EFIS Safety Study: Hardly a Surprise

  • E-Mail this Article
  • View Printable Article
  • Text size:

    • A
    • A
    • A

After I read the NTSB's findings on glass cockpits and safety, I felt the urge to stifle a yawn. My initial reaction was, is this all there is? My next reaction was to try to recall if glass displays were ever pitched as significant safety enhancers. Maybe I missed school that day or perhaps the notion that EFIS displays are just naturally safer has become so intuitively ingrained that we don't even discuss it anymore. I wouldn't have been a bit surprised if the study—which involved very few accidents as these things go—revealed a negative safety benefit for EFIS.

Either way, the NTSB's findings confirm my own research into the subject. I'm currently poring over Cirrus accident data to gain some sense of how this airplane—which really pioneered EFIS for everyman—fits into the overall accident picture. My initial findings are in tune with the NTSB's work: The Cirrus accident rate is about in the middle of GA in general, but the percentage of accidents that are fatal is higher than it is for other models. The boilerplate reason for this is that the airplane is used for different types of missions that involve transportation flights in challenging weather. That may be true, but a number of Cirrus accidents occurred on flights that don't meet this profile.

Another way of looking at the NTSB's report is to turn it on its head. Accepting that the Cirrus is often used essentially like a business jet is, it has significantly more exposure to high-hazard weather than other models. Yet in that high-hazard environmental, the Cirrus (globally) has turned in an accident rate similar to the fleet in general. Following that logic, the airplane as a system may in fact be statistically safer. "System" includes the airframe, the pilot and the EFIS. I'd add another element to that: the Cirrus Owners and Pilots Association, whose members have a dramatically lower accident incidence than non-COPA members.

I've always felt that EFIS for EFIS's sake is overblown. My view is that its real benefit is in training pilots to fly on instruments, not necessarily keeping them current. As an instrument instructor, an abnormally high percentage of my total time is instrument time in actual IMC because I did that instruction in the northeast and mid-Atlantic states where the weather abides the serious IFR student. I've never trained a student using an EFIS panel and I'd like to hear from instructors who have.

Given the centered, logical scan opportunity inherent in a PFD, I've also always assumed that it's just easier—and perhaps safer—to fly behind a glass panel, operability and complexity issues notwithstanding. I'm not so sure that's true, however. I was recently offered the left seat of an Entegra-equipped Cirrus and we launched into really low IMC. At the time, I was way out of currency, but with a qualified CFII in the right seat, I figured this would be the perfect laboratory to test my EFIS-easy-scan theory. I figured that a couple of thousand hours of steam-gauge IMC time was bulletproof insurance.

We disappeared into the overcast at 200 feet and, just as I predicted, I had the headings and airspeeds nailed. It was an illusion, however. When I transitioned into the more dynamic maneuvering of the approach, I made a complete hash of it—headings all over the map, yo-yo descent rates—all the stuff of a rusty, undisciplined scan. So much for bulletproof. I doubt I would have done any better on steam gauges, but at least for me, EFIS didn't offer sufficient magic to overcome lack of recent training, despite my expectation that it might.

I continue to believe that glass panels represent real progress and offer the potential for lowering accident rates. But like every other new technology, it takes time to integrate the advantages into daily operations—maybe a lot more time than anyone thought.

The all-purpose bromide that we need better training is just part of it. Our publications have written widely on the various EFIS systems and, frankly, I have found many of these articles to be quite irritating. They tend to deal with flyspeck minutiae at the expense of the big picture and often tell the reader how to adapt to the machine's software foibles while losing sight of the basic idea of flying the airplane in clouds. It's as though the purpose of flying is to twirl knobs and try to digest a lot of cool data that's not even relevant to getting there.

So I don't expect much out of "improved training." Frankly, I am more intrigued with the basic attitude that causes a pilot to join an organization like COPA and remain engaged with it as a means of sharpening proficiency. The give-and-take and informational exchange with like-minded pilots clearly has more efficacy than another dreary acronym like FITs or TAA or whatever the hell the industry comes up with next.

When you think about it, things like this have to do with personal responsibility. With the right attitude, even high-risk flights can be conducted safely and most of us can figure this out without someone else dreaming up special training.

Comments (62)

I was with you until the statement, "With the right attitude, even high-risk flights can be conducted safely..."

A high-risk flight, by definition, cannot be conducted safely. Maybe successfully (with a little luck). But the "right attitude" that causes you to depart on a high-risk flight is the same attitude that will eventually get you killed.

Posted by: Brad Koehn | March 11, 2010 11:39 AM    Report this comment

Maybe I should clarify the term. I view high-risk as night, in icing in a single-engine airplane like the Cirrus. I believe it can be conducted relatively safely and luck has less to do with it than skill, proficiency and judgment. It is, on its face, higher risk than a day VFR flight in good weather.

That doesn't mean it cannot or should not be done. It depends on one's personal tolerance for risk. "Safely" doesn't mean no risk, it means tolerable risk.

Posted by: Paul Bertorelli | March 11, 2010 12:12 PM    Report this comment

My analysis of accident data has repeatedly shown that "maintenance" plays an extremely high part in accidents. Granted I look at Beech mostly, but this week is very typical... 2/3 were engine failure & gear collapse. However the "Lesson Emphasis" was 1) Crosswind takeoff, 2) Are we safer? Parts I/II, 3) Too tired to fly?.

