Cirrus SF50 Vision Jet

An epic flight in the first truly personal jet

Up where we were, it was smooth, cruising along at Flight Level 270, Utah’s Painted Desert, swaths of reds, browns and golds panning behind us as we flew. Just like the light, we were, as they say, golden, with pilots in other planes at altitudes both below us and above squawking nonstop about the rough rides they were suffering. Here in the 20s, not only were we by ourselves—the airliners thousands of feet above us and the pistons thousands below—but we were also apparently the only ones enjoying a smooth ride. Not a bump anywhere.

It’s an apt metaphor for the remarkably successful Cirrus Jet, a single-engine, 300-knot 5-7-seat personal jet, that exists in a space all its own. There’s no airplane that’s a direct competitor, only ones that are possible alternatives.

Courtesy of Cirrus Aircraft

Part of this is because, well, the SF50, for which Cirrus earned type approval late last year and production approval, impressively, just a few months later, is the only single-engine jet. And at a price of $1.96 million, the air gets even more rarified. The airplanes that are competitive in terms of price aren’t as fast or can’t go as far, or both, and the would-be competitors that can match or beat its performance cost a lot more. It’s been a kind of sport to beat up on the SF50’s performance numbers, but that takes performance in a vacuum, removed from the very real issues of purchase price and direct operating costs, not to mention the very real and more difficult to quantify quality-of-life issues.

I’ll admit that for years I misunderstood the aircraft. I was confused when the company announced the jet and its performance targets, and I was baffled as orders built, eventually rising to around 600 orders today. Still, I wanted to fly the jet. I have seldom, if ever, been so anxious to get to fly a new model. In part, I admit, it was because I figured I must have been missing something.

I was right. I was.

The Cirrus SF50 Vision Jet is one of the most talked about airplanes in GA history, and its on-again-off-again business history underscores the great risk and challenge of bringing a jet of any description to market. The costs, complexity, regulatory hurdles and engineering budgets surrounding an effort like this would tempt any financial analyst to abort takeoff before decision speed, that is, the speed at which you need to hit the brakes right now or commit to taking off.

I was up in Duluth, Minnesota, the then-home of Cirrus Design, not yet Cirrus Aircraft, when the company launched the jet with a full-sized mockup. Over the intervening years, the company went through struggles related to the economy and to bringing a jet to market. It was a rocky road getting there, a subject about which I could devote a long article, but the bottom line is, they succeed.

Courtesy of Cirrus Aircraft

Singular Challenges

When it comes to designing a single-engine jet, there are a number of engineering challenges that one can best describe as treacherous.

The Cirrus brain trust, in fact, designed several of those problems out from the get-go, giving the jet a ceiling of 28,000 feet, the same as single-engine turboprops, which routinely fly in the mid- to-high 20s, where a pressurization failure, while still an emergency of the first order, isn’t nearly as potentially lethal as such a failure would be in the high 30s. The ceiling also eliminates the need for the manufacturer and owners to get RVSM approval, which is a difficult task.

It’s also important to remember the SF50 isn’t a particularly powerful jet. Its single engine, a Williams FJ-33, is a smaller version of the hugely popular FJ-44, which debuted on the CitationJet in the early ’90s and which has racked up an impressive record of reliability and performance. It’s safe to say that many of the potential problems for the FJ-33 were discovered and solved decades ago on the larger engine. So, in this case, the introduction of a new engine didn’t come with the same degree of risk as it might otherwise have carried with it.

The Williams FJ-33 hugs the top of the fuselage and creates very little pitching moment when power is applied or reduced.

The 1,800-pound thrust Williams powerplant is also, well, sized just right for the Cirrus jet. With its fine specific fuel efficiency numbers, the jet gets good range even in the 20s, where turbofans are less efficient than turboprops and where the air is slightly thicker, allowing slower true airspeeds. Bottom line: Jets typically belong in the mid-30s and above, but with less power comes less potential, so the fact is that the Cirrus Jet probably isn’t leaving too much on the table by not sticking with sub-RVSM flight levels.

Oh, and for ice protection, Cirrus went the conventional route, with boots for the wing leading edges and tail leading edges and TKS for the other vulnerable areas.

