I've just given you have to be adjusted for certain factors like during extremely cold weather operation we want to limit the CHP s even more than the mat and in very high altitude operations typically in turbocharged airplanes when we're up at the flight levels and the air density is very low we may have to live with ch TS a little bit above that so we do need to adjust those targets a little bit for operating conditions but those are the general ballpark maximums that that we want to keep CH T's at if the CH T's get too hot we need to do something to bring them down if you're climbing lower the nose and increase the airspeed if you have cal flaps open them to get some more cooling air or the cylinders if you're if you're climbing rich of peak or cruising Richard peak and the CH PS get too hot you want to you want to enrich in to increase fuel flow and and bring the CH TS down conversely if you're cruising Lena peak and the CAHPS get too hot you'll want to lean some more to reduce fuel flow and and bring the the cylinder head temperatures down but we want to keep a very close eye on CH TS we want to set alarms to make sure that we're alerted to any high CH TS I'll talk about that in a few slides and when we see a CH T get too high we need to do something to bring it down in crews will will lean and and I'm not going to get into leaning procedure here I did a webinar on that called leaning basics it's available on the ei video site um we want to cruise with the egt display normalised uh which cranks up the HT sensitivity and levels all the bars at mid scale so that if anything goes wrong with the engine it will be really obvious and we want to continue to monitor CH T's and and and keep them in those zones that we talked about in legacy aircraft preferably at or below 380 fahrenheit in a modern design aircraft with very efficient to cooling systems lower maybe 360 degrees or so and adjusted as I said if necessary for cold weather operation or very high altitude operation and again if the CH TS get too hot we want to rich in if we're richer peak or lean if we're leaning a peak in order to bring the CH TS back down I like to do an ignition system stress test which I'll describe a little bit later on most flights I typically get in the habit of doing it towards the end of the cruise phase before I start down and we'll talk about how to do that but it's a it's a it's an in-flight magnets that's a much more discerning test of ignition system condition than anything we could do on the ground and I like to do it every every flight of every few flights just to keep tabs on the condition of my ignition system when it's time to to start descending as we approach our destination obviously lower the nose if you are flying a normally aspirated airplane without an altitude compensating fuel system which is the case of most normally aspirated engines as you descend the mixture is going to get leaner and leaner as the manifold pressure increases in the fuel flow remains the same so we're going to have to occasionally rich in the mixture manually as we descend and one good way to do that is to rich in it in order to maintain approximately constant exhaust gas temperature in the descent if you're flying a turbocharged airplane like mine or if you're flying an airplane with an altitude compensating fuel system like the cirrus sr20 or certain bonanzas that that are equipped with an l2 compensating fuel pump then the mixture change during the descent is is handled automatically by the equipment so you just have to lower the nose and you don't need to touch the mixture all the way down which is very convenient as we're doing our descent and especially when we get to the point where we start reducing power we want to monitor the cooldown rate most engine monitors will report cool-down right and if possible we want to limit the CHD cooldown rate to about thirty degrees Fahrenheit per minute or less in my experience it's pretty hard to shop cool cylinders and get above that cooldown rate unless you do something very dramatic like yank the throttle way back as long as you're reducing power fairly slowly shot cooling does not seem to be a problem but the engine monitor will monitor that for you and if you set an alarm it will warn you if cooldown rate is excessive it's not nearly as big a problem as as most people attend thing it's pretty hard to get a cool-down rate that's that's abusive let's talk a little bit about the alarm capability of engine monitors most modern engine monitors that are installed as supplementary equipment allow you to program the alarm levels for various parameters some of the engine monitors that that were instead are installed as primary replacement have their alarms hardwired to the manufacturers red lines which makes them pretty useless I mean the CHT red line for lycoming engine is 500 degrees Fahrenheit and if you get an alarm don't get an alarm until you get to 500 degrees that's too late you've already abused the engine pretty seriously I was actually talking to the president vice president of JPI at AirVenture two weeks ago and they recently received approval from the FAA to enhance their software on their primary replacement instruments to provide two sets of alarms a red alarm that's hired hardwired to the manufacturers red lines and what they call a white alarm that is user programmable and that's that that's a that's a great improvement because unless you can program the alarms that they they really are pretty useless they don't give you a warning and time to do anything about it any case here are some alarm settings that I recommend that you use if you're assuming that your engine monitor alarms are user programmable the CHT alarm I set mine at 390 degrees Fahrenheit so I get a little bit of warning before I get to 400 degrees and if you have a modern design airplane with a very efficient cooling system you might want to set the CHT alarm as low as say 370 degrees Fahrenheit for oil temperature the ideal oil temperature changes between 180 and 200 Gries Fahrenheit I set my high temple alarm well temple alarms at 210 and the low alarm to 90 degrees