Better (Safer) Instrument Flying!

Last week’s “VFR into IMC” blog generated a lot of comments and discussions online. Many readers were surprised at how poorly instrument-rated pilots do statistically with a simple 180-degree turn in IMC (when surprised) – usually not much better than VFR pilots. This problem illustrates some interesting – and scary – facts about most instrument pilots.  Many IFR-trained pilots are pretty bad with manual control; either never trained well or rusty. Many IFR pilots have also never actually been in a cloud (not FAA-required). The trend away from hand flying is also accelerating as avionics become more capable. The modern pilot in most planes is increasingly becoming a “programmer.” There are endless YouTubes online covering “buttonology,” and all hell breaks loose when “George” goes away requiring hand flying to survive. These pilots might as well be remotely operating their (MQ-9?) “aerial vehicles.” Automation dependency is not just a problem with GA, but started with automated airliners and is now a problem in aviation at all levels and starts with basic flight training.

In nearly 100 million flights by United States passenger airlines over the past decade, there has been a single fatality. Other than most landings and takeoffs, the planes have largely been flying themselves.

The origin of this problem is with the basic training and testing of pilots. One of the weakest parts of most IFR checkrides is the applicant’s demonstration of basic instrument flying skills (without the automation). This is an increasing trend despite the FAA’s urgent plea to develop and maintain manual piloting skills (and not just IFR). If you want to be safer as an instrument pilot – and have more fun – please get some actual IMC flying and work on your hand-flying skills (I am guilty too). We already know that the autopilot can fly in IMC just fine, we need to focus on keeping the hand flying skills sharp too.

The pilot’s role has moved from “physically manipulating flight controls and interpreting cues into a role where they ‘interact and control complex systems and play a central role in system safety. ERAU white paper

In a good instrument training course, the first 1/3 of the training should be learning to control the aircraft entirely by instrument reference and without automation. This involves discovering and utilizing standard power and attitude references for performance targets. This is naturally disappointing for many pilots because there is an urge to play with all the fun computers in the panel they paid so much for. But only when an IFR trainee can hand fly as confidently by instrument reference as by visual outside cues are they ready to move on to tracking, holding and approaches. (BTW, there is a completely analogous situation in VFR training where every new learner wants to start off with landings, before the basic skills are mastered).

So if your CFII starts your IFR training off with approaches (as is the case in most quickie “crash courses”) have a discussion, or just fire them. And saving money by hiring the cheapest CFII you can find is also a dumb idea; it takes years to learn to teach instruments well with a perspective of actual experience; hire a pro. You are developing skills that will save your life in an emergency (or not). Fly safely out there (and often)!

Apologies to my MQ-9 buddies (instructors and pilots) nearby! Apparently the MQ-9 is entirely "hand-flown?" (Bad assumption on my part). And I have never been invited to try one out...for some reason?!

Join SAFE and get great benefits (like 1/3 off ForeFlight!) Your membership supports our mission of increasing aviation safety by promoting excellence in education.  Our FREE SAFE Toolkit App puts required pilot endorsements and experience requirements right on your smartphone and facilitates CFI+DPE teamwork. Our SAFE CFI insurance was developed by SAFE specifically for CFIs (and is the best value in the business).

Author: David St. George

SAFE Director, Master CFI (12X), FAA DPE, ATP (ME/SE) Currently jet charter captain.

4 thoughts on “Better (Safer) Instrument Flying!”

