Dangerous Flying: Abrupt Control Inputs

How sloppy flying could get you killed.

Recently, while flying with a 200-hour helicopter pilot, I was startled when he rather abruptly shifted the cyclic to make a turn. I didn’t say anything then because it wasn’t too abrupt (whatever that means). But when he did it again later in the flight with an even more abrupt movement, I spoke up and told him not to do it again.

Understand that we were flying a Robinson R44 Raven II, which has a rather unforgiving semi-rigid rotor system and very long rotor blades. We’re taught — or should be taught — during primary training to use smooth control inputs, especially when working with the cyclic.

I’m not a CFI and I don’t feel that I have the right to tell someone how to fly, but when a pilot does something I believe is dangerous, it’s my duty to speak up. So I did.

The trouble is, I’m not sure if he believes what I told him — that abrupt inputs are dangerous — or if he thinks I was just nitpicking his technique. (I let it go the first time partially because I didn’t want to be seen as a nitpicker.) Since so many pilots seem to read this blog to learn — or at least to get my opinions on things — I thought I’d discuss it here.

What Robinson Says

Section 10 of the R44 II Pilot’s Operating Handbook includes safety tips. Here’s the one that applies:

Avoid abrupt control inputs or accelerated maneuvers, particularly at high speed. These produce high fatigue loads in the dynamic components and could cause a premature and catastrophic failure of a critical component.

What Robinson is saying is that when you make abrupt control inputs you put stress on various aircraft components. They’re likely concerned about the rotor blades, mast, transmission, and control linkages most. This makes perfect sense.

Robinson Safety Notice SN-20, titled “Beware of Demonstration or Initial Training Flights,” includes these statements:

If a student begins to lose control of the aircraft, an experienced fight instructor can easily regain control provided the student does not make any large or abrupt control movements. If, however, the student becomes momentarily confused and makes a sudden large control input in the wrong direction, even the most experienced instructor may not be able to recover control.

And:

Before allowing someone to touch the controls of the aircraft, they must be thoroughly indoctrinated concerning the extreme sensitivity of the controls in a light helicopter. They must be firmly instructed to never make a large or sudden movement with the controls.

Of course, what worries Robinson here is that student pilots may make erroneous control inputs beyond what an instructor can fix to regain control of the aircraft.

What Worries Me More

January 31, 2012 note: Since writing this, a friend on the Rotorspace site has brought the topic of Mast Rocking to my attention. Apparently, some folks think that this accident may have been caused by Mast Rocking rather than an abrupt cyclic control input. I’m not convinced. Mast rocking supposedly does not cause the main rotor blades to diverge from their normal plane of rotation. How else could the tail be cut off in flight?

But what worries me more than putting stress on components is an accident report from 2006. I read this report on the NTSB Web site not long after the accident occurred. Back then, there was no known reason why an R44 helicopter with just two people on board for a long cross-country flight should fall out of the sky with its tail chopped off, but I had my suspicions. After my recent flight with the new pilot, I looked it up again. Here’s the probable cause (emphasis added):

The Canadian certificated commercial helicopter pilot was conducting a cross-country delivery flight with a non-rated passenger occupying the copilot seat. The passenger and pilot together had previously made delivery flights from the Robinson factory to Canada. Two witnesses saw the helicopter just before it impacted the ground and reported that the tail boom had separated from the fuselage. No witnesses were identified who saw the initial breakup sequence. Both main rotor blades were bent downward at significant angles, with one blade having penetrated the cabin on the right side with a downward slicing front to rear arc. The primary wreckage debris field was approximately 500 feet long on an easterly heading. The helicopter sustained damage consistent with a high-energy, fuselage level, vertical ground impact. Detailed post accident investigation of the engine, the airframe, and the control systems disclosed no evidence of any preimpact anomalies. The removable cyclic was installed on the left side copilot’s position, contrary to manufacturer’s recommendations when a non-rated passenger is seated in the left seat. The removable pedals and collective for the left side were not installed. The cyclic controls for both the pilot’s and copilot’s positions were broken from their respective mounting points. The copilot’s cyclic grip exhibited inward crushing. The Safety Board adopted a Special Investigation Report on April 2, 1996, following the investigation into R22 and R44 accidents involving loss of main rotor control and divergence of the main rotor disk, which included a finding that the cause of the loss of main rotor control in many of the accidents “most likely stems from a large, abrupt pilot control input to a helicopter that is highly responsive to cyclic control inputs.”