the low alarm is set fairly low because if you don't do that it'll scream at you pretty much during all ground operations and and then you'll start to ignore it so if you have a turbocharged airplane that has a TI T instrumentation I recommend you set the engine monitor TI T alarm to 50 degrees below the manufacturers red line typically the red lines of the TI t red lines on most airplanes are either 1650 or 1750 depending on what kind of turbocharger is installed and so I would recommend setting the red line I mean setting the alarm about 50 degrees below that on my airplane the TI T red line is 1650 and so I set my TI t alarm to go off when the TI t rises above 1600 some engine monitors will allow you to set an alarm based on the difference between the highest and lowest egt if your monitor has that capability I would recommend setting the alarm at about 150 degrees Fahrenheit for injected engines possibly 200 degrees Fahrenheit for carbureted engines again we want to set that alarm high enough that we don't get a lot of false alarms because if you do get a lot of false alarms you just start tuning them out and then when you get a real problem you may not notice it once again if you do have the ability to alarm on cooldown rate I recommend setting the alarm to go off when the cooldown rate exceeds 30 degrees Fahrenheit per minute and finally many monitors have the ability to alarm on when the bus voltage the aircraft bus voltage either gets too high or too low recommended alarm settings for bus voltage you have a 28 volt airplane would be 29 and a half volts on the high side and 25 and a half volts on the low side if if you have a 14 bold airplane the recommended alarm settings are 15 volts on the high side and 13 volts on the low side okay let's talk a little bit about flight test profiles we have a write up that we give to our savvy manage maintenance clients and I've also placed that same right up on the Savi analysis comm site that I'll talk to you about at the end of this webinar so you can go a download download the the flight test profile document from Savi analysis site and there are our three tests that we that we recommend that that are that our clients use or perform to maximize the the data that's captured by the engine monitor and give us the best chance to to troubleshoot engine problems the first one is a test of the ignition system that we call the ignition system stress test it's basically an in-flight Lena peak mag check as I mentioned I do this one on almost every flight and anytime that an any kind of an engine anomaly is suspected and we always ask our clients to do this prior to each annual inspection so that we can tell whether whether there were any getting ignition issues that need to be addressed during the annual the second test is a mixture distribution test otherwise called a gammy lien test because it was developed by a general aviation modifications Inc the the game ejected guys and again this is a test that we would recommend running every hundred hours or one year prior to each annual inspection to see whether it's necessary to clean nozzles or to or to tweak fuel nozzles and anytime engine anomalies are suspected and the last one is an induction leak test which we normally recommend doing only when anomalies are suspected we don't do that on a regular basis so let me just go through those those flight test profiles briefly and again the detailed documentation for them can be downloaded from our savvy analysis comm website that I'll give you the URL at the end of the at the end of the of the webinar we do the ignition system stress test in normal lien crews preferably aggressively Lena peak the the leaner you do this test the more demanding a test it is on the ignition system simply because lien mixtures are more difficult to ignite than rich mixture so the leaner the mixture the the the better the condition your ignition system will have to be in in order to pass the test and we really want this test to be as discriminating as possible if your engine monitor allows you to program the sampling rate we want you to crank it up to the maximum sampling rate that the monitor provides in order to get the as much data resolution as possible some engine monitors have a hardwired sampling rate and and some allow you to to program the sampling rate normalize the egt display again by pushing a button that levels all the egt bars and at the center of the of the display and then go through the normal mag check sequence both left both right both for singles and turn off each mag switch in sequence for twins and you when you do this test you want to remain on each individual mag for at least ten sample intervals so that we get enough data points to make sense of the data so if your engine monitor is sampling once per second you want to be on on the single mag for at least 10 seconds on the other hand if it's sampling once every 10 every 6 seconds which is for example a default rate for for most JPI engine monitors then you'll want to be on one mag for a full minute so that you get 10 samples before switching back to both and then and then going to the to the other mag so when we do this in flight mag tech we want to as usual verify that all egt bars rise that none of them fall that when they rise they remain fairly stable at their at their higher values normally egt will rise somewhere between 59 degrees Fahrenheit when you turn off a mag it is fairly normal for the odds cylinders and even cylinders to rise more than the than the others and it's fairly normal for se evens to rise when you operate on one mag and odds to rise more when you operate on the other mag that's not a problem that's kind of what we expect to see and make sure that when you're operating on each mag individually the engine is running smoothly with with no unacceptable roughness I use that phraseology because the engine will always run a little rougher on one mag than it does on - but if it's running rough enough that that your wife pokes you in the ribs and says what did you do that's probably too much so we want to just keep an eye out for the engine running