The passengers in the SF50 have huge amounts of legroom and headroom, and fantastic views of the outside world.

There’s also, as you probably know, too, a whole-airplane-recovery parachute system, which features a giant canopy. Unlike the mechanically deployed chute in its singles, the chute in the jet is tied in with the autoflight system. When the pilot pulls the big red handle, the system slows the plane to the max deployment speed of 135 knots before the rocket is fired. The interval between pull and the rocket firing can be lengthy, but automating the speed target, says Cirrus, takes a lot of the risk out of the deployment.

Another good thing about the lesser power of the FJ-33 is that the Vision Jet doesn’t exhibit a lot of pitch deviations when power is applied or reduced, a result, in part, of putting the engine very close against the fuselage and, again, with it being a lower-powered turbofan.

The other part of the equation is the design of the SF50’s tail. As you see in the accompanying photos, the tail, while reminiscent of the empennages of some other models, is unlike any GA plane ever produced. The V-tail shape isn’t necessary for a single-engine jet, but unless you want to put the engine inside the fuselage or mount it as part of the vertical tail (as Piper did with its short-lived PiperJet), the V-tail is the route to take, as it frees up the real estate for the turbofan’s exhaust to shoot out.

This isn’t unique to light GA but another feature of it is: the dual strakes on each ruddervator. The strakes are controlled not by the flight controls but by computer, to help control the tendency of such tails to exhibit instability in Dutch roll.

In addition, Cirrus designers canted the engine’s outlet, based on the results of flight testing, to further fine-tune stability. I would soon see how well it all worked, well, with the exception of the chute, about which I hoped I’d learn very little.

The V-tail features a pair of strakes that are computer-controlled to manage yaw and dutch roll. It works great.

Cabin Rules

The SF50’s cabin, like many other elements of the jet’s design, is unlike anything you’ve seen before. There’s a single door on the left side of the plane, so everybody boards via that portal. It’s not easy to make a great door, and it’s even harder to build one that’s light in weight, but Cirrus pulled it off. The bifold door opens to reveal a good-sized opening. The pilot’s seat slides far forward to give the back-seat passengers easy entry, and there’s plenty of room for the right-seat occupant/co-pilot (it is, of course, a single-pilot jet) to get into the right seat.

Cirrus left a huge open area in the center of the cabin. The two middle-row seats are situated at the side of the cabin. Behind them are three seats up against the rear bulkhead. The two outward seats in back offer enough room only for kids or the occasional flying-friendly pooch, with the center seat, located just behind but directly in the middle, leaving the space in front of it empty and affording that middle-back row seat about six feet of leg room. For all intents and purposes, the SF50 is a five-seat airplane with two jump seats.

When you’re flying a jet you don’t want just any kind of door; you want an airstair door, and Cirrus knocked it out of the park on this one. Light, easy to operate, impossible to close incorrectly, and still big enough to allow easy entry, the SF50’s door makes getting into a Cirrus easier than ever.

Up front, the pilots’ seats are well situated, with a good deal of headroom, and the shoulder room is so good Cirrus probably could have wedged a third seat in there. The windscreen is divided in the middle by a structural member, but the forward visibility is nevertheless quite good while the side view is nothing short of spectacular.

The avionics suite is the Garmin G3000 system, which Cirrus has tweaked to be a great fit for the Vision Jet. There are two big, 14-inch displays up front, while a step below and closer to the pilots are three touch controller displays mounted sideways, the far right one being a more or less dedicated comm controller. Like its piston-single counterparts, the SR22 and SR20, the SF50 features side-sticks, though these are actual sidesticks and not side yokes, which travel fore and aft like a conventional yoke, as is done on Cirrus’ piston planes.

The range of capabilities of the autopilot is astounding. It will even slow the jet down to optimal chute deployment speed.

Without going into a great deal of detail on G3000, suffice it to say that the system is a great fit for the Vision Jet, and that this installation has a number of features that I’ve never seen on Garmin’s touch-controlled system before. A few of them would come in very handy down the road.

