The Cirrus Airplane Has Serious Problems
Born of the 1994 NASA Advanced General Aviation Transport Experiments (AGATE) program, an attempt to re-energize the rapidly shrinking general aviation market, Cirrus became at once the darling child of aviation. Acclaimed by Popular Science magazine and every aviation journal of the day; it was the newcomer to aircraft manufacturing with aspirations of mass production. Cirrus was credited with bringing forth breakthrough technologies; and unlike other AGATE aircraft, it did not evolve from a lineage of experimental kit planes, from lessons learned over decades. It was a clean sheet design. Aviation writer, Greg Laslo, quoted Cirrus executive Dean Vogel, on April 26, 1998, as saying, "In concept, the goals we strive for with the SR20 is what AGATE is striving for. It will be easier to fly and to maintain currency in nearly all weather conditions." Although there were skeptics of this new company, it had no critics. The SR20 first flew on March 31, 1995, received certification October 23, 1998. Production was underway.
Over a decade later, the plane touted on mountaintop as delivering unparalleled levels of safety has not, in fact, performed as advertised. Many Cirrus pilots have now encountered their worst fear: loss of control. Even COPA the Cirrus Owners and Pilots Association a group characterized as being fanatic about the safety of their planes issue an unprecedented alarm.
Loss of control accidents are complex and most difficult to comprehend – especially when flown by factory-trained pilots and instructors, well-qualified in hours flown, and having lots of recent experience. Standard assumption has been that the Cirrus plane is faultless.
Delve into NTSB accident reports, however, now that there are ten-years of them; you’ll discover more-than-properly qualified pilots are being killed, not the fearless and inexperienced ones as you might expect.
Here's an example:
the flight instructor's failure to maintain control of the airplane, which resulted in an inadvertent stall while maneuvering.Yet the accident occurred in excellent weather, practically no wind. Both pilots were by any measure experienced and qualified enough to fly the plane safely. The owner was instrument rated with 1,327 hours total time of which 230 hours were in a Cirrus SR22 airplane. The flight instructor was no spring chicken either. According to the accident report:
The flight instructor held a commercial certificate for airplane single-engine land, and instrument airplane. He also held a certified flight instructor rating for airplane single-engine land, and instrument airplane. In addition, he was a certified airframe and power plant mechanic, with inspector authorization (IA). His last FAA second-class medical was issued on January 30, 2007. A review of his logbook's revealed that his last entry was made on October 30, 2007. At that time, he had accrued a total of 1,438.8 hours.Did the flight instructor "fail" to maintain control, or could it be possible the airplane was uncontrollable?
In defense of the dead, Jim S, Oklahoma City, argues against an assumption that his friend was at fault:
The pilot was a personal friend of mine. I have over 6,000 hrs of time flying time as a commercial pilot. I make my living flying close to the ground, and have for several years, starting out as an ag pilot, and now seving as chief pilot for one of the largest aerial patrol operations in the US. I am VERY selective about who I will fly with. Bronc Jackson was one of the best pilots I have had the privelege of knowing, and I can guarantee you he would NOT have done anything in that aircaft that he knew of felt it was incapable of, REGARDLESS of who was giving him instruction. Some of you Monday morning quarterbacks need to ease up and wait on the NTSB reports before you go casting judgement. Anyone who says the Cirrus is "unrecoverable" from a stall/spin has no business flying one. But, the fact is, NO aircraft, be it a Cirrus, Cessna, Bonanza, Etc, will recover from a stall/spin at 200' off the ground, there just isnt enough time or altitude for the recovery. It is possible to recover from a stall at that altitude if you are VERY skilled in low level flight ops, but add a spin into that equation, and it's pretty much all over but the crying. Both these men were experienced pilots, both DID have experience in this particular aircraft. In my professional opinion, this WILL end up being a mechanical failure of some sort.To research the question, I pulled NTSB reports that showed probable cause of Cirrus aircraft having to do with loss of control. What I found was surprising. It does not agree with the popular notion that pilot error was to blame.
A former NTSB-trained accident investigator myself, I assure you, prejudice during investigation is always to blame a pilot. There is inadequate focus on the aircraft. Investigation parties of small aircraft consist of an NTSB investigator, possibly an FAA Flight Standards Inspector (there to violate the pilot), manufacturer reps (there to defend the product); there is never a human factors expert.
As you accuse me of bias being a Cessna sales rep myself, know yourself that NTSB reports and probable causes can themselves be biased and shortsighted. We're trained to believe flying is generally safe, and that pilot error or rare mechanical defect is only to blame for accidents.