noticeably rough on one mag or the other here's an example of the analysis of data dump from from a an in-flight magic or ignition stress test you'll notice that that the eg T's on the on the right mag look pretty good but on the Left mag you'll notice that the orange trace which is EDT number 5 is very unstable during the period that you were running only on the left mag so it's clear that the bottom plug on cylinder number five which is the one that's connected to the left mag is is not igniting the mixture reliably and is is is probably bad maybe it has an excessive gap maybe it's it's resistor is has an excessive value but there's something wrong with that plug and it either needs to be cleaned and gapped or maybe it needs to be replaced also notice in this particular match check that the eg T's rose more when we are operating on left mag only then they rose when we were operating on right mag only that normally that differential normally indicates that the two mags are not timed the same on most engines make the magneto timing for the left and right mag is supposed to be identical and if you see uneven Rises like this that normally will indicate that that the mag timing split and that the two mags are not firing exactly the same time so again that's something that that that you would want to check the timing or have your mechanic check the timing and and bring them right on spec the mixture distribution test or gammy lien test is normally performed at moderate cruise power typically sixty or sixty-five percent no more than that because we don't want to generate abusive temperatures when we do this test the airplane should be operating at or near wide open throttle in a normally aspirated airplane that means climbing up high enough that you can have the throttle all the way in without exceeding 60 or 65 percent power so it's best to perform it at a relatively high cruise altitude and again if the engine monitor has a programmable some sample rate we want to crank it up to the maximum right so we get as much data resolution as we possibly can then starting with the mixture full rich we want to do a mixture sweep we want to lean very very slowly and very very smoothly until all the cylinders have passed peaky GT and are well into the lean of peak area we normally ask our clients to to lean very very slowly until the engine gets to the point where it's running quite rough so we know all of the cylinders are on the lean side of Pete and as you are doing this if your engine monitor records fuel flow then you don't need to do anything more than this mixture suite because everything is being recorded by the instrument if your engine monitor does not record fuel flow then as you perform this mixture sweep that's very slow mixture sweep from full rich to very lean will need you to manually record the exact fuel flow at which each individual cylinder reaches peak egt again if the engine monitor captures fuel flow you don't need to do it manually because the engine monitor is capturing the data and that's very convenient um then we'll ask you to repeat that procedure several times to make sure that we get several good sweeps normally we will will run from full rich - very lean back to full rich then - very lean back to full rich and do that several times very very slowly and very very smoothly and need any jerkiness on the mixture control is going to mess up the data pretty badly so it has to be very slow and very smooth and once we're done gathering the data then we can calculate what's known as the Gami spread which is the difference between the this slide as a as an error but it's basically the difference between the fuel flow on the first cylinder to peak the the leanest cylinder and the fuel flow on the last cylinder to peak which is the richest cylinder and let me illustrate that here with with data that's been been captured this particular engine monitor did record fuel flow the fuel flow line is that blue line that descends from left to right the other six lines are the egt lines and you'll notice that the very that as we lean the engine in other words reduce the fuel flow from rich to to lean the very first egt to reach peak is uh is the cyan a trace which is cylinder number three which reached a peak egt at a fuel flow of thirteen point four gallons an hour that means number three cylinder is the leanest cylinder so it it CGT reached peak first as we were leaning the engine the next cylinder to peak is number four which peaks at thirteen point one gallons an hour and the other four cylinders all peak pretty much together at twelve point nine gallons an hour so those four cylinders are the richest of the cylinders number three is a lean outlier compared to those other four and then we we note the difference in fuel flow between the lena cylinder the first one to peak and the richest cylinder the last ones to peak in this case four of them peak together and we see that the difference in fuel flow is a half a gallon an hour which is a very respectable Gami spread generally for engines that that are operated Lena peak we want the Gami spread to be a half gallon an hour or less if the Gami spread is a gallon an hour or more then the mixture distribution is poor and generally speaking the engine will not run very well Lena peak it'll start running quite rough when you get on the lean side peak and if the Gami spread is somewhere be one gallon an hour and a half a gallon an hour then it's it's mediocre you may be able to live with it but generally we would try to first clean the injectors and if that doesn't solve the problem then tweak the injectors in order to get the big a mean spread down to a half gallon now or less if the engine is carbureted rather than injected there are several things we can do to improve the gamma spread and notably use some partial carb heat which tends to to improve the mixture distribution on carbureted engines but anyway that's the the mixture distribution test or gamma lean test and and how it's done if the engine monitor didn't record fuel flow then you would have to be reporting it by hand and finally the induction leak test we we