There’s also a first for a Cirrus, real-time radar, which is a great addition but will certainly present a learning curve for new jet pilots, and I suspect that some time will be spent during training for the type certificate on just that subject.

Everything from startup to shutdown is done through the Perspective Touch cockpit. The Garmin suite features three landscape touch controllers, the far right of which is a dedicated comm panel. With large, multi-pane-capable displays, the avionics suite gives SF50 flyers the option to customize the view to meet the needs of every phase of flight.

Finally Flying The SF50

I had agreed to meet Cirrus’ Matt Bergwall in Van Nuys, California (KVNY), the famous jet port, the perfect spot to fly the latest in jet technology. It was Sunday morning when we climbed into the jet for the trip, and from the start flying the jet was filled with new, unexpected experiences.

The flight-planned mission was to fly back to Austin, Texas. The idea was to give me the chance to get a feel for how the jet would perform in real-life missions and not as a chariot for getting to the latest classic car show, golf weekend or marlin fishing tournament.

Even the preflight was a pleasant surprise. There are a number of tasks that you need to do in a jet that are either not required or are routinely shrugged when flying a piston single. In the SF50, a number of those tasks normally done with pencil and paper are automated, including weight and balance, which you can do graphically, as well as the calculations for takeoff and landing speeds.

Cirrus’ Matt Bergwall (near right) and the author at FL270, cruising along at 300 knots on 67 gallons of Jet A per hour.

Likewise, again, thanks to ingenious design by Cirrus and Garmin with the Perspective Touch system, the start sequence on the SF50 is just as easy. After you initiate the start, the system runs its own series of checks on everything from TAWS to fire suppression. The start itself is just as easy. Move the dial from “OFF” to “RUN,” push the start button just above it, and watch the Williams FJ-33 start itself. In case of a bad start, it even will shut itself down. For those observers who understandably questioned how well less experienced pilots might transition to the many technical details of flying jets, the user-friendly design of the start sequence indicates clearly that Cirrus was ahead of that issue.

Once you get rolling, forward taxi visibility is terrific. You steer the jet with its steerable nosewheel, but if you’re coming from an SR22 and are used to using differential braking, well, that will actually help matters considerably.

We were pretty close to max takeoff weight for our departure from the 8,000-foot runway, and as we got cleared for takeoff and I poked the takeoff/go-around button on the thrust lever, which made the flight director chevrons magically appear on the PFD, showing me graphically what I knew already, that my pitch attitude was going to be substantially less than on any jet I’d flown. I looked over at Matt, made a quick check of trim and flaps and engine parameters as I pushed the thrust lever to takeoff power, and away we went.

The start sequence on the Cirrus Jet will make pilots type-rated in most other jets long for such simplicity.

Rotation speed that day was 90 knots, and it took a while to get there on that warm morning. I, again by training, couldn’t stop myself from looking for V1, the speed at which you’re committed to go flying in the event one of the engines fails. There’s no V1 in this jet because if the one and only engine quits on you, you’re not going anywhere except for the brakes regardless of what speed you’re at. This is one of the downsides of a single-engine jet, the lack of redundancy, and nowhere will this be more acutely felt than at this juncture of any flight.

Needless to say, it wasn’t an issue, and I rotated the jet at 90. While still a very satisfying feel, the control response of the SF50 is heavier than I’d expected, though I’m not sure why I expected differently. The procedures for takeoff are more like a high-performance single than like any jet I’ve flown. Positive rate, gear up, flaps up at 115 knots and pull back to maximum continuous power when it’s safe to do so. I felt supremely un-busy, though I did have the advantage of Matt running the radios while I got to know the airplane.

Unlike on other jets, where busting the speed limit for the airspace you’re in is a real risk, you don’t need to worry so much about that in the Vision Jet. Its one 1,800-pound thrust engine isn’t going to rocket you through 250 knots in a hurry, so that’s one less thing to worry about. In fact, I found it really easy to manage while hand flying the step climb imposed upon us on a surprisingly busy weekend morning in the Valley.

FMS waypoint information displayed on the far-left controller makes it easy to keep track of the flight plan and make changes.