I admit myself a prejudice that Cirrus marketing is focused on selling high performance planes to newbie pilots . What surprised me, though, in studying the accidents was to find that pilots who lost control of the Cirrus plane were by in large pilots with great experience, pilots you would never think could get themselves into surprisingly difficult situations.
How many hours doth make a safe pilot, one you wouldn't blame? What is your total time? How many hours should you have to fly a Cirrus?
The average pilot time in loss-of-control Cirrus accidents was:
a) 100 hours
b) 200 hours
c) ...
Nope.
The average Cirrus loss-of-control pilot had 1,312 hours total time and 249 hours make and model. Although it makes no sense to do so, I’ll take out the pilot with 12,000 hours and 1,000 make and model, and the pilot with 12,773 hours and 681 hours make and model. The average still results in 954 hours total time and 228 make and model.
| Total Hours as Pilot | Number NTSB Findings Cirrus Loss of Control |
| < 100 | 0 |
| 100 - 249 | 2 |
| 250 - 499 | 12 |
| 500 - 999 | 20 |
| 1,000+ | 19 |
It takes the NTSB months, sometimes more than a year to complete an investigation and name a probable cause. To date, since 1999, there have been one hundred sixty-eight NTSB investigations of Cirrus accidents, of which one hundred twenty are complete and indicate probable cause. I examined only the completed reports.
Fifty-three finalized NTSB reports (1999-2010), with one exception, show loss of control or an inability to maintain control of a Cirrus aircraft. As you read, be a good investigator. Most reports conclude the pilot "failed" to maintain control. Take exception to the easy-to-use word, "failure", and consider substituting that the pilot had difficulty or couldn't maintain control.
Read these twenty-six examples of Cirrus loss-of-control accidents to get started. In each scenario, avoid prejudice in an assumption that blames the pilot. You may find it especially interesting the cases where pilots live to tell their story. Try to imagine the story of dead pilots too. Make effort to empathize with them. Try to believe pilots possessed some measure of skill as they struggled for their lives.
Until you spend an hour reading those, I'll summarize: Experienced Cirrus pilots have been unable to escape icing conditions. Experienced pilots, even in good weather conditions, inadvertently entered stalls that developed into unrecoverable spins or sprials to the ground. Experienced pilots crashed while landing in regular and gusty wind conditions.
As you question, what might be wrong with the plane? When will the NTSB look deeper? Here is an example. Last year the NTSB reacted to a string of in-flight structural breakups of Zodiac CH-601XL airplanes designed by Zenair, Inc.
What it takes for the NTSB to become aroused is a high profile accident or continuously high fatality rates. Save this post. I tell you now that the Cirrus fatal accident rate will fall under increasing scrutiny, and serious attention will be given to the Cirrus flight control system.
Excerpt of the National Transportation Safety Board Safety Recommendation regarding Zodiac aircraft fatalities, dated April 14, 2009:
A sufficient stick-force gradient is required for pilots to maneuver an airplane safely. The Safety Board recognizes that experimental and light sport airplanes typically exhibit lighter stick forces than airplanes certified under 14 CFR Part 23 and that, if properly trained, pilots can safely maneuver airplanes with relatively shallow gradients. However, even experienced pilots may find control difficult if the gradient is not constant but instead lessens as Gs increase. With a lessening stick-force gradient, it becomes easier to inadvertently overcontrol the airplane and reach higher acceleration forces than intended.
Uneven stick force gradient was partly blamed in Zodiac accidents. Low, possibly no stick force gradient is a problem with the Cirrus aircraft. It's prone to overcontrol also, a factor in many of the cases you just read. The Cirrus control system offers no “feel”, very little aerodynamic resistance because its control mechanism is centered by springs, not by aerodynamic pressure. A Cirrus control in flight feels the same to a pilot at any airspeed. Furthermore, small hand movements command full flight control deflection. The Cirrus joystick is so touchy that Cirrus instructors teach pilots to squeeze the control handle instead of pull back to rotate at takeoff.
Control force to manipulate a stick should not be so low that an airplane is inadvertently overcontrolled. If you're a non-pilot reading this, set your computer mouse so it's too sensitive, so that a small hand movement sends it across your screen. Now, bounce around in your chair and attempt to highlight this sentence. You get the idea.
