Vance Harral

The advantages of a feathering prop are a tradeoff against complexity, and new potential problems. Even if the feathering mechanism was free and weightless, some designers and owners might still choose against having one in a piston single. In a piston prop twin, single engine service ceiling and - more importantly - controllability, are so severely effected by a windmilling prop that feathering capability is a no-brainer even with the complexity and potential problems it brings. But in a single, the only thing you're buying with a feathering prop is a slight improvement in glide range, in the event of a total engine failure that doesn't affect the ability of the propeller to rotate. Note that feathering in a single doesn't help you at all if the engine is still making partial power (often the case, and in that case you want all the thrust you can get); or if it seizes suddenly enough that you don't get it feathered before rotation stops. In exchange for the benefit of feathering in the small number of cases where it actually helps, you get all the failure modes of the feathering hardware . Those include sticking of the anti-feather shutdown pins, such that the engine feathers on shutdown and is hard on the engine and battery on the subsequent restart; as well as the unlikely-but-catastrophic case of accidentally feathering the propeller in flight at full power, which tends to damage the engine. And because it would be very high risk to practice feathering and securing the only engine in a single-engine airplane in flight, you'll never train to do this "in real life", only in a simulator or on the ground. Most of my multi-engine students have a pretty difficult time getting the feather/secure sequence exactly right on their first "real life" attempt, and I suspect the error rate would be even higher amongst pilots of single-engine feathering propellers, who do not hold a multi-engine rating and regularly train the feathering sequence in multi-engine airplanes. I'm not really arguing that a feathering prop in a piston single is a bad idea. Just that it's not the no-brainer it might seem to be at first glance, even setting aside the large cost and small weight penalties.

Coincidentally, this week's update of the AOPA McSpadden report re-organizes the data in a way that specifically breaks out collisions, including by phase of flight. Here's the fixed wing, non-commercial data for the most recently available full year (2022): https://www.aopa.org/training-and-safety/air-safety-institute/accident-analysis/richard-g-mcspadden-report/mcspadden-report-figure-view/?category=all&year=2022&condition=all&report=true For those who don't want to pore over the charts, here's a quick breakdown of collisions in 2008 vs. 2022, which arguably spans the time period when in-cockpit traffic displays became ubiquitous: landing: 2 nonfatal in 2008, 1 nonfatal in 2022 takeoff and climb: 1 nonfatal in 2008, 1 nonfatal in 2022 maneuvering: 1/2 nonfatal/fatal in 2008, 1 fatal in 2022 descent and approach: 4/2 nonfatal/fatal in 2008, 2 fatal in 2022 enroute: 1 fatal in 2008, 1 fatal in 2022 taxi: 13/1 nonfatal/fatal in 2008, 10/0 nonfatal/fatal collisions in 2022 It's interesting to note that the overwhelming majority of collisions occur while taxiing. It's unclear to me if the landing and takeoff collisions occurred on the runway surface, or just after takeoff/just before landing, but I'm guessing they're probably on the surface, especially given that... ...there is an important note in the data stating that each aircraft involved in a collision is counted separately in the data. e.g. a classic "midair" is reported as two incidents, and any data point showing only one incident involves a collision with something other than another aircraft (probably a tower or cable in the air, probably lights and signs on the ground). So I think - my guess only - that the midair collision concern that traffic displays address, is contained in the fatal maneuvering, descent/approach, and enroute data, where there are an even number of incidents. That's 4 incidents (2 collisions) in 2008, and 2 incidents (1 collision) in 2022. It's debatable whether traffic displays have driven a 50% reduction in midairs, or if the numbers are so small that it's just noise in the data. It would be interesting to create a separate graph for every year to see how things move around. But what's indisputable is that that the average pilot is ten times as likely to kill themselves due to loss of control (the breakout says that in the descent and approach phase, for example, there were 29 fatals in 2008 and 17 fatals in 2022), as they are to die in a midair. I think one should plan their risk management focus appropriately. All this only applies to the average pilot, though - not gods of the sky who would never lose control of a perfectly good airplane. For that latter group, obviously "the other guy" is their biggest risk, and it makes sense for them to focus a lot of attention on midairs, and little on stick and rudder. Up to each of you to decide which group you're in.