set up the airplane and cruise at a fairly low altitude with a full rich mixture and then we run two tests the first is the high manifold pressure test which for normally aspirated airplane would be wide open throttle for turbocharged airplane we would set the throttle so that the manifold pressure is approximately equal to outside ambient pressure and then write down the egt for each cylinder or we'll be able to see it on the engine monitor data if you if you dump the data then we reduce manifold pressure by about 10 inches and then again record the eg T's for each cylinder or or let the engine monitor record them and look at them afterwards and then we calculate the amount of egt change for each cylinder from the high manifold pressure run to the low manifold pressure run low where we reduced manifold pressure by about 10 inches and the deltas should all be about the same for all the cylinders and if one or two cylinders have a much larger excuse me a much significantly smaller eg T change than the other ones that's an indication that we have an induction leak in the vicinity of that cylinder or of those cylinders so that's the induction leak test we normally only do this when we suspect that there might be an induction okay at this point I'm going to transition into talking about how we use engine monitor data for troubleshooting and show you a bunch of problems and talk to you a little bit about how we figure out what what the problems are because with little practice you will find that that engine monitor data is an astonishingly powerful tool for for diagnosing various engine problems let me start out with this one this is a data from assessment t 210 it was about a four hour flight and you'll notice the number three egt trace which is the one that's kind of gray starts going more and more unstable as the flight progresses and towards the end of the flight it's it's it's it's quite unstable now at first glance you might think that this is an indication of of failing eg T probe or maybe a loose connection on that EGT probe but if we take a closer look at that unstable layer here you'll notice a couple of interesting things you'll notice that the variation in egt is very slow and almost perfectly rhythmic the egt went up and down 20 times in 30 minutes and it went up and down 10 times in the first 15 minutes and 10 times in the second 15 minutes in other words this egt oscillation is extremely regular and extremely rhythmic and it's very very slow these this is not cycling up and down rapidly like you would expect if you had a loose connection its cycling up and down a little less than one cycle per minute very very slow and very rhythmic you'll also notice that the amplitude of the variations is only about 30 degrees Fahrenheit peak to peak this is on a 1500 degree roughly egt value so it's a very small percentage change and it's so small that you'd never notice it on the engine monitor unless the engine monitor were in in normalized mode where the egt bars are much more sensitive so it's in order to be able to detect something like this it's very important that whenever you're in cruise you operate the instrument in normalized mode otherwise you would never even see a situation like this you'd never even notice it what what is the cause of these egt variations well it turns out that there's only one thing and one thing only that can produce slow rhythmic variations in egt with a frequency somewhere around one cycle per minute or so and that is a burr or an exhaust valve and this is the exhaust valve that came out of that session to teach you 10 and you'll notice on the left side that that the valve has a very severe hotspot is that that area on in in the roughly the the 3 o'clock position where the valve was running so hot that literally all the exhaust deposits were burned right off of the valve we could see this very clearly if we looked into that cylinder with a borescope and took a look at the valve in the upper right photo of the valve where we're looking at it in cross-section you'll notice how badly warped the valve is and in the lower right part of this photo again a different view the same valve you'll notice that the right side of the valve there's a nice shiny ring that is that shows that the valve was in good contact with the valve seat and the valve seat was was burnishing the the the the backside of the valve and and creating that nice shiny ring we in a good valve we should see that shiny ring all the way around the circumference of the valve but on the left side you'll notice that it's dull and that there's a tremendous amount of metal erosion in fact there's so much metal erosion that that particular edge of the valve has almost gotten to to to a knife-like thinness and had this valve remained in service very much longer that edge that knife edge would would would break off and the cylinder would shut down completely and and you'd be flying a 5 cylinder engine and probably declare an emergency but this is what an exhaust valve burned exhaust valve looks like five or so hours before it would actually fail and this condition showed up very clearly on the engine monitor if the owner of the pilot knows what to look for which you now do and and keep the engine monitor in normalize mode so that you can see things like this again the unique signature of a burnt exhaust valve is a low amplitude egt oscillation that is very slow and very rhythmic typically the variation is only about 30 to 50 degrees Fahrenheit so unless the engine monitor is in normalize mode you won't notice it in real-time typically the oscillation is somewhere in the vicinity of one or two or two cycles per minute quite slow and here's a really interesting example from an f3 33 a bonanza a friend of mine Eliot Schiffman gave me these these traces from his bonanza and it traces the progression of a failing number to exhaust valve over a period of several months and as you as as you go to the later and late months you'll see that the oscillation in the egt on cylinder number two the red trace becomes increasingly rhythmic and increasing and increasing in amplitude until on the