While I appreciate all the study material concerning safety, constantly read it, take the test(s) available... none of it has ever helped in near accident conditions. Flat spin due to W&B problem (maintenance), 1 mag failure on take off over a lake at 200 feet with wife (low time) flying (3 hrs. on rebuilt mag), stuck main gear & loss of directional control ( 1st landing) following an annual on test flight (failure to grease nose gear) etc.

Why would I ever think that another "Star Wheel" like EFIS would save my bacon? My bacon is my responsibility and that is why I fly with a constant healthy appreciation (fear) as to how my little bucket is trying to kill me... Being rusty, as Paul points out, is my biggest safety problem. My last aerobatic foray was a good example... wow, what a ride. I did double snap roll, never saw the first one, but caught the 2nd perfectly... claimed I planned it. Not funny. I am my worse enemy...

Posted by: David Spencer | March 11, 2010 1:56 PM    Report this comment

Yeah, I hate those invisible snap rolls, too.

Posted by: Paul Bertorelli | March 11, 2010 2:14 PM    Report this comment

The element of "risk" and the perception of risk seems to be a moving target depending on circumstances. A friend of mine who now flies a Cirrus has made a number of flights he would not have made in his old Bonanza. He has even commented on how much more "capable" the Cirrus is. Okay, the Cirrus has a couple of TV sets in the panel and a parachute. How does this change the pereption of risk from high risk to reasonable or acceptable risk?

Posted by: Richard Montague | March 11, 2010 3:03 PM    Report this comment

We might see improvement out of "improved training" if it's the right kind of training - and this should include teaching skills for adapting to software foibles and overall button-pushing. From what I see in the avionics retrofit world is a huge learning curve for otherwise experienced pilots switching from familiar steam gauge to glass. A curve so high that basic ILS and GPS approaches are completely blown because the data is either not properly set up on the PFD, the wrong data is loaded, deer in the headlights reaction when it comes time to load this data and all of the above. There just isn't enough quality instruction out there to make a difference in the safety numbers and few pilots can safely learn it on their own.

Posted by: LARRY ANGLISANO | March 11, 2010 3:31 PM    Report this comment

I think training is everything in this case. Can advanced equipment help-sure, but it is no replacement for good instrument currency and understanding your equipment. When learning to fly a high performance TAA (I've never flown glass) it was amazing to me how much there was to learn (how to use the rate function on the autopilot, how to couple approaches, limitations, how to program and use all functions of gps, etc.) I've noticed pilots with limited ability (some even VFR only) who either buy the fanciest glass-equipped airplane money can buy or drop mega bucks at the avionics shop, then blast off in marginal conditions because of the extra "safety" margin of their new avionics. I think these are the guys clouding the statistics. I'm willing to bet a proficient pilot in a 172 with two nav-coms and an adf is significantly safer than a marginally proficient pilot in his TAA. That being said, I love the new technology, and not going to give up my 430.

Posted by: Josh Johnson | March 12, 2010 8:08 AM    Report this comment

Several years ago I made the decision to buy a Columbia 300 rather than the all glass 350. Why? As an "old" pilot, I determined that a steam gage panel with HSI, a Garmin 480 and Apollo 200 provided situational awareness upgrades and the older panel was much easier to maintain instrument proficiency (and cheaper to maintain). I had flown the Cirrus 22 and formed the opinion that the human being would have no choice but to watch the "TV" screens rather than look out. Why the Columbia? The aircraft would pass certification without a mandated ballistic chute. However, the Columbia/Cessna isn't satisfactory as a primary trainer because of the lack of stall performance. This leads to my opinion that we aren't training competent and safe pilots with the latest generation of fancy paneled aircraft. We need to train pilots to fly, to fly the basic stick and rudder and to understand flying the wing. How to do that? Easy. All candidates for powered flight would be required to first solo and accumulate 15 hours in sailplanes, then move up to a minimum of 15 hours and a solo in tailwheel aircraft. Then, the candidate could move to the "panel" of choice to finish the rating. In conclusion, until we return to the basics our infatuation with the best panel will continue to provide the best unsafe accident statistics.

Posted by: Barton Tate | March 12, 2010 9:03 PM    Report this comment

"I figured that a couple of thousand hours of steam-gauge IMC time was bulletproof insurance." Honestly, the very limited experience I have with simulators makes me think that steam-gauge experience counts for something, but not likely to save the day if one had to fly by the glass panel. I wish the avionics manufacturers would make a "Southwest Switch" - Southwest Airlines for a time had presentations of the typical 6-pack of instruments on their glass cockpits. At least this way, when we purchase a new aircraft, we have a choice. This wouldn't fix all the transition issues with glass panels, but might help a little bit.

Posted by: Josh Johnson | March 13, 2010 7:52 AM    Report this comment

As a flight instructor at a school with 60+ glass cockpit aircraft I have many opinions about this topic. I do believe that EFIS GREATLY increases situational awareness, especially in IMC conditions.
However, I do agree that students tend to look inside not out. Because of this I will usually turn off the monitors(when practicing PPL maneuvers) and force the student to look outside.
I think the biggest thing for improving safety with EFIS panels is improving training. I can imagine instructors who don't routinely fly these aircraft know as much as they should about the advantages of "glass cockpits"

Posted by: Ryan Konrath | March 13, 2010 5:05 PM    Report this comment

I'm somewhat astonished at the number of instrument pilots (and instructors) that can't use anything other than Direct-to on a 430 - I can only imagine the situation with a glass cockpit. In fact, there are entire online courses on the 430/530 series as well as the G1000. If we want the airline's safety record, we're going to have to start agressively training on the new avionics.