Old instrument instructor's axiom: the more ridiculous the hood looks on you, the better it works. Anything that looks like a cool pair of shades is marginal at best. We're jiffyhood fans ourselves, just pull the strap over your headset. Trivial to raise it up at the end of an approach when you "break out".

Sure. A couple of months ago I was asked to assist with recurrent training for a client who was returning to aviation after a long hiatus. Despite the layoff, he was a good stick, and flew the pattern nicely. But only after I took his iPad away. On the first downwind leg, he has his head absolutely buried in the iPad, and was extremely concerned about a couple of other airplanes on or entering the downwind. I s**t you not, he spent more than 50% of the flight time on the entire downwind leg with his head in his lap (I know because I watched his eyes, like I do with all students). Meanwhile, he allowed the airplane to accelerate to over 100 KIAS (this was a 172), climb almost 200' above traffic pattern altitude, and was unable to conduct a stabilized approach. Again, he had no problem maintaining reasonable speeds and altitude profies in the pattern once I took the iPad away. When I asked him about this, he said he felt it was important to get acquainted with the "new technology", and talked about how much safer he felt with it. But no one had ever given him any training on how to use it, and he was stunned to hear me say he was making himself less safe. I did my best to do so politely, using some of the data discussed here. He finished the checkout with another instructor, and it's unclear to me what impact my training had on him. Another: last week I was established on downwind with a student in a specific training scenario (power-off 180 in a significant crosswind), when an airplane from the flight school down the road reported inbound on the VOR-A approach to my home drome. I'm very familiar with the approach. The MDA is 600' AGL, and flown properly, inbound aircraft are established at 600' AGL well before crossing the downwind leg for the runway on a perpendicular course, enroute to a midfield flyover and missed. There is no conflict between the VFR traffic pattern and this approach, when everyone has eyes outside (and augmenting that with ADS-B traffic data is great). But rather than continuing to simply report position, the inbound aircraft identified us by call sign and issued us an instruction to do a 360 to accommodate them. I've had this happen a couple of times, and I'm always incredulous. But that's actually not the relevant part of the story. Doing what they asked would have both compromised my training scenario and created another conflict with an aircraft behind me, so I replied "unable", to which the response was, "we have you on ADS-B, we'll blah, blah, blah", at which point their aircraft blindly maneuvered to head directly toward the location of our downwind-to-base turn, and climb right through the pattern altitude. I presume this is based on what they saw on ADS-B based on their last announcement, but who knows? In any case, they clearly had positional awareness, but not situational awareness. Another example, less interesting but still to the point: traffic is congested enough in our metro area that a "common training frequency" has been established, which all the local flight schools would like pilots to monitor if more than a few miles from any particular airport. I'm unconvinced this is helpful, but I'm not a jerk, so I do monitor it, and do my best to be polite. I routinely get calls to my N number from aircraft that are multiple minutes away from a possible conflict, asking me to "say intentions", and wanting to negotiate separation even greater than what the professionals who staff ATC require. Again, I try not to be a jerk about it, but the minor irritation is that it unnecessarily distracts from training; and the major issue is that I've seen it create bizarre panic on a couple of occasions when more than two aircraft happened to be vaguely in the vicinity of each other, but not anywhere near a real threat. If you've never seen or heard things like this, I respect that. But don't tell me I haven't seen and heard it with my own eyes and ears. There is definitely fixation and distraction going on out there. The bottom line argument in this thread isn't whether ADS-B is "bad", no one is saying it is. The opposing positions are between "Traffic displays are easy to use effectively, never a distraction, and represent a huge safety improvement", vs. "Effectively using a traffic display is complex and can be counterintuitive, and it provides a moderate benefit against a tiny risk, at the possible expense of a small increase in larger risks". In this respect, your comments about airspeed indicators are actually a great conversation piece. We have a finite amount of time to be "heads down" in the cockpit, so what do you think the risk is of taking time away from checking the ASI while in the traffic pattern, in order to look at your traffic display? Yeah, yeah, I'm sure you can do both at the same time - you're an excellent pilot, you can pat your head and rub your tummy, etc. But the accident data is right there in black and white: huge numbers of takeoff and landing accidents, very tiny numbers of midairs, and this hasn't changed since ADS-B became commonplace. As an instructor, should I really be telling other pilots, "I see you're looking at the airspeed indicator a lot, you really should spend more time on your ADS-B display"?