final trace before the cylinder got pulled the the oscillation was almost a perfect square wave with a frequency of literally exactly one cycle per minute almost as if it came out of the signal generator it's just amazing and this is the exhaust valve in its last stages of burning right before the valve would fail and of course the Elliott noticed this on his engine monitor verified it with a borescope had the cylinder pulled and the valve repaired before it it actually failed in flight what happens if you let things go too far well this is a an actual valve that came out of my own Cessna 310 about 20 years ago back in the bad old days before we had engine monitors and before we used Bora scopes and where these this kind of failure was relatively routine and as you can see a piece of junk broke off of the valve and and the cylinder shut down this is the sort of thing that was fairly common 20 years ago but today there's really no excuse for it happening because we we have the technology to detect it well before failure both through borescope inspection which typically gives you a hundred hours or more warning before failure and the engine monitor readouts which if you know what you're looking for will give you a heads up twenty or twenty five hours before the the valve actually gets to the point of fail this this next series of engine monitor data illustrates probably the most dramatically violent and destructive thing that can happen to an aircraft engine and that's a pre-ignition event this particular one happened in an SR cirrus sr20 that that my company manages fairly uh and it occurred just this last January the the pilot had flown the airplane up to Santa Barbara for $100 hamburger turned out to be one of the most expensive hamburgers of his aviation career by the time it was over when he took off out of Santa Barbara to head back for home he got a CH T alarm hi CH T alarm on his engine monitor and felt a little roughness and very wisely decided to quickly throttle back declare an emergency to Santa Barbara Tower and put the airplane back on the ground promptly which was was quite a good thing if you take a look at this engine monitor data it's very dramatic he applies takeoff power all the CH T's which are the upper graph these graphs by the way are from our savvy analysis dot-com site that I'll talk to you about at the end of the webinar but you'll see that all of the CH T started to come up together and then very suddenly the CH T in cylinder number three started running away very very rapidly it rose from three hundred and seventy four degrees which is a normal CH T on takeoff all the way up to six hundred and fifty three degrees which is the hottest CHT I think I've ever seen recorded on any airplane and it did it in a period of less than two minutes it happened very very quickly at at that point the pilot the throttled back aggressively but not quite in time to save that cylinder all I'll show you that later when my account manager first looked at this data he sent me an email asked me to look at it and said I'm having a hard time believing that this is real data because I've never seen a CHT of 653 degrees I didn't even think that was possible and I'll have to admit it was the highest CHP I'd ever seen recorded as well but when I took a look at the data it was very obvious that this this dáil data was real it was not an instrumentation problem and the way you can immediately tell it was real is that it was confirmed by a big anomaly in the egt of cylinder number three so it couldn't just be a problem with a bad CHT probe or a bad connection or something like that we knew immediately that it was real we we told the shop to to pull the airplane in and do a borescope inspection of cylinder number three and I told the shop to expect that that that that the piston and cylinder number three was probably melted and in fact the piston was severely melted around about 2/3 of its circumference the the number one compression ring was was compromised and pretty much welded into place it was a was a real a real mess fortunately the damage was limited to the number took to that particular piston and cylinder and although there was a tremendous amount aluminum in the oil filter from the melted piston we were able to verify that the filter caught all the metal and we were able to just replace the piston and cylinder and and not have to tear down the engine so the fact that the the pilot pulled up pull the power back quickly was really helped save the engine here's another one in and another sr20 this one was actually an airplane in England and it was it was sort of interesting again it happened on takeoff in this case it was cylinder number five that had a thermal runaway the CHP got up to five degrees and then flat-topped and the reason a flattop was not that it really did that but this particular series had an avid I'm MFD with the engine monitor function in it and for reasons known only to avid I'm their software doesn't register CH PS above 500 degrees it assumes that that's as high as it can get we're guessing that the CH T actually got up to somewhere above 600 degrees but the engine monitor didn't record the actual temperature just because of a software issue with the Aventine software again from the time that the CH T that number 5 CH T got over 400 degrees Fahrenheit to the time that it peaked and actually melted the piston was less than 2 minutes the CH T again rose at a very rapid rate of more than one degree per second here's another similar run away this one is quite complicated to look at it it came out of a bonanza but again CH T 5 ran away shortly after takeoff until it melted a hole in the piston this one rose so rapidly that that the piston was destroyed inland in about one minute after after the CH T reached 400 degrees and that was that was the result this this engine was was trashed and had to be torn down so the moral of the story on on these violent events we want to set the C HT alarm at no higher than 400 degrees I said - I mentioned at 390 and if the CH T exceeds 400 you want to act immediately to prevent it from going any higher and