Any lack of stick feel with various load factor or speed changes makes an aircraft more difficult to control. Overcontrol during normal cruising speed can result in overstress of an airframe. Overcontrol at slower speed can result in stall. A Cirrus control stick inadvertently slapped about during turbulence or by the pilot searching for controllability further exasperates a stalled condition. Full down aileron more deeply stalls a stalled wing. A raised aileron has no affect on recovery.
What's more, an aircraft that requires extra skill to hand fly substantially occupies a pilot's attention affording less ability to maintain situational awareness. The more difficult a plane is to control or trim, it makes it substantially more difficult to navigate, communicate with air traffic control, interpret flight instruments and manage passengers. God be with those at night, in IFR or dealing with any other complexity if the autopilot is not used.
Ray Hecker, Experimental Aircraft Association (EAA) Flight Advisor, Flight Instructor, and FAA FAAST Representative (Aviation Safety Counselor) warns about flying aircraft with low stick force gradient in an online article, “Flying By the Numbers”, February 27, 2009:
When you have excess power and a very slippery designed airplane, or you have low wing loading, getting away with maneuvering without adding power can be achieved initially, but it will catch-up to the unwary pilot in the long run. When flying a light stick-loaded (low stick force gradient per G) airplane, like most of our higher performance experimental designs, BEWARE; you may be operating close to the performance envelope. You are inviting a stinging snap-role or at least a departure into an incipient spin due to mismangement [sic].Cirrus control feel is reported by Cirrus instructors and Cirrus aircraft owners, as surprisingly light and unconventional although no one has been willing, aware enough of its consequences, humble enough, or bold enough to label it as unsafe.
NTSB report CHI01FA169:
On June 13, 2001, the pilot began a three-day ground and flight training program conducted by Wings Aloft at CDC [Cirrus Design Corp]. The first day of training consisted entirely of ground training due to weather. The pilot stated she was told that the airplane was "finicky" on landing and you don't want to flare but rather land flat.Exerpts from: "Cirrus SR20", an owner's review by Philip Greenspun, ATP, CFII, in July 2005, updated January 2010:
An SR20 owner, who was involved in a propeller strike [not related to this occurrence], said that the he thought that the transition to flying an SR20 is the toughest aspect of flying the airplane. He thought that the controls are very sensitive and also noticed that you can reach the end of control in pitch and remarked, "all of a sudden you're pulling back and there is nothing left."
In terms of avoiding an accident, one problem with the Cirrus is its unforgiving handling compared to other basic four-seaters. For pilots accustomed to learning about an impending stall by feeling reduced airloads on the flight controls, the Cirrus provides much less stall warning. This is due to spring cartridges that continue to resist flight control movement even when the airplane is not moving. In other words, the flight controls feel similar whether you're flying or stalled.Cirrus sales pilots tout lightness of control as being maneuverable and fun to fly. Typical flight demonstration shows how stable and controllable the aircraft is in full stall. Sales pilots demonstrate aileron control left and right as the aircraft buffets. How? It's possible at 3,000 feet above ground where the air is smooth and laminar airflow is assured.
Once a pilot has gotten sloppy with airspeed, the plane is harder to keep level with rudders in a stall than a Cessna or Diamond; if in a deep uncoordinated stall, the Cirrus wants to drop a wing and go into a spin. Thanks to a "split-airfoil" wing design, in which the inner portion of the wing has a higher angle of attack than the outer portion, the Cirrus gives more of a stall buffet warning than many airplanes. The outer portion of the wings, which are in front of the ailerons, are still flying and permitting the pilot to control roll with the yoke, even as the inner sections of the wings may be stalled and creating a warning buffet. This illustrates one of the advantages of composite construction; you could build a metal wing like this, but it would be very costly.
A pilot with 800 hours in the SR22 noted that in his experience it is not nearly as docile as the Cessna 172 and Piper Arrow that he had trained on. A CFI ("certificated flight instructor") who now flies the $3 million Pilatus PC-12 says "The Cirrus is a plane designed to go fast. You shouldn't be flying it slow. It is trickier to handle in a stall than a 172 or the Pilatus."
Once in a spin the SR20 and SR22 are virtually impossible to recover, according to the test pilots. Remember that spin testing in certification is done with a special tail parachute for breaking the spin that can then be cut away inflight.
A high-time CFI friend of mine had the opportunity to buy a low-time SR20 for around $135,000 from a guy who was basically afraid to hand-fly the airplane (note to shoppers: look for people like this at your local airport!). Despite the unarguably fabulous value, he decided not to get the plane because he didn't like the springs. "I can feel the air load on the flight controls in a regular airplane," he noted. "And when I get a Cessna or Grumman slow the controls begin to feel sloppy so I know it is time to pitch down for some more airspeed. With the Cirrus there was still a lot of resistance from the springs even at zero airspeed so it was hard to tell whether I was feeling the air loads or the springs."