The National Transportation Safety Board determines the probable cause(s) of this accident as follows:
a loss of control and the divergence of the main rotor blade system from its normal rotational path for undetermined reasons.

(A full narrative is also available.)

This is pretty much what I’d imagined. The helicopter is cruising along at 110 knots in a very boring part of the California desert. For some reason, the pilot (or his passenger, who has access to a cyclic control), jerks the cyclic one way or the other. Maybe he was trying to dodge a bird. Maybe he was goofing off or pretending to be Airwolf. Who knows? The sudden input is enough to cause the blades to diverge from their normal path. One (or both) of them dip down and chop off the tail boom. The result: two dead bodies in a 500-foot long debris field.

And this is what was going on in the back of my mind when the pilot beside me made those sudden inputs.

Anyone who has flown a Robinson helicopter can tell you how responsive the cyclic control is. It wouldn’t take much effort to knock the blades out of their path. That’s why we’re taught — or should be taught — to use smooth control inputs.

Other accident reports like this one include: CHI05CA267 and MIA00FA102 (which is a “watch this” moment).

Other Concerns

Rotorcraft Flying HandbookThere are at least two other reasons to avoid abrupt cyclic movements. You can find all these in the Rotorcraft Flying Handbook, an FAA publication that’s a must-have in any helicopter pilot’s library.

Under the “Retreating Blade Stall” heading (page 11-6):

High weight, low rotor r.p.m., high density altitude, turbulence and/or steep, abrupt turns are all conducive to retreating blade stall at high forward airspeeds.

Personally, I don’t think retreating blade stall is an issue in Robinson helicopters, except, perhaps, at high density altitudes and high speeds. But in that case, you’d be exceeding Vne.

Under the “Low G Conditions and Mast Bumping” heading (page 11-10):

For cyclic control, small helicopters depend primarily on tilting the main rotor thrust vector to produce control moments about the aircraft center of gravity (CG), causing the helicopter to roll or pitch in thedesired direction. Pushing the cyclic control forward abruptly from either straight-and-level flight or after a climb can put the helicopter into a low G (weightless) flight condition. In forward flight, when a push-over is performed, the angle of attack and thrust of the rotor is reduced, causing a low G or weightless flight condition.

You can find an account of this (with a lucky pilot and passenger) in this accident report from July 22, 2010. Indeed, the problem may have occurred during the right turn the pilot initiated — did he jerk the cyclic over as my companion had done?

Another accident report that suggests mast bumping is SEA03FA148 (which took the life of a pilot I knew).

I’m Not Just Nitpicking

The point of all this is that I’m really not just nitpicking a fellow pilot with limited flight time. He performed a maneuver which I consider dangerous and I have all this information to back me up. It’s important for him — and for others who might not know any better — to avoid abrupt control inputs.

Robinson helicopters aren’t capable of safely performing aerobatic maneuvers. Don’t fly them as if they are.

Update, March 17, 2012: Here’s another example of an accident likely caused by an abrupt control input. This one resulted in mast bumping.


Discover more from An Eclectic Mind

Subscribe to get the latest posts sent to your email.

12 thoughts on “Dangerous Flying: Abrupt Control Inputs

  1. This note does not directly pertain to this post, but I just had to tell you – I just talked with my daughter and she told me that she will start helicopter lessons. She made her single-engine pilot license last year and now also adds IFR and qualify for both. She lives in the UK, at times also in France and also the US, so IFR allows her to fly in these places. Naturally I had to let her know about your site and your flying. Not to worry – she will not ask you for advice or otherwise lean on your shoulders. So now I have two ladies in my life that break the mold. Til another time – Eberhard

    • Charlie: Thanks. I felt it was pretty important to make my case. I seldom comment on other pilots’ flying techniques because they often think I’m just trying to find fault with them. I’m not. But I might read too many accident reports.