Posted by: Josh Johnson | March 14, 2010 6:44 PM    Report this comment

First of all, I am irritated by the use of the term "glass cockpits." The multi function displays are constructed of and best described as plastic cockpits. No glass in the Garmin displays at all. If you are going to communicate with words, why not make them accurate. Kind of like Peter Jennings saying "black box" when we all know they are orange.
I feel a better use of assets where flight safety is concerned would be to install a minimum of 2 fully capable digital autopilots with auto throttle. The autopilot should have the ability to fly a flight from takeoff to landing.
The use of the term "tail dragger" is also used incorrectly today. Tail dragger aircraft have tail skids. The correct term is tail wheel aircraft. Almost every aviation writer is guilty of misusing words like the above.
I want each of you who reads this post to look at his wrist watch right now. Is it a digital format or the analog format. Most of my friends who fly have recognized the analog version is superior to the digital presentation.
Attempting to stay current for flight in instrument meteorological conditions using the computer in the cockpit is impossible unless you fly a minimum of 200 hours per year.
Just my opinions...What do you think?
Dick Siano

Posted by: Richard Siano | March 15, 2010 8:32 AM    Report this comment

>>Almost every aviation writer is guilty of misusing words like the above.

Posted by: Paul Bertorelli | March 15, 2010 9:03 AM    Report this comment

I have no problem with the recommendation for more training, even if it is the answer to all questions.

The statistical analysis from which the NTSB headline derived is flawed. A well-designed cohort study has both groups equivalent, except the variable under study. In this study, mission, weather, and crew composition were all different. The only valid conclusion is that a glass cockpit was installed in more aircraft that suffered fatal accidents. It doesn't tell you whether the instrumentation made the difference.

The slide that got all the attention was one that described the fatal accident rate in glass-cockpit equipped aircraft from 2006-2007 as 1.03 per 100,000 flight hours. The Nall report describes the fatal accident rate for the entire GA fleet for the previous ten years as 1.11 to 1.38. New aircraft, equipped with glass cockpits ARE safer.

Posted by: Robert Hadow | March 15, 2010 9:21 AM    Report this comment

Just about the time the US Army Air Corps became the US Air Force, a study of fatal accidents in IMC among high-time pilots while transitioning to new aircraft concluded that the difference in instrument scan patterns required for instrument flight from one cockpit to the next was a major factor. That observation led to the adoption of the "standard tee" instrument layout we steam gauge pilots have grown to know and love. That Air Force standard instrument layout was eventually adopted by civilian manufacturers and almost every instrument panel designed after the early sixties incorporated the tee layout creating a safer environment for instrument rated pilots encountering instrument flight conditions while transitioning from one aircraft to another.

Transitioning into an EFIS equipped cockpit requires an adjustment in scan pattern that the Air Force concluded is dangerously difficult even to highly trained, skilled pilots. It makes me wonder whether the departure from standardization in the EFIS world is a factor somewhere near the core of the accident rate issue.

Posted by: William Harper | March 15, 2010 10:04 AM    Report this comment

I think that the comments about the aircraft mission and environment get to the real issue. People who want to take a quick trip around the patch on a nice day mostly aren't the ones dropping lots of money on fancy glass panels. The ones who do want to use the airplane for transportation, and with that comes additional weather exposure, night and other risk factors.

Not sure if the data exists to do it, but it would be interesting to see of the accidents analyzed for the conclusions about glass panel airplanes were also analyzed for how many of those accidents occurred in which types of operation (IMC, night, icing, versus VMC, day, etc), other trends might emerge that were at least as enlightening.

Posted by: JON CARLSON | March 15, 2010 10:21 AM    Report this comment

R. Siano: "Most of my friends who fly have recognized the analog version is superior to the digital presentation."

Especially when you factor in a little far-sightedness.

Posted by: Mike Holshouser | March 15, 2010 10:27 AM    Report this comment

Duh, The NTSB's recommendations were spot on. I believe there should be a logbook endorsement requirement for glass cockpits AS INSTALLED on a given brand aircraft. For instance, a Garmin 1000 in a Cirrus has a different emergency electrical backup system than a Diamond. I also believe recurrency on desktop simulators would greatly improve pilot knowledge as of airborne software version upgrades change the "switchology: and alert thresholds from time to time. There is no need for any new FAA "ratings", but the same credo holds true as it did for the Wright Brothers, the person operating the airplane, NEEDS to know the aircraft systems intimately to be considered a safe pilot.

Posted by: Brian Baldridge | March 15, 2010 10:35 AM    Report this comment

The first concern I have about sophisticated glass panels in the cockpit of General Aviation aircraft are as follows:
1. Training in glass aircraft tends to use up an inordinate amount of cockpit time trying to learn the right buttons to push and when. Two problems with this:
a. This tends to keep the eyes inside the cockpit.
b. It reduces the time for what should be more important concerns to the student, especially a primary or commercial student.

