Glass Panel Failures

I recently returned from Duluth, MN, where I had the pleasure of picking up a brand new Cirrus SR22 from the factory.  The experience left me feeling that we’re becoming a bit too complacent about the miracle of glass avionics. 

During the obligatory tour of the Cirrus plant, I was surprised to learn that starting with the 2006 models, they no longer manufacture airplanes with any analog engine instruments whatsoever.  Instead, they make critical engine data (RPM, MP, oil pressure and temperature) available on both displays.  I guess I shouldn’t have been surprised.  After all, the Garmin G1000 – general aviation’s first and thus far only true purpose-built all-glass avionics suite — has had fully digital engine gauges since day one.

Nevertheless, I question the wisdom of this approach.  On the return flight, we were in solid IMC west of Pierre, South Dakota.  Suddenly all the engine gauges stopped displaying data.  The displays were still there, but no data appeared.  Here’s what we lost:  RPM, manifold pressure, pressures and temperatures, CHTs, EGTs, power output, fuel flow, and fuel totalizer.

Was this a concern?  Sure, but not nearly as much as it would have been if the analog gauges hadn’t been available.  A call to Cirrus revealed that some moisture had probably gotten into one of the processors.  As soon as it dried out, everything came back online.  To it’s credit, the Avidyne primary flight display did annunicate the loss of the engine data processor at the time of the failure.

John over at Freight Dog Tales writes about discovering a G1000 failure mode the hard way:  by having a student relate it to you on the phone after experiencing it first hand.

The autopilot was flying the plane and the PIC was flipping through some approach plates when both the primary flight display (PFD) and mulit-function display (MFD) went black.

The intercom, radios and autopilot continued to function, but the pilots had no navigational display, no way to change radio frequencies, no engine instruments, no indications of the health of the electrical system, and only the back-up steam gauge instruments with which to control the plane. Luckily they were in VFR conditions and were already talking to approach control, who helped them land at a nearby airport to sort things out.

The cause of the blackout and the solution turned out to be both simple and unexpected. Something of which I was not aware, nor were several other pilots and instructors who were consulted on this, is that Cessna still has an avionics dimmer knob and, here’s the kicker, it will override the G1000 screen brightness settings when it is adjusted to any setting other that off. 

I experienced this while doing a ground training session in a G1000 equipped DiamondStar hooked up to an external Ground Power Unit.  I couldn’t figure out why the screens wouldn’t come up after plugging in the GPU.  Eventually I figured out that the dimmers had been turned on and were set to the dimmest setting.  The DA40 is nice in that the rheostat seems to be very beefy, and when the dimmer is “off”, it clicks into place quite solidly in such a way that it would be hard to accidentally bump it out of position.

Cessna SE rheostats have a bad reputation, but the new ones they’ve put into the glass panel planes are different.  They seem quite solid.  In fact, all the knobs and switches appear to be Citation jet-quality hardware.

But yes, that’s definitely a single point failure.  All the more reason to carry a Garmin 396 or other capable handheld GPS.  With that and the standby flight instruments, you could still keep the plane upright and navigate.  You’d either have to talk on the current frequency, or hold down the flip-flip button long enough to set it to 121.5.

I don’t think this is all that uncommon.  I’ve recently read about Airbus aircraft having problems with displays disappearing in flight.  I predict we’re going to hear a lot more about this scenario as analog gauges fall by the wayside.

After the factory tour, I have even greater confidence in the strength and engineering quality of the Cirrus airframe, but my misgivings about the increasing reliance on computer displays remain.

Side note: the good folks at Cirrus were busy expanding their factory.  Current production is 5 airplanes on Monday, 5 on Tuesday, 5 on Wednesday, and 4 on Thursday. The plant uses Friday to perform maintenance and catch up on any backlogs. 

They’re also busy building a new airplane:  a single engine jet that looks an awful lot like the SR20/22.  The prototype is under construction and the first flight is scheduled to take place in the fall.