This is a cool picture and certainly shows that the Stanfield stack is busy. But it also illustrates the emotional effect of zoom scale. In this clip from the image above... ... there are four airplanes inbound on the approach to KCGZ, and another two in the vicinity, in addition to the ownship. Sure looks busy. But the distance between the VOR and the airport is 8 nautical miles. The very closet airplanes in this image are about 2nm apart, and the case that is nearest to head-on probably has a closure rate of about 200 knots groundspeed, or about 30 seconds of time at 2nm. Certainly worth paying attention to, and having the depiction on ADS-B is useful. But none of the airplanes in the picture are (yet) in a "near miss" scenario by anyone's definition, even if they were all at the same altitude; and everyone is less close than the traffic pattern at many uncontrolled airports on a Saturday morning. It's also worth noting that while your average GA airplane is about the size of a house, the icons in the depiction suggest a single airplane is about as big as the entire city limits of Casa Grande. My experience working with clients is that these issues of scale/zoom are often poorly understood, leading to inappropriate fixation. For better or worse, this is what I think about every time someone posts an ADS-B depiction of all the airplanes inbound to Airventure being "crazy"; or says something like, "you just don't realize how many airplanes are out there"; or expresses certainty that a threat they saw on ADS-B would have speared them but for their superior situational awareness and valiant maneuvering. Of all the bad things that could happen to the airplanes in this picture - basic loss of control, mechanical failure, etc. - a midair is the lowest risk by far, even given what it looks like on the display. Presumably the pilots in this picture are all working on their instrument skills, and that training results in significantly more risk reduction for them than the risk of the stack itself. To hear some on this thread characterize it, operating in the Stanfield stack or something like it without devoting a significant amount of attention to traffic displays is a death trap; and yet as has been noted, the stack was around for decades before ADS-B traffic became commonplace. That doesn't mean ADS-B traffic is worthless or that pilots shouldn't use it. But the folks saying, "If it saves even one midair it's worth it" are ignoring a crucial part of the calculus, which is, what other risks increase and how much additional maiming and death is caused by fixation on a very small threat, to the exclusion of more significant ones? In a world of finite resources, fixation on shark repellent instead of swimming lessons may be a bad idea, even if it conclusively prevents a single shark attack.

The caveat is that this information can be of limited value unless you're already pretty close to the airport. Seeing the direction and density of pattern traffic at an airport a few minutes away is helpful. Less so when you're 10+ minutes out, because things can change a lot in 10 minutes, at least in some areas of the country. This is especially true in the dense metro area where I teach, hence my sensitivity to it. There are no less than 7 GA fields one could choose to practice pattern work at, within reasonable flying distance of each other, even on a short lesson. I can't tell you how many times I've had an instrument or commercial student say something like, "Hey, it looks too busy at KXXX, let's go do our pattern work at KYYY where it's easier", only to find that in the short time it takes to get to KYYY, that airport has become a beehive while KXXX has gone quiet. Because of this, I no longer base my planning on what traffic appears to be doing on ADS-B at an airport that's more than a few miles away, and that just feeds back into my arrogant opinion that traffic displays should prohibit zooming out more than a few miles. More generally, one of the common mistakes I observe training advanced students is a stubborn assumption that airport conditions won't change between the time they first think about them, and the time they actually get close to the airport. Ironically, this problem gets worse the better the student is at staying ahead of the game. It's admirable to check winds, etc. via ADS-B or ATIS/AWOS when you're 20+ minutes away from an airport, but if you do that, don't forget to spot check them again when you get closer.