  1. I have gone inadvertent IMC and I had little choice.
    I was arriving at an airport just outside a class C airspace. It was faster to cut through the class C, so I contacted the controller with my destination, and was given a squawk code. There were scattered clouds around, but the bases were high enough that I intended to descend between and then below them before arriving at my destination airport. This was south Florida, so scattered afternoon showers were a normal occurrence I had seen for years.
    I was now VFR inside a departure corridor in class C airspace for a busy commercial airport talking to ATC, on a squawk code, when the departure controller (I knew her) apparently cleared the tower for an IFR departure of a Boeing sized aircraft to take off toward me.
    Between my aircraft and the Boeing was a good bit of IMC, clouds that obscured my little C182RG from the Boeing now climbing out towards me. I wasn’t on the tower frequency, but heard the Boeing contact departure. ATC directed me to turn in a direction that would put me in IMC. I announced ‘unable’. ATC again, told me to turn to the heading. I again said unable and explained I’m VFR the turn would put me IMC. She said it again with a bit of panic in her voice and I realized a Boeing was going to exit out of the cloud near me and I did want to be there when he popped out. So, I announced my compliance with the directions, and turned into the cloud thinking it was much softer than a large Boeing aircraft.
    I knew I could fly on Instruments, but really didn’t like flying IMC without preparation. Now I was IMC, unprepared, in controlled airspace, I don’t even think I had instrument charts in the aircraft. My only saving grace was I maintained at one time or another all the equipment used for the approaches at the airports around me. Once in the clouds I couldn’t really descend out of them (I was well aware of the surrounding towers) and I couldn’t do a 180, not that I would have tried one.
    Just like the 180 deg. turn back if your engine quits on take off, the 180 doesn’t get you pointed back to the runway. It doesn’t matter if you are instrument rated or VFR pilot. Depending on speed even if you could fly the perfect 180 level turn… you could be several miles away from where you entered the IMC and really don’t know what the conditions are you are going. Popping out of IMC after a perfect 180, is just as much ‘luck’ as continuing straight. I’ve always thought it was better to know what is in front of you, than hoping nothing is in front of you. I now had to leave the class C in IMC, fly out into the Everglades then descend out of the clouds on an ILS I didn’t have the plates to tell me the decision hight, I had to ask. I made it back down to VFR well before decision hight, but is was eye opening experience. VFR into IMC might not be avoidable. The plane I was flying didn’t have a working autopilot or anything the fancy new glass of today. I had nothing to ‘program’. Everything was already programmed in my head.
    I had to give the incident a lot of thought. Almost becoming a bug on the windshield of a Boeing is an unnerving event and my only choice was to go inadvertent IMC. A 180 was what I remembered being told to do. What if that isn’t even a choice?
    This happened to me around the time Kennedy crashed in inadvertent IMC. I was flying in from the Bahamas, so I was out over the water in hazy conditions quite a lot. I was often flying the same Saratoga type over the Bahamas that he crashed. I went out to the FAA training facility in Oklahoma frequently for training. I would go over to aero medical to experience hypoxia and vertigo. I wanted to know what it felt like to get vertigo, and could I still fly with forced vertigo, I found it hard to even keep the wings level, much less make a coordinated turn.
    This is why I have always questioned the idea of doing a 180. It makes little sense. Yes, do a 180 before you go into a cloud. But what if you have little choice? What then? Like stalls, we don’t want to stall the plane. But, we train what to do if a stall does occurs so you will survive. As with a stall, sometimes it is straight, sometimes it is in a turn, sometimes it is accelerated, sometimes it is slowing down. Each requires a solution so you don’t fall out of the sky. The same various solutions are needed for IIMC. The pilot needs to understand the situation and taught react appropriately. I don’t think a 180 is a good idea in almost every situation of IIMC.

    On your current article, Many years later, (decades) my instructor and DEP on my last instrument check ride, both said, they know Cessna can fly on instruments, let’s see if you can. Flying by hand in actual IMC for a couple of hours is very hard mental work. I’m usually very tired after making a few approaches very close together, flying by hand, sometimes bouncing around, trying to prepare in flight for the next approach, both looking for the approach plate and programming, (Yes, my instructor and DPE changed airports and approaches mid flight) while I was hand flying most of the time in actual IMC with my instructor.
    Many times it seemed the controller was in on the testing with a few changes to my arrival to the selected approach, while flying my arrival. Once I remember the controller changing my arrival three times within a few miles of starting the arrival. I found this tough after flying for decades on six packs and using paper plates. Smooth air is hard. I’m not sure if I could do it in turbulent air, without being on autopilot at least while ‘programming’ the approach.

  2. I completely agree with the main point of your article. Where you lost me was the line about MQ-9 operators standing on the ramp flying their planes. The caption under the picture “totally autonomous” was equally false.

    I can assure you that MQ-9 pilots fly and land their planes from inside a control station with a live video of a forward looking camera i.e. a first person view. The crosscheck of flight instruments and visual references is similar to flying a manned plane. Additionally, the stick and rudder skills required are like a manned airplane right down to the rudder and aileron inputs to land on the upwind wheel on a crosswind landing. It could even be argued that an MQ-9 pilots task is more demanding since he lands this 8000+ pound, 900 HP plane with the aspect ratio of a glider without the benefit of peripheral vision, seat of the pants feel, or sound cues.

    CFII,MEII, MQ-9 Pilot, Retired USAF Manned and remotely piloted pilot

    1. I agree, I’ve been flying drones and RC for about thirty years. It is harder to fly them than a manned plane… but you don’t have the same risk of killing yourself if you screw up.

  3. I trained in the 1970’s with 2 instructors: one was a former Royal Canadian AF Captain on C-130’s; the other was a former Project Engineer for Douglas. Between them, I was trained to ROUTINELY do under-the-hood takeoffs AND under-the-hood landings, from either left or right seat. My friend and I could talk one another from either seat, through under-the-hood takeoffs and landings. Today I hear that most airline pilots hand-fly their aircraft less than several minutes per flight. I was on a major airline, and an FAA check-pilot was riding jumpseat, evidently assessing one of the pilots. As I sat in the cabin, I noticed many uneven power applications, sloppy airmanship etc. I mentioned all this to the FAA check-pilot after the flight. He was well aware. It appears that most pilots today, are far behind the curve in IMC skills. Unimaginable deterioration of basic airmanship skills.

Tell us what *you* think!

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: