Is it a mirror on the great debate?
In 2012 I posted an article about what might be wrong with Cirrus pilots. That attracted a lot of attention and is third on the list of most-read AIR FACTS posts. Four years later it is still in the week-to-week top 20.
A lot has changed since 2012, the Cirrus safety record has improved dramatically, and what has happened seems directly related to the great debate about flying through computers v. basic flying. I’ll explain that in a minute.
First, though, a review and the record.
It is not unusual for new airplane types to have problems in their early life. Some of these are related to the design or construction of the aircraft, as was the case with the Cessna Cardinal and 441, but most are related to the interface between pilots and the new design, as was the case with the V-tail Bonanza and the Cirrus. The Cessna P210 and Piper Malibu featured a bit of each.
Those of us who follow aviation safety watch new types carefully. If there is a problem, it is important to get word on it out there as soon as possible. The first I wrote of this about the Cirrus was in the September, 2002 issue of FLYING.
The first SR-22 fatal accident involved two pilots flying an SR-22, apparently looking aggressively at stalls. The airplane entered a flat spin at 5,000 feet and spun to the ground. Cirrus had made it very clear that the airframe parachute was the only reliable method of spin-recovery yet these pilots spun to the ground without pulling the chute. What I wrote then examined why this might have happened.
The comments on my 2012 Cirrus post show that there is still a great lack of understanding on the subject of spins in general and their relationship to the Cirrus in particular.
I closed my 2002 FLYING comments by saying There’s a lot left to learn here. That was quite obviously true ten years later and it is probably still true.
In the June, 2003 FLYING the question was What’s going on with the Cirrus? From April 24, 2002 until January 23, 2003 there were five fatal accidents in Cirrus airplanes. Because the fleet was quite small at the time, maybe about 500 total airplanes, that suggested a really bad fatal accident rate.
At that time I pretty much gave the airplane a pass and concentrated on the pilots. The primary question was whether or not the parachute emboldened pilots and caused them to do things with the airplane that they would not have dared do without the parachute. They were, in particular, having trouble dealing with weather and when things went to hell in a handbasket they were not deploying the chute.
I closed that discussion by saying Cirrus pilots will have to realize that the parachute addresses only a tiny percentage of the bad things that pilots do to themselves in airplanes. In retrospect, that was not entirely accurate because the percentage is not so tiny.
Apparently these and other admonitions fell on deaf years because 2011 was the worst year for Cirrus accidents. There were eight fatal accidents in the fall of 2011, which is pretty bad. The causes were much the same as they had been since the beginning. The 2011 record was what prompted the 2012 post.
Fatal accidents outnumbered chute saves through 2013 when the most dramatic change I have ever seen came to the Cirrus fleet. Chute saves soared and fatal accidents dropped. The fatal accident rate had been above the average for all of private aviation but starting in 2014 it appears to be among the best.
The folks who build Cirrus airplanes had been grappling with the safety question from the beginning. They had good training programs, they offered the high-tech stuff earlier than other manufacturers, and, of course, they pioneered the chute.
It wasn’t until recently that they started a program that really encouraged pilots to deploy the chute if there was any question about the outcome of the flight. I think that is what reversed the relationship between fatal accidents and chute saves.
The other thing that happened early-on but that gained traction over the years is the Cirrus Owners and Pilots Association. Type clubs, as they are called, have always been constructive but the Cirrus group is the most aggressive one out there when it comes to looking safety questions squarely in the eye and being realistic about what is going on with the airplanes.
One pay-off for the effort is that the members of COPA have a much lower involvement in accidents than non-members. I think that is because COPA shoots straight about risks and how they might be managed. They have also been brutally honest about the safety record, from when it was bad to now, when it has become a lot better. The website is both interesting and good.
The Flying Physicians Association isn’t a type club but what happened there some years ago relates to this. Doctor-pilots had a terrible reputation, underlined by such things as the designation of the Model 35 Bonanza as the V-tail doctor killer. The FPA launched an aggressive safety campaign and soon its members had a far better safety record than the doctor-pilot group in general—even when they were flying Bonanzas.
I said in the beginning that the Cirrus record might be a mirror on the great debate about basic flying v. flying through a computer. I’m old but I didn’t forget that because it is always fun to throw a bucket of 100LL on that smoldering fire.
I was not alone in thinking that the Cirrus was oversold as a Boeing replacement when it first came out. It was implied in promotional material, but there was simply no way the average instrument-rated private pilot could use a Cirrus to challenge weather that was being flown daily by the crews of Boeing airliners.
They did try, though, especially as high-tech avionics came to the Cirrus after the first few years of production. There were a lot of weather accidents and a lot of them were clearly the result of the pilot biting off more than he could chew.
In a few of the weather encounters the chute was deployed and when it was, it worked, with one exception. A pilot lost control of a Cirrus in icing conditions at high altitude and didn’t deploy the chute until the airspeed was way high. It didn’t save him. On the other hand, it appears that all chute deployments at or above 1,000 feet a.g.l. and below 200 knots indicated airspeed have been successful.
When you look at how Cirrus accidents have evolved, it is clear that the pilots are becoming ever better at using the high-tech equipment while at the same time becoming ever better at recognizing its limitations. The latter is probably the biggest factor in an improved safety record.
The campaign to get pilots to use the chute when in doubt has been quite effective but I think this does not have a whole lot to do with the improved safety record. Why? I looked at some of the more recent chute deployments and it appears that most are related to power problems.
When you look at fatal accidents in the huge fleet of Cessna 172s, almost none of the events are related to power problems. A 172 might fare a little better than a Cirrus in the hands of an average pilot in a forced landing but it wouldn’t be a big difference. That is why I don’t think the more frequent use of the chute has a big effect on the Cirrus fatal accident rate. I hasten to add that both the challenge and the risk do escalate when the (or an on a twin) engine quits and a Cirrus pilot can calm this escalation quickly with the chute.
There are those who would declare a Cirrus pilot to be a true aviation wuss for popping the chute over something as simple and survivable as a power problem. I’m not one of them. If a pilot pays for a chute, what he does with it is his business. It is even possible to relate use of the chute to using a flight control system to fly an instrument approach. In both cases the pilot is using the technology in the airplane to successfully reunite it with the ground.
As an aside, when the Cirrus was introduced it was said that any airplane that came down under the chute would be totaled. Not true. In 14 of the 68 successful chute deployments, the airplane was repaired and returned to service. Ironically, two of the 14 were later destroyed in a fatal accident that had no apparent connection to the repair.
From the record it is obvious that Cirrus pilots are doing a good job of using the technology built into their airplanes to avoid certain risks and that is why the fatal accident rate has improved markedly. Score one for high-tech.
On the other side of the coin, when you look at more recent Cirrus accidents, a lot occur after a low-speed loss of control at a relatively low altitude. That can only be attributed to a pilot who comes up short in the basic flying department. Score one for seat-of-the-pants flying.
A classic example of basic flying trouble was seen in a recent accident where the crash was captured on video. It was hard to watch. What preceded the crash was an apparent series of problems with a visual approach that led to an eventual low-speed loss of control.
What all this does for the Cirrus is put it with most of the rest of the fleet where the low-speed loss of control is the most frequent cause of fatal accidents. In case you forgot, that was also true in 1938 when AIR FACTS was founded. Progress.
As far as the safety record goes, COPA uses data from Cirrus on fleet size and hours flown to come up with accident rates so the numbers are as close to accurate as you can get. They recently showed the rate to be 0.81 fatal accidents per 100,000 hours over the last 36 months and a rate of 1.18 over the last year.
The FAA is projecting a general aviation rate of 1.03 for 2015 but I think that is suspect. The number of fatal accidents is an absolute but the number of hours flown is a wild guess and the FAA has always seemed to estimate on the high side, substantially so at times. However you look at it, I think COPA is correct in saying the Cirrus rate is better than average.
How does it compare with other types? I developed some rates a while back. The methodology likely wasn’t as accurate is that used by COPA but the rate I calculated for Cirrus airplanes was quite close to the rate they had calculated at that time. My Cirrus rate was 1.52. I showed the Bonanza A36 at 1.81, and the Cessna 182 at 0.74 as matters of comparison. At that time the Cirrus was still settling in where the pilots of those other two airplanes had been flying them for years so I thought the Cirrus rate was not bad considering that new airplanes always have teething problems with pilots.
I think it is obvious what is right with Cirrus pilots. They have become much more realistic about the balance between the risks and rewards of using airplanes like this for purposeful private air transportation. Once that is done, there isn’t a better way to go.