The flight director will give you 160 knots indicated for the climb, which is a typical figure for the jet, and at that setting and MCT, we were burning 75 gallons per hour. In jets you typically go by pounds per hour, which takes a while for pilots to get used to when transitioning to turbines, but on the Vision Jet, Cirrus chose to go with the more piston-pilot-friendly gallons per hour index, which seems suitable since the figures are so low.

In no huge hurry, we enjoyed a long, vectored circumnavigation of the south San Fernando Valley, step climbing all the way. The visibility was so good we might have been able to see, forget Catalina, Hawaii, had it not been for the curvature of the earth. After a bit, we got turned northward, in smooth air, and continued our climb above and beyond the San Gabriel Mountains, passing just south of Rogers Dry Lake and Edwards Air Force Base, where they’ve tested a few jets in their day as well.

One of the greatest and most overlooked achievements of the SR22 is the simplicity of the power and fuel management chores. These same tasks in the jet are, if anything, even easier. The power management is via FADEC, so that’s even easier and smarter than the simplified power management on the singles, and keeping an eye on the fuel involves pretty much just keeping an eye on it, as the system automatically switches the fuel tanks it’s drawing from based on the levels.

The FADEC Williams FJ-33s keep the temps in the green with margins much tighter than most humans are comfortable with.

The trip was around 1,200 nm, and with a forecasted 25 knots of tailwind, we could have probably made it nonstop. Matt likes the word “probably” about as much as I do, and I was looking for a little more hand flying anyway, so we scheduled a no-brainer stop about halfway to Austin, in Taos, New Mexico. But as we flew, the temps out that way were rising faster than forecast, and with moderate turbulence and wind shear expected, we decided to go to Santa Fe instead, for its longer runways and slightly lower elevation.

Cruise in the SF50 is much to my delight exactly what Cirrus said it would be…no, actually, it was consistently a little better even than that. Heading east at FL270, we were truing 300 knots or a couple of ticks higher while burning just over 65 gallons per hour of jet fuel. The view was amazing. I took a lot of photos with my iPhone, some of which featured the striking landscapes of the American Southwest as framed in the cool oval windows of the jet.

The Vision Jet features large, nicely placed windows, good for passengers and pilots alike. An emergency escape hatch is seen at the right in this shot.


As we got cleared for the descent into Santa Fe, we made use of the Garmin G3000’s Vertical Descent Profile display on the MFD to see exactly where we were on the descent and where the terrain was, too. Going into Santa Fe from the west, we flew directly at the ridgeline to the east of the airport, which was more disconcerting in visual conditions than in IMC, I’m certain, before getting the turn to final.

As we descended, we encountered strong moderate turbulence below the ridgelines. Matt and I were both happy we had opted for Santa Fe. As we descended on the arrival and turned final for the RNAV for the 8,300-foot-long Runway 20, I extended the first notch of flaps and watched the airspeed melt away. For those pilots used to the relative challenge of speed control in the SR22 (which requires more art than technique, I sometimes think), nailing the speeds in the SF50 is cake. With flaps that can come in at 190 knots and go to full at 150 knots, along with the gear, which you can extend at a high 210 knots, you’ve got tools galore.

In fact, it’s so easy to fly the SF50 in its slow speed regime it has powers that previously only turboprops could boast. For lack of an official term, I call it helicopter mode. At dinner the night before our long flight, we ate with a number of Cirrus demo pilots who were in Van Nuys for a sales event. A couple of them referred to the nose-down attitude you can cop in the jet, and in Santa Fe I saw it firsthand, and frankly, it will take pilots with experience in other jets some getting used to as the windscreen fills with a view of the terrain below and the speed somehow magically refuses to build much. In my notes for the flight, my one remark about helicopter mode was comprised of a single word, “Wow.”

Our reference speed for our approach was just 90 kias, so when Cirrus jokes about the SF50 being the slowest jet, it’s a backdoor brag, too. The jet doesn’t need a lot of runway. My first landing was terrible, and I made every mistake Matt warned me not to make, which shows that Matt pays attention and also how slow a learner I am. I flared too high, as Matt predicted I might, and my recovery was a bit ham-handed, too. The entire approach, in my defense, was Mr. Toad’s Wild Ride, with multiple episodes of plus or minus 10-knot wind shear and squirrely, gusty winds once we arrived above the tarmac.