if it continues to rise rapidly and reaches 420 degrees you want to throttle back to idle immediately to brake the run to break the run away cuz you want to stop this phenomenon before before something melts and then once you've once you've broken the the thermal runaway and the CHT start to come down then bring in just enough power to keep the airplane aloft and put the airplane down at the very next Airport and pull the top spark plug stick a borescope in the cylinder to see whether whether there's damage to the piston or the combustion chamber but these are very violent events they happen very rapidly and you have to respond to them almost immediately you have maybe 30 seconds to respond to this sort of thermal runaway in order to prevent damage to the engine and if you don't respond very rapidly and have an alarm that gets your attention immediately you're going to at least lose the cylinder and very likely lose the whole engine finally I'm going to take you through a couple of a couple of diagnosis situations to illustrate the the way we we do what's called a differential diagnosis by by systematically eliminating possibilities and until we've we've narrowed the the cause of a problem down and I'll just show you a couple of examples or very quickly to illustrate kind of the way this this works I'll start off with them this one that depicts about one minute of of engine operation the particular minute that it illustrates is the point where the pilot leveled off in crews and decided to do what we call a big mixture pull to transition from a richer peak climb mixture to a lean of peak cruise mixture something he had done many many times before uneventful II in this case when he pulled back the mixture to Lena peak the engine started running rough and two of the eg to the 6eg T's cylinders number two and four started dropping rapidly it was pretty obvious that those two cylinders had stopped producing power he rich in the mixture a bit and the the two cylinders came back so at the end of the flight he captured the data and and sent it to me to see if I could figure out what was wrong so how do you figure out what's what's wrong in a situation like this well you figure it out by the process of Illumina of elimination it takes three things for a cylinder produced combustion fuel air and spark air is typically never a problem when only one or two cylinders are affected because almost anything that would prevent air from getting into the engine would prevent air from getting to all the cylinders at once so typically we need to figure out whether this is a fuel problem or an ignition problem so it's possible that that the cylinders to enforce toward to flame out because they were running leaner than the other cylinders or it's possible that they started to flame out because there was a problem with their spark plugs or something that caused an ignition problem again remembering that it's much harder to ignite a lean mixture than it is a rich mixture so if there's a marginal ignition problem it's going to show up with a lean mixture and it might look fine with a rich mixture so how do we figure out which one that is well this one's fairly easy if we take a look at the what the eg T's did when he did that big mixture pull from Richard Piech Galena peak we can see that all six cylinders reached peak egt at just about the same time which suggests that all of the cylinders were running roughly the same mixture if two and four were running leaner than the other cylinders then they would have reached peak before the other cylinders and it appears that's not happening so if we again eliminate a fuel problem like like a clogged injector or something by the process of elimination this needs to be an ignition problem and so in fact I told the owner that it looked like an ignition problem and he needed to check the spark plugs in cylinders 2 & 4 he he actually had those spark plugs replaced and the problem went away it would have been better had he reacted to this anomaly in flight by doing an ignition stress test an in-flight Lena peak mag check because that would have verified that it was an ignition problem and would have probably let us know whether it was the topper mount of spark plug that was the culprit but he didn't do that any rate we figured out that it had to be ignition problem simply because all the symptoms indicated that that fuel was was fun here's another more complicated scenario this is about 20 minutes of of a flight again it starts where the pilot levels often cruise and transitions from Richard Piech to Leena peak everything looks fine at first but over the the succeeding seven or eight minutes the egt for the number five cylinder the orange trace starts creeping higher and higher and the purple trace which is the turbine Inlet temperature creeps up along with it eventually the pilot notices this sort of panics and shoves the mixture control full rich of course all of the EG T's come crashing down when he does that and he decides that that was a little bit of an overreaction so he leans a little bit to a reasonable Richard peak cruise mixture everything seems fine and then later he he leans back to a Lena peak mixture everything still looks looks okay to him but he was sufficiently unnerved by this experience that that he dumped the data and and sent it in to be analyzed to figure out what was wrong this was a fairly complicated and fairly unusual case so you know I looked at it to try to get some sort of Clou is what could be wrong and again we're trying to figure out is this an ignition problem this is a fuel problem what's what would cause this anomaly well the first thing I noticed was that during that seven or eight minutes where the number-5 egt was climbing higher and higher the number 5 CH T was moving lower and lower so the two CH T and EG t were moving in opposite directions all of the other EG T's and CH TS were relatively constant I know from experience and this is a basic rule of thumb that any time we see eg T rise and CH T fall that is almost always indicative of an ignition system problem of a spark plug that's not firing again it would