My 800-hour-SR22-pilot friend says that one of his main complaints with the plane is how difficult it is to trim for straight-and-level flight. "The plane really needs a micro-trim switch," he says. His CFI confirms this impression and notes that the plane is much easier to fly if you back your hands away from the yoke and don't grip the control. "Just fly with pressure from your fingertips," he says. My first efforts were very unsuccessful. What seemed to me like the smallest nudge forward put the SR20 into a 300 fpm dive. The smallest nudge backward and it was in a 300 fpm climb. The smallest nudge left or right and the airplane was in a half-standard-rate turn. After 20 hours I was able to trim the plane for a hands-off straight-head 500 fpm climb through some clouds in Quebec. An airline pilot buddy of mine flew one vectors-to-ILS approach in the SR20 and said "My arm is tired; our Embraer regional jet has a trim speed that varies depending on airspeed; this plane really needs that."
The SR22 CFI related a story about taking friends up to Montreal for the weekend. "We land at a few places and all of the non-pilots take turns sitting in the left seat. I do this trip every year in a 172 and keep my arms in my lap while the non-pilot does all of the flying, including the takeoff, though I'm ready with my feet on the rudders." One year he took an SR22 instead. "It wasn't as much fun. None of them could fly the plane. They couldn't get it trimmed right. They couldn't do the takeoff. They couldn't handle the side yoke."
What's convinving in demonstration does not compare with aircraft performance in turbulent flight conditions typical at pattern altitude and below. Airflow, eddies and currents near the earth are entirely more severe than homogenous air mass at altitude. Have you ever encountered a stall horn chirp on downwind? Without a compounding effect of load factor or pilot overcontrol, eddies, turbulence and sheer can put a wing or some part of it at sudden and excessive angles of attack.
Experienced pilots, accustomed to traditional airfoils underestimate the devastating effect that airflow disruption has on super-smooth NASA AGATE-designed natural laminar flow (NLF) airfoils. Experienced pilots know high lift wing characteristics like those of Bonanzas and Cessna offer lots of safety margin. They also know to give greater care with semi-laminar flow wings like those of Centurians and Mooneys. Pilots are trained to believe the Cirrus wing is more safe, even as purported in Mr. Bertorelli's video, yet this is not true. What it is is high performance.
A laminar flow wing does not recover from stalled condition readily as traditional high lift designs do. The more efficient NLF airfoil gains a larger proportion of its lift from Bernoulli's principal (the suction effect of smooth airflow atop the wing). When smooth airflow is disrupted either by high angle of attack or by ice build-up however small, air does not suddenly re-adhere to the wing surface. The loss of lift is more devastating than pilots are prepared to handle.
Accident reports like this one record the disparity between what pilots are used to experiencing compared to what they experience in the Cirrus. Again, this pilot is considered at fault:
ATL06LA035 The pilot, a factory-trained Cirrus Standardized Instructor Pilot (CSIP), had 12,773 hour total time and 681 hours in make and model.
Upon reaching 7,000 feet the airplane encountered icing conditions. The pilot informed the controller he was encountering trace icing conditions at 1553 and requested clearance to climb to 9,000 feet. The controller cleared the pilot as requested and informed him, "cirrus eight seven hotel kilo uhh that's been uhh pretty much the uhh norm all day climb and maintain nine thousand." The pilot entered a 500 foot per minute climb into the autopilot and initiated the climb to 9,000 feet. As the airplane reached the clouds tops at 8,000 feet in visual flight conditions, the airplane began to buffet. The pilot looked at his airspeed indicator and it indicated 80 knots. The airplane stalled, the nose pitched down, and the airplane started spinning to the left while reentering instrument flight conditions. The pilot reduced power, neutralized the flight controls, and applied right rudder with negative results. The pilot activated the Cirrus Airframe Parachute System, and the parachute system deployed.Do not construe this as a defense of flying in icing conditions, but the above pilot’s encounter started at 7,000 feet in what he reported as trace icing. Tops were 8,000, which means the clouds were less than 1,000 feet thick.
My story, published in AOPA Pilot Magazine, April 2006, was eerily similar to the above, except I did not enter a deep stall. I did, however, discover another hazard:
The Cirrus elevator forms with the stabilizer leading edge where it's exposed to icing. It is completely prone to ice bridging that acts as a control lock. This elevator control froze in actual icing conditions.