  2. Really helpful Maria. For those of us who are low-time ppl’s every bit of sound advice like this is helpful. To be constantly reminded in your blogs of how to fly the 44 safely is a great help. I’m 100 hours but only 50 or so p.i.c, so every time I fly I try and think safety first. I enjoy all your posts, thanks from London, England.

  3. You’re right, of course, and that comes from someone who IS a CFI and has been, for 42 years.
    You and I know you can fly most any helicopter from L.A. to New York and not move the cyclic control more than an inch from the neutral position. For this reason, some say “You don’t really need to move the controls, just THINK about moving them”. Pressures. Finesse. That’s all it takes, and more than that is not better.
    I hope he heard what you said and took it to heart, Maria.
    You may have saved his life.

    • He did. I hope others did, too.

      Sometimes on flights with curious passengers, I demonstrate how I can make a turn without apparently moving any of the controls. Of course, I’m putting just the tiniest bit of pressure on the cyclic when I demonstrate. Passengers are always amazed.

      To me, that’s the hardest part of flying a helicopter: using a light touch. Once you get a feel for it, it’s easy. But as you know, the first 10 hours or so of flight training can be brutal.

  4. As a crew member on a Bell 412. I can tell you that you are not nitpicking. It is in your best interest to speak up if you feel uncomfortable with what anyone else is doing. It doesn’t matter if they are new or extremely experienced. If your butt cheeks are pulling the fabric off the seats from clenching, it may be time to reassess the situation. When I fly I follow a simple rule – I don’t want to die today so Im not going to let you kill me! I know it sounds silly but it reminds me to step in before things get out of hand.

    • I hate telling other pilots how to fly, but when they do something dangerous I really need to get over that and do what needs to be done. Thanks for taking the time to comment.

  5. Hi Maria –

    I’m an old dog learning new tricks; high time guy (>60),
    1200 hrs fixed wing, low time helo student.
    Love your blog, found it when searching for info on R44
    and really appreciate your comments
    and POV on flying and training.

    I tried to go to the reports listed here:
    “Other accident reports like this one include: CHI05CA267 and MIA00FA102 (which is a “watch this” moment).”
    but got an error from the NTSB server.
    Might be my browser but thought you should know.

    Rick

  6. Smooth control inputs are not just a good idea from an airframe perspective, they also help keep your passengers from getting sick. Always a good idea to keep their tummies happy, unless for some reason you like cleaning up barf from your helicopters interior.

    I would however disagree with the proposition that you can fly any helicopter coast to cost with one inch of cyclic travel. While it’s probably true for most helicopters with 2-bladed semi-rigid rotor systems (especially those with hydraulic boost), it’s definitely not the case with fully articulated or rigid rotors. The Huges 269 / Schweizer 300 series that I used to instruct in is a case in point, They require substantial amounts of cyclic movement (especially fore / aft) during several maneuvers, noticeably more than the R-22, which I have also instructed in.

    If you switch back and forth between different types on a regular basis the difference in required cyclic inputs can be quite noticeable, especially when one is something like a Jetranger/ Longranger and the other is an AStar or a Hughes 500. At one point I was checked out (Part 135) simultaneously on a Longranger, an AS350, a BK117, and a Bell 212 (too many types, really) and it was occasionally a bit sporting (especially the initial lift-off) when I hadn’t flown one type for a while.

    Oddly enough, the “backwards” pedal inputs required when switching back and forth from French to U.S. types wasn’t an issue at the hover, but instead manifested itself typically when making big power changes on decent. In a hover, your feet know what to do to make the nose point where it needs to go, even if it does seem “wrong”.

What do you think?

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