have been good if the if the pilot confirmed this with an ignition stress test he could figure out exactly which spark plug was was non firing but he didn't do that in any case I indicated to him that one of the spark plugs in number 5 was was bad and needed to be replaced again a general rule of thumb that really helps a lot when we're doing this sort of diagnosis that if we have a fuel related problem say an injector that's clogging up or something like that fuel related problems almost always cause CH T mu G T to move in the same direction they might both move up they might both move down depending on whether you're operating Richard Piech or Lena peak at the time that the problem occurs but they virtually always move in the same direction on the other hand if we have an ignition related problem that almost always causes eg T to rise and CH T to fall so that's a real easy rule of thumb that frequently helps us distinguish between fuel related problems and ignition related problems but as I was looking at this same data I noticed something else kind of interesting and and and that was that this number 5 CH T was sometimes the highest CH T of all six cylinders and at other times it was the lowest CH T of all the cylinders that's kind of unusual because generally speaking the the rank of the cylinder ch TS tends to remain pretty constant the hottest cylinder is usually the hottest and the coolest is usually the coolest in this case all of the ranks remain pretty much the same except for number five and number five kept trading positions between being the hottest and being the coolest and if you look at it you'll notice that what the pattern is anytime the pilot was operating this engine rich a peak the number 5 CH T was the hottest any time he was running Lena peak the number 5 CH T was the coolest well that clearly indicates that the number 5 cylinder most likely is running leaner than the other cylinders when richer peak it's it's leaner so the CH T is hotter when Lena peak it's leaner than the other cylinders so so the CH T is cooler because because the at the leaner mixture the cylinder is producing less power now how can we verify that diagnosis well we can take a look as we did before at the point where the pilot transition from richer peak - Lena peak and if we blow that up we can see very clearly that cylinder number 5 peaked well before the other 5 cylinders so cylinder number 5 is running considerably leaner than the rest of the cylinders so either the number 5 fuel nozzle is dirty or the number 5 fuel nozzle is the wrong size so I recommend it to the owner that he had number 5 fuel nozzle claimed and and do a lean test and if the number-5 cylinder was still leaner than the others after cleaning the nozzle that he needed to install one larger sized nozzle in the number five position to get its mixture in line with the other four five again as a general rule of thumb if the CHT rank of a cylinder changes between Richard Piech and Lena peak that usually indicates a mixture imbalance and we can confirm that by doing a gamine test if a cylinder has is the hottest when operating Rina richa peak and coolest when operating Lena peak then then then that cylinder is running too lean on the other hand if a cylinder has the is hottest all the time both when Richard Piech and Lena peak then there's normally a cooling error issue and we need to inspect the baffles and if all cylinders have excessive cylinder head temperature we would suspect that the ignition timing is advanced because advanced ignition timing will cause all cylinders to have high CHT and all cylinders to have low egt so again from the engine monitor data we can easily determine what the cause of the problem is here's a situation where where the number three egt suddenly rose by about a hundred degrees Fahrenheit and the number three CHT became lower than the other cylinders and then all of a sudden all by itself the situation corrected itself the number three egt went back down to normal and the number three CHT went back up again we've got egt and CHT moving in opposite directions so it indicates an on firing spark plug in this particular case the spark plug was fouled and then it spontaneously got unfiled which is a fairly common occurrence and we see that sort of thing all the time here's a case where we are seeing number one CHT with kinds of noise on it if we look closer we'll see that that noise is not slow and rhythmic it's it's rapid and random and this is an indication of either a failing probe or a bad connection CHT cannot rise and fall rapidly the way egt can the cylinder head has a lot of thermal mass EHT changes quite slowly so if you see CHT changing rapidly like this that violates the laws of physics and it can't possibly be true it has to be an indication problem again if an engine monitor if engine monitor data appears to defy the laws of physics it's probably not real it's probably an instrumentation issue here's a case with eg T's that are showing spikes in the downward direction CHT looks normal so we're pretty sure that these spikes aren't real if we look at them more closely they're quite random there's nothing rhythmic about them and we can conclude that this is a problem with the number three EGT probe or possibly a bad connection at the number three EGT probe where it connects to the harness just here's a funny one we're on takeoff egt suddenly shoot up way too high they're way up over 69 degrees Fahrenheit which we don't want and then suddenly return to normal if we take a look at the fuel flow and this engine monitor recorded fuel flow we'll discover that the initial fuel flow on takeoff was way too low and then suddenly it went up ten gallons an hour to 29 gallons an hour which is about normal for this engine and then subsequently had the pilot reduced power and lean and it went back down to 12 gallons an hour which is normal for Lena peak cruise what happened the pilot took off and forgot to push the mixture in and then he realized his error and shoved it in that's when the fuel flow suddenly jumped up by 10 gallons an hour and the EEG T's went back to what they