An SR20 Elevator tip leading edge showing ice bridging that caused an elevator control problem.
Here is a TKS-equipped SR22 elevator tip leading edge. Notice the great leverage, distance forward of the hinge, ice formation has against pilot control.
Rear view of ST22 elevator configuration.
This is a similar aircraft, the Columbia 400's elevator which is not designed in a manner that's subject to leading edge ice formation.
A Cessna Skylane elevator is protected from leading edge ice obstruction due to a large gap between the elevator tip leading edge and the stabilizer.
The FAA’s stance: The Cirrus was not certified for flight into known icing conditions.
My position: That’s true, however the Cirrus is IFR-certified, equipped with pitot heat, and offered with FAA-certified TKS ice protection. Any IFR-certified aircraft should afford a pilot reasonable opportunity to escape unexpected and inadvertently encountered light icing conditions. The Cirrus elevator design is totally exposed to ice bridging that results in frozen and obstructed elevator control.
Cirrus has been given a pass for bringing revolutionary design and techonologies to market. Yet, there’s something odd about its total lack of critics, in spite of its track record. Cirrus Design will continue to lure pilots into situations that are beyond what reasonable skill level and human attention can be expected to handle.
Be warned Cirrus pilots. You are at more risk than you may know.
Posted by Stephen T Wilson at 12/20/2010 8:15 PM 
Categories: Flight Instruction, Air Safety
Tags: Flight Instruction
Categories: Flight Instruction, Air Safety
Tags: Flight Instruction
Trackbacks
-
12/21/2010 5:13 PM
Aviation Law Monitor wrote:
Steve Wilson argues that there are safety issues with Cirrus airplanes. First, Wilson feels that the Cirrus is more prone than your typical Beechcraft to crashing due to the pilot losing control of the aircraft. Second, Wilson feels that the...
Cirrus aircraft remind me of the time when Bonanza's were called "doctor killers". It seems like the Cirrus is the new "doctor killer" of today.
Reply to this
Stephen . . .
Your analysis is thorough and compelling. The danger of leading-edge ice obstruction was poorly considered in the Cirrus design. It should be well considered by any Cirrus pilot who might encounter such conditions.
As you say: "Be warned, Cirrus pilots. You are at more risk than you may know."
Reply to this
Stephen,
I couldn't agree with you more. The Cirrus planes are dangerous and kill people.
Reply to this
Stephen,
You are so completely incorrect. Your "expert" opinions are totally without fact or merit. It is unfortunate that you pretend to be so well informed but yet have very little knowledge in flight test or design; or apparently as a pilot it seems.
In full disclosure, I am bias to the Cirrus, with well over 3,000 hours flying the Cirrus in all types of conditions (including ice) and as a former factory demonstration pilot. Prospective owners have put the Cirrus in many unusual and unrecommended attitudes in an effort to "test" the Cirrus (not to mention the FAA through certification). I have never once seen the Cirrus "loose control." The wing cuffs or spring loaded flight controls do not dampen the flight characteristics of the plane - they have in fact made it safer and more enjoyable to fly!
As for icing, I suggest you speak to any flight test engineer who has take any plane through Part 23 FIKI certification. You will find your conclusions to be totally off base and incorrect. Or better yet, let's go for a flight!
Reply to this
Thank you for the comment. I appreciate your testimony that’s exactly consistent with every pitch of demonstration pilots with which I've been accosted.
Reply to this
It will be intersting to see how the new "stability control" feature available on new Cirrus airplanes, will help the safety record.
http://www.avweb.com/avwebflash/news/cirrus_perspective_esp_hypoxia_stability_203851-1.html?CMP=OTC-RSS
Reply to this
Dear Sir, I have about 1,000 hours with over half in multi engines flying in Northern Alberta and the Yukon. I am also a former CFI with hundreds of hours in Cessna's. I recently took my first flight in a SR20. My first impression is that this is NO CESSNA! This is a complex, high performance airplane and I was just in the 20. What I think would be interesting is to know how many accidents in the Cirrus were with pilots who did not go through the Cirrus recommended and recurrent training? Cirrus uses a different philosophy in its design so to should the pilots who fly it. If you are properly trained in the airplane, and undergo the recommended recurrent training this is a perfectly safe airplane to fly. If you think you can fly a Cirrus like a Cessna you are not getting the point, this is not a Cessna and do not try to fly it like one.
Reply to this