2. Is the situational awareness advantage of the glass justify the time spent learning how to use the glass?

3. Then, what do you do with a pilot who flies twice a month or less often? How does that pilot keep current in how to make the glass do what he wants it to do? Would we be better off training that pilot with steam gages or with glass?

Button pushing training would be better suited to some kind of a desk top or dedicated simulator or computer driven trainer that allows the student to learn the button pushing without wasting aircraft time that would be better spent teaching the pilot how to fly. While these simulators are “around”, I have yet to see one, unless attached to a FTD with the actual head and knobs that really did the job properly.

Posted by: THOMAS OLSEN | March 15, 2010 11:56 AM    Report this comment

Thomas, that's a good point. When my CAP unit got assigned a C-182T G1000, the first thing we did was arrange with a local flight school to use their excellent G1000 FTD. I spent three hours with one of their CFIs (who was also a squadron member) in the FTD before I ever touched the real airplane. It made huge difference when I did my first flights. Ground training is essential for someone transitioning from steam gauge. Otherwise you do spend way too much time heads-down. If your local flight school doesn't have an FTD, the next best thing is to hook the aircraft up to a battery cart and sit in the airplane practicing the basic functions. The new technology has the potential to be safer than steam gauge...the increase in SA is amazing...but not if we try to use 1950s training methods. The way we train pilots is going to have to catch up to technology if we're going to realize any safety gains.

Posted by: Chris McLellan | March 15, 2010 1:04 PM    Report this comment

I am a CAP member and a CAP Check pilot and agree with Chris's comments. Good ground training and flying the computer is essential. Part of this is also recurrent retraining on the computer. Garmin provides with most of their systems a computer program that allows learning and relearning the G1000 operation. While NOT a simulator, it goes a long way in teaching and learning the various buttons and knobs before getting into the cockpit. If an owner or renter of such systems are not using this valuable tool then accidents/problems will occur while the heads are down looking for the right knob or button.

I do not know if Avidyne or other manufacturers have the training aid that Garmin has, but I hope they do.

Posted by: Bobby Picker | March 15, 2010 2:54 PM    Report this comment

Learning to fly glass is just that: Learning to Fly Glass. Problem is there are pilots who do not take the time to learn, thinking "this shiny piece of glass will keep me safe". There are also pilots who take the time to learn where to park eyes for information.
Pilots with knowledge and experience will have a better chance of taxying after returning to the earth's surface.
There is a lot of training material to help a round dial pilot transition to flying glass. What is missing is the knowledge in how to learn. I know pilots who will buy the book or "King" course only after flying 20 to 30 hours. Too late.
After market glass accidents is an area that was not addressed and will probable not be able just for the very reason that there is no way to identify an aging TAA.
Insurance companies will get on the ball requiring training and set amount of hours before you can act as PIC. And then will be born the phrase Learnt to Fly Glass

Posted by: Neil Keller | March 15, 2010 4:00 PM    Report this comment

Well said Neil. Glass or steam, it still comes down to basic airmanship, and that includes learning your aircraft. I've only got about 25 hours in glass and I still consider myself a newbie. What I hope to see in the future is some attention paid to the system interface. The G1000 (I don't have any Avidyne experience) has a lot of functions that use less-than intuitive controls. I'd like to see future generations have a simpler interface...A touch-screen MFD would be a good start.

Posted by: Chris McLellan | March 15, 2010 4:15 PM    Report this comment

Perhaps not only a simpler, but a standardized interface would be nice. In the bad old days, just about anyone could figure out how to work two nav-com's, but today specific training is really necessary to make our new avionics work.

Posted by: Josh Johnson | March 15, 2010 5:16 PM    Report this comment

Years ago I had the privilege of regularly flying a Mooney 231 so over-equipped with King stuff that it was reduced to a 2 person airplane with full fuel: HSI, flight director, 3-axis autopilot, RMI, radar altimeter, radar, even a "flite fone"--better than many of the airliners of the day. It was the equivalent of the over-equipped Cirrus/Bonanza/Cessna/Piper/G1000/Avidyne/whatever of today, just round gauges instead of square ones. It took me only a couple hours to get comfortable VFR--but it took a good 10-12 hours and a lot of regular practice, not to mention pouring over the manuals for all of the avionics, to get comfortable IFR, not because of the displays but because of having to learn all the nuances, i.e., what would the switches actually do? That's the problem with today's "glass" cockpits--they are so capable that it's hard to transition to them--too many switches and knobs which can provide too much information, unless the pilot has taken the time to learn. And there does appear to be some standardization issues as well, but that was/is true of "steam" gauges too.

Training and currency are the keys to safety (minimization of risk, I suppose), not what the instruments are made of or how they are shaped, or how much information they can display. So the NTSB's findings aren't surprising.


Posted by: Cary Alburn | March 15, 2010 7:56 PM    Report this comment

A problem I've seen with flying glass panels is that we get so accustomed to the great situational awareness that we get mentally lazy about keeping good SA in our heads. So then when we go back to flying a steam gauge panel in IMC we lose the bubble. Had this happen to an instrument pilot with whom I was doing an IFR currency/proficiency flight. He had lots of G1000 time but we were flying with steam gauges on an ILS in IMC...he got so totally disoriented that I had to take over the airplane. So unless a pilot is sure he/she will never again have to fly in IMC with steam gauges, he/she should maintain steam gauge proficiency as well.