Absolutely. The graph you posted shows the accident rate per 100K hours has declined from 5.82 to 4.84 in the last 10 years, a roughly 20% improvement. The fatal rate is also about a 20% improvement. No argument there. But you seem to conclude this is mostly or at least largely a result of improved technology. The accident data doesn't support that at all. The graph that's relevant to this thread is this one: That's for 2022. You can take my word for it (or not) that the graphs for the past 20 years are essentially identical. Midair collisions are in the "other" category. It's difficult to tell how many of these were actually midairs vs. other "others". But it's probably reasonable to estimate that midairs are roughly represented by the fatals. That's 3 in this case, which is about 0.3% of all accidents, and about 2% of fatals, both of which are statistically insignificant. So that 20% reduction in overall accident rate isn't coming from a reduction in midairs due to ADS-B improving threat awareness. I'm sure some of it comes from other technology - onboard weather, better maps and backup instruments in IMC, etc. But it's worth noting that the majority of accidents occur during takeoff and landing (557 out of 965) , and by definition fall into the "loss of control" category (again, collisions are in the "other" category). Technology essentially never helps you avoid a takeoff crash. It can help avoid a landing crash to the extent you might divert somewhere else with better conditions to land in, but I'd wager only a very small number of those landing accidents fall into the, "I wouldn't have tried to land here if only I'd known earlier in the flight that the conditions were challenging" category. I think you're misrepresenting my position . I never said that no one should be allowed to spend money on technology, or that it should be banned/shunned. I said that when you have X amount of dollars and or time to spend it on "safety", upgrading the avionics in your airplane - and demanding others do the same - is almost never the best use of that money and time. Now, there are lots of good reasons for avionics upgrades other than safety. They're fun (I'm an electronics nerd, I really enjoy playing with them); they can make a flight more comfortable; and to the extent they make you feel safer, there is some indirect safety value in that even if you haven't actually changed your direct risk profile. But what gets my hackles up is stuff like this: To the extent this is true of ADS-B, it's at least as true of training with an instructor, and per the data above, arguably much more so. So... the bare minimum training requirement for Part 91 ops is a 1-hour flight review every two years. Is everyone who pursues no additional training beyond that also doing a disservice to themselves and everyone around them? To be clear, my point is no so much to shill that everyone should get more flight instruction, but just to point out the curiosity of the passion around technology - particularly that designed to reduce midair collision risk - vs. the "meh" attitude toward everything else that is much more likely to hurt you, and somewhat more likely to hurt others. Bottom line, if you're going to start throwing around opinions about what you think every other pilot should be investing their time and money on, it better be backed up with data; and mid-air collisions ain't. The fact that one can post an occasional story about a midair where ADS-B might have made a difference doesn't change the risk profile. As tragic as those accidents are, they're still a drop in the bucket vs. the things that actually put us most at risk. Focusing on them is akin to investing a bunch of money in shark repellent for a beach vacation, instead of on swimming lessons.

This is a great question, and I think it's really the crux of the matter. It's certainly the reason why I tend to get my hackles up about this stuff. An aircraft that is "near" certainly presents more risk than one that is "far". But my experience flying with a lot of other pilots is that many of them are unable to appropriately internalize what actually constitutes "near". A comical example of this - but an honest-to-god-true-life-story - is a pilot I flew with whose display showed no threats at a particular time during the flight... so he kept zooming out until he found one (any aircraft looks "near" if you zoom out far enough). Then he got fixated on maneuvering to avoid this wildly distant threat, instead of paying attention to something more important. This resulted in him basically losing control of the aircraft (uncommanded attitude change under the hood) while trying to figure out what direction to turn to avoid an aircraft that was multiple minutes away (and likely was going to turn away from us anyway). That's an extreme example, but I see variants of this quite a lot - pilots who are very concerned about midair collisions, who spend an inordinate amount of time looking at traffic displays trying to analyze threats that are objectively tiny, to the detriment of managing other risks. It turns out there's a pretty narrow range in which traffic displays actually help. If you're 5 seconds from a conflict, they're unhelpful because they tend to indicate bad position data, just like @MikeOH noted above. The delays in the ADS-B network are indeed very small, as the reference posted by @Aaviationist indicates. But that study ignores delays in the display device (maybe a couple of seconds for a Bluetooth-connected iPad which is the most common setup), and delays in the pilot's scanning of said device (maybe very large). At the other end, if you're more than 60 seconds from a conflict, the information just isn't relevant - you don't actually know if there's a threat, and the best course of action is to just keep doing whatever you're already doing. If I had my way, no traffic display would be capable of zooming out beyond a couple of miles. In the narrow range between "too far to care" and "too close to help", ADS-B traffic displays are really helpful when used properly, and all the arguments extolling their virtues make sense. Airplanes in the same traffic pattern or holding stack with you fall into that range, and people pointing that out have a strong argument. But it's worth noting that's also when all your other risks unrelated to traffic go way up. I flew with a pilot earlier this year for an aircraft checkout who was incapable of maintaining pattern altitude and appropriate heading and airspeed on the downwind leg of the traffic pattern, because he was completely heads-down on his iPad, and terrified of a couple of other airplanes trying to work into the pattern. That's just a single anecdote, but again, I fly with a lot of pilots as an instructor, and I see a lot of variations on this basic mismanagement of airmanship. One thing that would help is if the industry could put together an AC on how to effectively use traffic systems. Today, there's just an assumption that the way to use them is "obvious", and that's a lot of what all the arguing is about. One of my fellow flight instructors wrote a great article in the NAFI Mentor magazine about this. I don't have a link, because it requires a subscription. But the gist of it was to define certain ranges to the threat, and appropriate actions to take in response to threats in those ranges. Outside a certain range, no action is warranted and there is no value in paying attention to the threat. Once a threat moves inside a "noticeable" range (30 seconds or so of closure), a single, predictable, course or altitude change may be warranted before visually acquiring the target - especially if that change actually helps you and the threat acquire each other visually. But if the target moves inside a "critical" range, blind maneuvering is as likely to cause a collision as it is to prevent one, and the best action is to maintain course and altitude while working especially hard to visually acquire the threat, perhaps to the extent of paying less attention to your usual fly-the-airplane duties. Only after visual acquisition is maneuvering appropriate inside this range. Everyone is free to disagree with this strategy, and would be even if there was an AC. But @EricJ is asking a really good question: not if ADS-B traffic data is helpful, but when. Especially relative to other risks.