The learning curve has to do not with the landing qualities of the jet—it’s really very conventional—but with the sight picture. First, the visibility is so good it’s hard to get a good read on how high you are—apparently, the effect is that you think you’re lower than you actually are. The good news is that even with a high flare, a short recovery and a bit of a bounce—the trifecta—we still used well less than half of our runway length. And it was in spite of and not because of my brilliant technique.

Into The Storms

After lunch and much talk of the jet, Matt and I headed out again. After a routine takeoff, we climbed out, direct to FL270, which was in stark contrast to our earlier step-climb tour of the Los Angeles basin. Expect around 2,000 fpm for the climb, again, substantially less impressive than in any entry-level twinjet.

We also saw that on the Sirius/XM weather that the earlier FAA forecast for possible convective activity in Central Texas, where we were headed, was looking like a big understatement, as multiple lines of storms formed and marched southwest to northeast.

As we flew what was planned with an ETE of one hour and fifty minutes, we watched the storms on XM grow. First, they engulfed San Antonio, just south of our destination of San Marcos, and we began looking for alternates farther north and farther west. To avoid a particularly nasty area of weather just southwest of the Dallas Class Bravo, we got a longcut (is that a word?) to take us east with a planned 75-degree turn to the right. By then, however, the storm had moved farther north and San Marcos was getting hit hard. Our next alternate, Austin Executive, north of the international airport in Austin, was our next great idea, until the storms started moving north toward it, too.

Our flight in the Cirrus SF50 Vision Jet was far from an afternoon stroll in the park. Our last leg was into an area of fast-building convective activity, but with a plethora of information on the status of the storms, we kept the wings dry and made an uneventful landing near our original destination.

In the stormy darkness, we finally decided on Temple, about 60 miles north of Austin, because it still had marginal VFR conditions and no nearby storms…yet. As the controller descended us down to 3,500 and toward our alternate, we broke into the clear below what looked to be a uniform ceiling above us at around 4,000 feet. With good visibility and no sign of storms visually or on either radar between us and the airport, we decided again to give Austin Executive another look. Sure enough, as we approached we saw a clear path with storms visible but not in the way. We cancelled IFR in visual conditions and I hand flew the approach, going directly to the runway. As we approached, Matt demonstrated again the helicopter mode descent and despite our cutting the corner, the glidepath marker was soon centered and our approach stabilized. Matt handed me back the plane and I made my second landing in the jet, lightning blazing along a storm wall 10 miles to the northwest.

My second landing of the day was, well, not perfect, but really good, and I braked aggressively and exited the runway at an intersection that piston singles sometimes miss.

I was home and I was sold. All of a sudden, the success of the jet made a lot of sense to me. Here you have a 300-knot jet with economy and comfort that rival or surpass some turboprops while offering an ease of operation comparable to some high-performance piston singles.

Courtesy of Cirrus Aircraft

Cirrus has 600 orders for the jet, and while a lot of those early orders were locked in at a price of around $1.4 million, there’s a healthy market for those positions with buyers gladly paying a premium—sometimes over the current $1.96 million asking price—to get their SF50 sooner. The company hopes to turn out 40 of the jets in 2017 and 80 in 2018, ramping up from there.

Once again, as it did with the introduction of the SR20 and SR22 with their revolutionary design innovations, Cirrus has disrupted the aircraft manufacturing industry. Today, there’s a sub-$2 million option for proficient pilots of high-performance singles and twins to move into the world of flying their own single-engine jet. And after my flight, I saw that it did, indeed, make a lot sense. Again, my one-word note on the subject applies. “Wow.”

2017 Cirrus SF50 Specifications

The Cirrus SFR50 Vision Jet we flew for this report was a nicely equipped 2016 factory demonstrator outfitted with TCAS, TAWS, Vertical Situation Display, Garmin color weather radar, Iridium satellite communications, envelope protection, and much more.

Price: $1.96 million

Want more first-hand accounts of flight experiences in interesting, new, and/or rare aircraft? Check out our Pilot Reports archive.