Posted by: Paul Gardella | March 15, 2010 8:24 PM    Report this comment

Hello Paul Gardella --
Good point. You said, "So unless a pilot is sure he/she will never again have to fly in IMC with steam gauges, he/she should maintain steam gauge proficiency as well."
Even if the G1000 never fails, the pilot will for sure have to fly on the steam gauges on his or her next IPC.

Posted by: Robert Hadow | March 16, 2010 8:43 AM    Report this comment

My two cents;
I watched some of the NTSB's broadcast and it centered on the fact that systems and displays have no "standards". And as almost all responders have noted, training must be required.

I think what NOT said was too loud to ignore. The thing that stands out for me is the deaths that resulted from VFR conditions. AND even with balistic 'chutes onboard!

I think what is missing here is simply PRACTICE. I mean too learn how the aircraft "feels" and reacts to the pilot's inputs and THEN how the panel reacts to the aircraft. We cannot remove our part of the flying system. It must also be noted that our part of the "system" requires PRACTICE and the more the better. Read the comments of Mr. Barton Tate (above) and relate that with Mr. Bertorelli's piece on his re-aquaintence with a Cub and THATS what I'm talkin' 'bout!

Posted by: Larry Fries | March 16, 2010 12:07 PM    Report this comment

I can't help but wonder if the accident rate for glass cockpits isn't just a spin off of the #3 big killer, failure to fly the aircraft in an emergency. Yes, glass is distracting, but are we looking at over confidence in the equipment, the need to "get there", and the pilot just flying at the outside edge of his/her capability when an emergency occurs?
In the end glass versus steam may be just a case trying to separate the fly poop from the pepper.

Posted by: RICHARD GIRARD | March 17, 2010 5:21 AM    Report this comment

The "V-tailed Doctor Killer" Bonanza of long ago was largely a result of very intelligent, highly paid, busy people getting in over their head in aviation. Does the glass cockpit present a similar situation? The new generation airplanes, and glass retrofits, are very expensive. Highly successful people that can afford these things may not have the time to become, or maintain, proficiency. Perhaps their hard-driving, self confident, nature is influencing their flying.

Posted by: Dan MacDonald | March 17, 2010 11:49 AM    Report this comment

I have little EFIS time, but from that, I found the moving bands in particular for airspeed, but also for altitude, much harder to read by merely glancing at them because they essentially looks alike all the time, only the numbers change. Old EFIS hands tell me that in principle that's true, but that speed/alt bugs and trend information help over that. I know that I never had this trouble with the old round gauges though, not even on my first flight lesson.

I have seen altitude bands now that have notches every 500 ft, which seems to be an attempt to address this problem by giving the eye something to hold on to.

Posted by: Arno Schoedl | March 18, 2010 3:25 AM    Report this comment

If one is truly trained in the likes of the G1000, I do think that TAA are a safer aircraft. I have over 200hrs in a G1000 equipped aircraft (that I miss since moving). Having gone through the Cessna G1000 course back in 2004 in Independence, KS and then flying back our C182, we were able to shoot a GPS approach to minimums (at night), go missed, and then head to our alternate with little issue. In fact when we landed on the ground we were not nearly as spent as you'd be flying a traditional approach to minimums. And all of this after flying from Kansas to Central Texas. The key is to be fully aware of the systems and how to manipuate them. As both myself and 'co-pilot' are in the IT industry, we were able to grasp both the 'switch-ology' of the G1000 and the computers that made up the system that much easier to comprehend. It made the trip much more enjoyable and we were able to maximize the use of the aircraft.

Posted by: Bryan Bogle | March 18, 2010 6:40 AM    Report this comment

Ditto Dan MacDonald for one possible aspect of this. Business guy in a Cirrus instead of Doctor in a Bonanza.

Posted by: Mike Holshouser | March 18, 2010 8:02 AM    Report this comment

Moving maps certainly improve situational awareness. But is all this new digital technology really targeting the causes of most accidents? This emphasis on glass is like the rush to electronic voting machines: low/no correlation between problem and solution.

As this technology (e.g. datalink, autopilots) expands "capability," you have to control for this new operating envelope when you compare accident/safety data (for improvements, etc.) or you are comparing apples to oranges.

I used to skydive a lot. The new technology was great: light & durable materials, elliptical planforms with incredible performance, turbine jump planes, etc. All of that led to more jumping, higher winds, and doing more with the new "capability." And this brought even more people into the sport. And yet there were still a lot of accidents -- some for different reasons -- but overall things really hadn't improved. So I too am not surprised by these findings.

BTW, my watch is analog. :-)

Posted by: BRADLEY SPATZ | March 18, 2010 10:03 AM    Report this comment

Actually, in skydiving, the total number of fatalities has remainded relatively constant against increasing numbers of jumps. The actual accident rate per 1000 jumps has declined.

What the technology has done--automatic deployment devices, reserve static lines, highly loaded canopies and fast-climbing turbines--has changed the nature of the accidents. (Far more open-canopy fatalities now than in 1974 when we jumped rounds that tended to...not open.)

I think the Cirrus experience and glass in general is a virtual parallel.