The problem with this kind of thinking is that all of us have finite limitations on the amount of time and money we spend on "safety", and those resources often wind up being misdirected. Consequently, the most important risks aren't mitigated, and the rate of maiming and death stays about the same. Scientifically, this is borne out in accident data. CFIs who enroll in a FIRC every two years get data jammed in our face about it. The rest of the pilot community could benefit from at least a casual look at it. AOPA does a good job with their annual McSpadden report. Here's a link to the data from 2022, which is the most recently compiled year: https://www.aopa.org/training-and-safety/air-safety-institute/accident-analysis/richard-g-mcspadden-report/mcspadden-report-figure-view/?category=all&year=2022&condition=all&report=true Anecdotally, in my neck of the woods, some of the biggest risks to the aviation community are the aircraft with $100K of avionics, flown by pilots with so little recent experience that they can barely keep the greasy side down, and have almost no ability to really grok where other traffic is in the pattern (or the stack, or the airway, or whatever). Imagine how much better things would be if those pilots kept their airplane on the flight line instead of in the avionics shop, flew more often, and spent a few bucks on training with the local CFIs, or even just flying with more experienced friends. But this seems to be a loser of an argument, primarily because humans are terrible at data analysis and risk management. For better or worse, those folks really believe their big avionics upgrade was a great investment in safety, and scoff at others who are less-well-equipped as being "too cheap for aviation". Those pilots aren't going to read the McSpadden report; or even if they do, they think it doesn't apply to them. So yes, I do think "all of the other safety precautions we routinely take to mitigate small risks" are often inappropriate. Not because it's wrong to mitigate all risks, but because we live in the real world where no one has infinite time/money/energy to actually mitigate all risks. If you can't mitigate all the risks, you should focus your time and attention on the biggest ones. That requires the curiosity to actually look at the data, and the humility to accept that you are not somehow a special pilot, whose risks are different than average.

Most of the value in getting an instrument rating comes from developing the ability to easily and precisely control the airplane with less and less cognitive power, such that you have more and more cognitive power to tend to other things: reading charts, programming navigation, obtaining weather, interacting with ATC, scanning for traffic, etc. Pilots who obtain an instrument rating get fairly good at this, at least at the time of their check ride (the skills lapse if you don't use them). These skills certainly make one a better and safer pilot, but there are other ways to develop them without getting an instrument rating of course - mostly just a lot of cross-country flying with flight following. The rating itself may not have much value to any particular individual - especially if they do most of their flying in a part of the country where the weather is not conducive to flying a piston single regardless of having the rating.