ought to be and finally this one is is really interesting this is a from my friend Elliot shiffman's of Bonanza f33 a where he ran a test flight where for the first I think eight minutes or so of the flight he was running the engine at 27 inches of malleable pressure and 2100 rpm Lena peak and then he transitioned to 21 inches of manifold pressure and 2,500 rpm the first setting was I guess what many people would call way over square and the second one was way under square because the fuel flow was exactly the same for both both segments of the flight we know that the engine was putting out exactly the same amount of power it's consuming exactly the same amount of ab gas with exactly the same number of latent BTUs per hour but look at the difference in the CH T's in egt these the the CH T's are about 50 degrees higher in the second part of the run the eg T's are about a hundred degrees higher in the second part of the run why would that be well it's that that's the case because operating at high manifold pressure low rpm is always much more efficient than doing it the other way around the CH T's are higher with the high rpm operation because friction varies with the square of rpm and so it at 2500 rpm the the friction is something like thirty some-odd percent greater than at 2100 rpm and the e GT was higher because at the higher rpm there's less time between the time the spark plug fires and the time that the exhaust valve opens and so a lot of the combustion event actually just gets wasted out the back door and yet doesn't get turned into into useful mechanical energy turning the prop in creating airspeed so the moral of this story and I talked about this in the leaning webinar as well is that whenever you have the choice always use maximum and manifold pressure and minimum rpm rather than the other way around this is probably the exact opposite of what your flight instructor taught you but it's the best way to run the engines and it's the way I always run my engine engines and minor at 200 and some odd percent of TVO's so obviously I'm not hurting anything just to conclude engine monitors are an amazingly powerful way of diagnosing engine problems suggest that that you dump your engine monitor data before every annual inspection and anytime you suspect any kind of an engine anomaly certainly before permitting any mechanic Terr to perform exploratory surgery on your engine like pull cylinders we don't want to diagnose engines through surgery we want to diagnose them through data analysis we want radiologists not surgeons to be doing our diagnosis if you if you can can figure out what's wrong yourself that's great it takes a little practice our savvy analysis site as the most powerful tools available for for analyzing this data and it handles data from virtually every brand of engine monitor ever built it's a free site we encourage you to use it to upload your data there and to use all the neat tools for analyzing the data and if you can't figure out what the data is trying to tell you then have somebody who has some expertise in this area take a look at it the savvy analysis dot-com site went on stream just before the 1st of July I announced it at at AirVenture and all my talks on the forums Plaza it's a free I encourage you to use it it's the most powerful data analysis platform that's ever been created it's completely web-based so it'll run on anything that has a browser max pcs iPads whatever and and if you go to the website eight you can just create a free account by just giving your your name and email address and the password you want to use and then you can log into the site upload your your engine monitor data analysis or edging monitor data that you download from your engine monitor and use all easy tools to analyze if there's a whole bunch of documentation on the site there's the flight test profiles all sorts of there's instructions for how to download engine monitor data from from many different kinds of engine monitors all sorts of great stuff if you're interested in engine monitor data analysis to be had at the site and once again it's free now it's beginning sometime in September I don't we will be starting to offer professional analysis services if if you want somebody to take a look at the data it'll be done on an annual subscription basis the professional analysis of course won't be free but it'll be very modestly priced but but the platform and all the tools and stuff are free and will continue to be free forever no ads no gimmicks it's we just want people to start using engine monitor data analysis instead of letting mechanics tear engines apart and with that Tam and I'm sorry this this ran a little bit long but but it was a long presentation if we have a little time I'd like to go ahead and open it up up for questions I'll leave my contact information here on the screen so if there are any questions that we don't get to during this session feel free to to email me and I'll try to answer every question and there are several URLs including the one for the the savvy analysis site great Mike thank you so much that was a excellent presentation very informative thank you for your time and sharing that knowledge and information you have about engine monitors we have gone 10 minutes over our allotted time so we won't be able to take questions tonight but as you said to anybody that would like to contact you will have your email contact up on that screen page right there so with that Mike I'd like to thank you again for your participation tonight very informative as always and for everyone in the audience thank you so much for attending and I hope to have you tune in to another via a webinar sometime in the future thanks everyone and have a good night thanks everybody

Channel: Savvy Aviation

Views: 17,817

Rating: 4.7894735 out of 5

Keywords: Lean-find, Cylinder, A&P, Mike Busch, general aviation, Busch, Lean of Peak, JPI, Mike, Continental, GAMI, Lycoming, Savvy Aviator, EI, Savvy Aviation, EGT, Insight, CHT, EDM, aircraft mechanic, engine monitor, Grand Rapids

Id: 0SvTESXqidM

Length: 89min 3sec (5343 seconds)

Published: Thu Dec 08 2016