Coincidentally, I have a analog watch, too.

Posted by: Paul Bertorelli | March 18, 2010 10:40 AM    Report this comment

Eeow! I'll stay inside start to stop. I broke my watch.

Posted by: Larry Fries | March 18, 2010 1:51 PM    Report this comment

The key slide in the NTSB hearing said the fatal rate in glass cockpit airplanes was 1.03 per 100,000 hours. The AOPA ASF Nall report describes ten-year results between 1.11 and 1.38. So, like skydiving, the accident rate is going down.
It's perfectly worthwhile to argue glass v round, but do so aware of the safer flying results. To imply that glass is making us less safe is disingenuous.
As we argue, let's also make sure we are using the right terms. TAA refers to a moving map. Glass Cockpit in this report, refers to a PFD.

Posted by: Robert Hadow | March 18, 2010 2:05 PM    Report this comment

I thought of the wristwatch analogy myself awhile back and had considered basing a magazine article on the concept. ...Look at any of the flagship watches from Omega, Rolex, Breitling, Tag Heuer, and Tissot etc. …They are all analogue and they don’t have a digital watch insight! …“Seems to me like those companies know something that aviation has lost sight of ”

…Analogue watches are better than digital watches quite simply because you can visualise time ahead and behind on them, but they are also nicer to look at on account of a 3dimensional quality that the digital equivalent can only replicate in 2d.

For similar reasons I believe airspeed and altitude indications/trends can be recognised more easily on the old style round ASI and altimeter gauges. As well the last of the mechanical gyro attitude indicator/artificial horizons had very much evolved to reflect a 3 dimensional quality that for me was much nicer to look at than the modern digital equivalent, which is certainly only a 2D reference.

A popular word often used these days to describe EFIS is that it is more intuitive. … “Whereas in fact I believe the truth is exactly the opposite”

Posted by: peter hirst | March 18, 2010 5:46 PM    Report this comment

Although I agree with you on analog presentations, the general view is that pilots prefer digital tapes. We did a survey of bout 700 users last year and the majority said they would stick with digital rather than analog if given the choice. (57 percent said digital, only 15 percent preferred analog.)

I'm in the 15 percent myself.

Posted by: Paul Bertorelli | March 19, 2010 5:12 AM    Report this comment

Arno - This is a problem that I see quite a bit. The Bugs are the best way to stay on headings/altitudes. Students, especially who have never used a flight director or autopilot, have trouble using the Heading/altitude bugs. The first thing you learn when using a flight director is set heading/altitude bugs before making any changes to flight attitude. For example before turning from a heading of 090 to 180 you twist in 180. This will have the flight director "show" you the turn. If you fail to do it in this order as you start to make your turn the flight director will try to have you turn back to your original heading, causing problems. Instructors NEED to stress the use of these "bugs"

Posted by: Ryan Konrath | March 19, 2010 6:55 PM    Report this comment

I am wondering if the industry has ever conducted a test of pilot performance with the analogue ASI versus the digital tape ASI in difficult or limiting weather conditions.

i.e. ...Trialled the two indicating systems with numerous experienced pilots being measured for their degree of control with similar jet aircraft and especially during a series of hand flown approaches with substantial wind shear and/or crosswinds. ...Somewhere like Gibraltar or Vagar in the Faeroes - both of which are notorious for fearsome wind shear.

Posted by: peter hirst | March 20, 2010 5:07 PM    Report this comment

>>I am wondering if the industry has ever conducted a test of pilot performance with the analogue ASI versus the digital tape ASI in difficult or limiting weather conditions.

Posted by: Paul Bertorelli | March 20, 2010 7:26 PM    Report this comment

>>Yes, some. Avidyne did some research in this area but it was not indepth enough to be conclusive.

Posted by: peter hirst | March 21, 2010 2:53 AM    Report this comment


Posted by: Larry Fries | March 21, 2010 11:38 AM    Report this comment

I've taught students in the G1000 C-182T with user hostile KAP 140 autopilot as well as various G-400/500 navigators with user friendly S-tec or Century rate-type autopilots (vacuum driven sensors are suspect until proven honest). IMHO knowing now to quickly enable the wing leveler function is as good as a copilot: Those with an 'Oh Shit' button on the yoke will at least hold the greasy side down so you can sort things out. A favorite unusual attitude recovery training technique is to get the airplane all upset and ugly, turn on the wing leveler and watch what happens. I have yet to be disappointed. Add a nav tracker so it can take you to safety. Sadly, some pilots will not turn on an autopilot to save their lives, literally, especially if it has previously scared hell out of them. But those scares seem to be related to altitude hold or rate functions. So don't enable them if it adds to the workload.

I said all that to say this: It shouldn't matter if the plane has glass, steam, spinning iron or laser nav systems when it comes to physically flying the plane. if there is a wing leveler and the pilot knew how to and was willing to quickly enable it, wouldn't that have a better outcome than hand-flying when the workload is high?

Posted by: Thomas Connor | March 21, 2010 9:15 PM    Report this comment

As an aside I wonder at the number of accidents per 100,000 hours claim. In over 10,000 hours of flying in various private and personal aircraft burning various fuels nobody has ever asked me how much I flew, so IMHO the basic accident data is suspect. But it appears few question its validity.