Shaw (now Parker Aerospace) made a lot of different fuel caps with a 3" diameter and 120-degree, three-tab locking mechanism. They're not interchangeable from a paperwork standpoint, but one flavor often fits in the hole designed for another... except for maybe the orientation of the tabs/holes. Between honest confusion over part numbers, willful ignorance, and the reality of maintaining aging aircraft, I think there are a lot of airplanes flying around with Shaw caps that are functional, but technically incorrect for the airframe. Incorrect "clocking" of the cap is possibly a clue you've got the wrong flavor of cap on the airplane, but not necessarily. I used to be all pompous about this, and say any airplane with Shaw caps that could not be installed with the tab pointing directly aft, clearly had the wrong caps. But I've never been able to find actual guidance from Shaw/Parker or any airframe manufacturer about the orientation of the tab, just a bunch of people on the internet who say, "I was taught ..." The closest thing I've found to official guidance is that the illustrated parts catalog for some airplanes depict Shaw caps in a particular orientation. But I'm not sure that's really intended to convey there is only one correct way to install the cap. Even if it did, IPCs for different airframes show different orientations.

I used to ask the A&P/IA who did our annual throughout our first 10 years of ownership, about our smoking wing rivets, basically every year. We have a lot of them, but he was completely unconcerned about it, particularly given the ugly cosmetic condition of our airplane. Finally, when I asked one too many times, he said something to the effect of, "Alright Junior, the straight dope is that if it's really bugging you that much, you can re-buck a rivet exactly one time, and it *might* stop the smoking, but no promises". So I gave it a try, and it didn't seem to help. For better or worse, I quit asking about it after that, which was over a decade ago. The fact the wing hasn't come off since then doesn't necessarily mean it's OK to ignore the smoking rivets, but it's a data point. The concept of re-bucking relies on the idea that the original installation didn't work harden the material enough to prevent it from further deformation, i.e. that there is still some "give" left in the rivet. It also assumes the problem is a rivet that was never set correctly in the first place, rather than something else causing the rivet to move. All of that is a crapshoot. If you really want show-plane quality with absolutely no smoking rivets, I think you're in for a long haul of drilling and replacing, and you have to carefully examine each drilled out site to ensure the hole in excellent shape to receive a new rivet (which may not have been true even when it was built at the factory). I've never been able to find official guidance that matches my mechanic's "you can only rebuck once" advice, maybe that was just lore. But if true, it's irritating, since even the best maintenance logs are not going to identify re-bucking of individual rivets, especially given that the factory might have done some re-bucking themselves. So in practice, I think it's essentially impossible to tell in advance if re-bucking is going to correct a smoking rivet or not - you're not necessarily going to have the same good/bad luck someone else had, there is too much variation in circumstance. All you can do is try it.

Just for reference, here's what mine looks like at present (no LASAR cowl closure).

@Ragsf15e's picture exactly matches what our 1976 M20F looks like today. I think the flat head screw with the countersunk washer at the rear of the cowl is OEM; and that the one at the front of the cowl is actually not. I think the one at the front originally matched the other two truss-head screws up there, closer to the prop; but that IAs get nervous about the underlying hole getting ovaled out much larger than the diameter of a #10 screw as the years go by, and go to the flathead/washer arrangement as a "fix". That's just a guess, though. The IPC for this vintage of Mooney doesn't actually detail any of these screws, see attached. Whatever the case, the important thing to understand is that the underlying nut plates are 10-32 threads. Or, at least they are supposed to be. In the early days of our ownership, I was chagrined to find a small number of coarse thread sheet metal screws jammed into places that were clearly supposed to receive machine screws. In most cases, I was able to use a tap to un-bugger the threads of the nut plate to receive the correct screw. This is yet another reason to bring a bin of correct screws to your annual or other maintenance. It's not uncommon for a less experienced or just less patient shop assistant, who gets the lowly work of re-installing cowls and inspection panels, to lose a screw, and in their embarrassment/frustration just jam in whatever random fastener in the shop doesn't immediately fall out and hope you don't notice.

By the way, these are the screws used for the vast majority of the belly and wing inspection panels, avionics access covers, etc. They're 12 cents a piece at Spruce, you can order a lifetime supply for practically nuthin'. Keep a bin of 'em in the hangar for use as needed. Bring that bin to your annual, so you can throw away any screw that's even the slightest bit buggered and replace with new. It goes a long way toward reducing the frustration of the annual R&R'ing-the-screws ritual.