Posted by: Thomas Connor | March 21, 2010 9:15 PM    Report this comment

Paul - a study of the literature on human factors and instrumentation would lead one to discover that mission critical instrumentation is always analog. Take one step and realize that human can't really multi-task. What we lay persons think of as multi-tasking is really high speed serial tasking. The more data you throw at the brain more time it takes to sort out what it means - so those old steam guages with the VSI headed south and the Glide Slope needle headed north are actually easier to interpret than that nice clean glass panel.

At the end of the day - you still need to fly the airplane - and the only thing that seems to keep me sharp is doing it regularly and regular recurrent training.

Posted by: George Mattingly | March 22, 2010 2:06 PM    Report this comment

One of my favourite cockpits for single crew IFR was to be found in the Cessna 421 Golden Eagle (last produced by Cessna in the mid 1980’s) and for me it remains perhaps the most ergonomic and intuitive cockpit arrangement that I have found.

…All flight, engine and navigation data was represented beautifully on analogue gauges. In particular as each gauge reflected data sourced directly from its own separate element, a gauge and its attendant box could most easily be regarded according to priority. “Moreover I maintain that as those separate values of flight data could be instantaneously referenced/altered/manipulated as well as disregarded for electrical/mechanical fault” …a Cessna 421 pilot really could operate his machine intuitively.

Of course aviation hadn’t latched on to the fashionable hype word “intuitive” in the 1980’s. My recollection is that word was first foisted on the industry as a marketing device to vaunt the joint Garmin430/Avidyne product in the Cirrus. In fact when I fly a Cirrus SR20 or 22 I feel very much trapped outside the Garmin430/Avidyne interface. In particular that interface feels like a bottleneck to me and through which I am obliged to input an unacceptably large proportion of the data required for safe operation of the aircraft. So it is that personally I think that Garmin/Avidyne combination is most certainly a lot less intuitive to work with than 1980’s generation cockpits.

…Continued below :-

Posted by: peter hirst | March 22, 2010 6:52 PM    Report this comment

...continued from the box above.

That said I wish to make entirely clear that my argument has nothing to do with GPS or moving maps, which viewed as a separate entity have made a magnificent contribution to aviation. …Only a fool would criticise GPS and/or a moving map, but still I would have preferred such displays to be kept separate from other cockpit functions including engine and fuel data.

Specifically I am unhappy with the computer programmers interfering with my basic six flight instruments and turning them into an inferior flat 2 dimensional PC style image, as well as obliging me to work through their avionic data bottlenecks.

Moreover I take the view that the EFIS cockpit whilst admittedly capable of representing large amounts of flight data also has the effect of blending that data into a smooth digital cocktail that has much potential to disadvantage the pilot in his ability to prioritise data. …I am sure that must be the case for trainee pilots. Many years ago I flew about 1600hours as a flying instructor teaching PPL’s and much enjoyed the experience, but I really think it must be a wearisome task these days trying to teach an ab initio pilot in a machine with two glass screens.

Posted by: peter hirst | March 22, 2010 6:55 PM    Report this comment

I have a suspicion that a lack of standardization in the instrument scan pattern between the glass equipped cockpit and the standard tee incorporated in most steam gauge panels may be close to the core of this safety issue, but there may also be other contributing factors.

I've flown with flight instructors who insist that the best way to approach the biannual flight review or an instrument competency review is to fly the simplest of the aircraft in which one is qualified to conduct the review.

Could it be that the opportunity for life saving feedback on the potential competency issues associated with the more complex aircraft is lost in this manner? After all, if I'm going to fly a G-1000 equipped SR-22 or a Chelton equipped Twin Comanche into actual instrument meteorological conditions, does it make sense for me to fly a steam gauge equipped C-172 when taking an instrument competency checkride or biannual flight review?

Somehow I don't think so and my question is, did the pilots involved in the reported incidents/accidents do their biannual and/or their last instrument competency check in the aircraft involved in the incident/accident or in a much simpler, steam gauge equipped aircraft? I doubt that the data is readily available, but it surely would be interesting to know.

Posted by: William Harper | March 22, 2010 11:07 PM    Report this comment

If the choice is yours then choose the aircraft in which you feel the most competent to complete check rides. And always specifically avoid flying any other machines/equipment that you are unsure of into difficult or testing situations/weather.

Posted by: peter hirst | March 23, 2010 6:16 AM    Report this comment

I'd hazard a guess that the CFI recommending a flight check in the simplest aircraft is so the CFI doesn't have to get smart in whatever higher grade of aircraft you fly. Both for hubris and liability reasons.

Posted by: Thomas Connor | March 23, 2010 10:41 AM    Report this comment

Our failure to teach proper risk management is likely the root cause of most TAA fatal accidents. The ability to identify, assess, and mitigate hazards is fundamental to reducing these accidents. As FAA's lead executive for general aviation from 2001-2005, I started the FAA Industry Training Standards (FITS) program to promote the training of higher order pilot skills, including risk management. After I left FAA in 2005, some additional progress occurred but, as an industry, we still cling to outmoded training doctrine.

Posted by: Robert Wright | March 23, 2010 11:08 AM    Report this comment


I agree that the choice of checkride aircraft is up to the pilot being checked and I also agree one should avoid flying aircraft that you are unsure of into difficult weather.

I'm suggesting that a simple aircraft isn't the best choice for the pilot who normally flies a more complex aircraft because he owns, normally rents or is paid to fly a more complex aircraft. The opportunity to get feedback is missed on issues the pilot may not be aware of when flying the more complex aircraft and that may not be sufficiently obvious to the instructor or checkride pilot to draw comment when the person being checked is flying the simpler aircraft.

I, for one, always want to be tested in the aircraft that I own and won't opt to rent an airplane that is easier to stay ahead of. If my skills need to be honed I want the feedback that will permit me to focus on the areas I'm demostrably weak in when flying the most complex airplane I will probably be flying, my own.

Reviews and checkrides should be opportunities for learning and improving safety, not check the box activities. A checkride you ace because you're so far ahead of the slower, more simply equipped C172 builds false confidence that the same good performance will apply in the faster, feature rich environment of the glass paneled SR22 you're about to fly from one coast to the next.

Posted by: William Harper | March 23, 2010 12:39 PM    Report this comment

A third category is the endless list of acronyms and processes to analyze the pilot, plane, environment, bank account and state of the universe. 'Safety' is a growth industry but for a primary flight student and instructor there is a point where it is overkill and both just want to grow some balls and go flying. As one crash inspector told me: Safety is a way of reducing the cost of doing business. Either train better or carry lots of insurance. And don't be surprised if the insurer insists on more training."

Posted by: Thomas Connor | March 23, 2010 12:47 PM    Report this comment

Just out of curiosity I searched on the Internet for cockpit photos of the Diamond DA20 aircraft acquired in the last few years by the USAF Academy. …“Well it’s interesting to note that the USAF Academy aircraft are equipped with round gauges”

…Surely implying the USAF prefers that its potential aviators be introduced to aviation with conventional non-EFIS equipment.

Posted by: peter hirst | March 23, 2010 5:05 PM    Report this comment

The USAF Diamond DA20s are equipped with standard six-pack instrument clusters ON THE RIGHT SIDE, engine instruments and circuit breakers on the left with G430 and GTX330 (I think) in the center.
They made their buy decision prior to the availability of the LCD panels.

Posted by: Robert Hadow | March 23, 2010 5:43 PM    Report this comment

Nevertheless if my memory serves me correctly the USAF ignored or turned down the Cirrus SR20 and that did come with a Garmin/Avidyne glass cockpit at the time.

Posted by: peter hirst | March 23, 2010 7:05 PM    Report this comment

Hello Peter --
For all of its great points, the SR20 is unsuitable as a military trainer. It is not type certified for spinning. You can be sure that the Cessna T-41 spun a million turns between 1964 and 1993. The Diamond replaced another aerobatic aircraft, which the SR20 is not. Whether the USAFA chooses to use a round-dial or PFD equipped aircraft for flight screening is probabably not the indicator for most of GA. *** We are complaining mightily about the G1000 presentation, but no one is focusing on the amount of data we are presenting. Here is some of the stuff I never saw while I learned. Each one, arguably, is valuable, but in aggregate, the list is quite long: (1) peak cylinder EGTs, (2) TFRs, (3) Traffic Alerts, (4) NexRad WX, (5) calculated wind display, (6) terrain depiction and alerts, (7) graphical fuel exhaustion ring ... need I go on? It's no wonder a new G1000 user's head spins.

Posted by: Robert Hadow | March 23, 2010 8:12 PM    Report this comment

Hi Robert, actually I believe the SR20 is unsuitable in any capacity as a training aircraft. Nor am I convinced that it has many great points. Beyond the fact that it looks quite pretty and has a good range (I have often flown SR20’s on 675 nautical mile legs and landed with 60minutes fuel) I can think of nothing else to recommend it.

For me the SR20 would be a much nicer machine if it came with conventional spectacle controls, manual aerodynamic trimmers (as opposed to electric/spring biased) and analogue flight and engine instruments. …Of course I would be happy to keep the moving map on the right hand screen.

Posted by: peter hirst | March 24, 2010 1:40 AM    Report this comment

I think that good instrument flying is more about zen then electronics. I remember one rather nasty evening training with Paul. I was flying the Mooney on a rather challenging approach to our home airport (you crossed three ATC sectors if I remember correctly) when I transitioned into the "zone". My subconscious mind was flying the airplane and I entered real situational awareness. I was free to think, plan and stay ahead of the airplane. It almost felt like an autopilot was flying. It's a known fact that when professional tennis players start thinking about each shot they loose. It's almost like they revert to being a beginner. After all, a beginner does think about each stroke. I think this is true in instrument flying as well. When an experienced instrument pilot has to engage the conscience mind to fly the airplane, performance declines. Glass or no glass. It takes many hours of instrument flying and constant currency to reach the point of when the subconscious mind can fly the airplane. It could be that people haven't been able to reach zen on glass yet.

Posted by: DANA NICKERSON | April 1, 2010 8:54 PM    Report this comment

Every site have their own goal in the making of it. This site is make fro the people that want to have a good time with all their citizen by telling them what they are feel bout the political view in their country. interior design services, architectural design services,

Posted by: monita smith | April 6, 2011 11:58 AM    Report this comment

Add your comments

Log In

You must be logged in to comment

Forgot password?


Enter your information below to begin your FREE registration