Philip Greenspun’s Weblog

My friend Tom and I made it back from London, Ontario in his new 2006 Diamond Star DA40 with Garmin G1000 glass cockpit. The design has improved a bit since my 2002 DA40 was built. Interior noise with the front vents closed is down to 88 dbA in front and 85-87 dBA in back depending on power setting (still way too loud to get yuppies out of their SUVs; will someone PLEASE make a quiet general aviation airplane?).

 

It was interesting to compare the Cirrus’s Avidyne glass cockpit, which is a couple of years older in design, to the fancy new Garmin.

 

Here are some things that I like a lot about the Garmin installation in the DA40:

 

 

• you have a custom settings memory for individual pilots; I would use this to store “philip-ifr” and “philip-vfr”. VFR Philip wants airspace alerts. IFR Philip doesn’t want to be bothered with 10 strident warnings while transitioning through the cone of Bravo airspace
• the audio panel, part of the G1000, has a “playback” button that plays back the last handful of calls from the radio (JFK, Jr. had a similar device in his Piper Saratoga, which led conspiracy theorists to talk about his plane being equipped with a cockpit voice recorder, like an airliner, and why weren’t the recordings recoverable? (possibly because his device cost $300))
• the airspeed tape has little adjacent v-speed bugs for Vx, Vy, best-slide, etc. (though sadly, though the Garmin has a fuel totalizer and could let you enter the weight of your passengers, the G1000 does not compute these for you and adjust them in flight)
• the multi-function display (MFD) displays an endurance circle of all the places you can go before running down to 45 minutes of fuel and a wider one of the places you can glide to with empty tanks; this “circle” is actually stretched out of shape depending on the winds alot
• you can switch the system from showing magnetic headings to true, critical for doing approaches in the Arctic and simply not possible with the Avidyne
• the primary flight display (PFD) offers an inset map that can show traffic, terrain, or just waypoints and the flight plan right next to the HSI
• the A/P gives you voice “leaving altitude” alerts, unlike the Cirrus
• lots of dedicated, single-purpose knobs, e.g., one knob that always does heading (unlike the Avidyne where the knobs’ functions are modal)
• soft keys, a first for Garmin and long overdue

 

Things I didn’t love:

 

 

• the fuel tank gauges are small, about the same size as the ammeter and other non-critical gauges (i.e., not prominent enough considering the dire consequences of running out of fuel); there is no digital readout of the fuel quantity in the two tanks unless you use the soft keys to go down a page
• the engine gauges display manifold pressure and prop speed, but not % of engine power, as the Avidyne E-Max system does on the Cirrus; in the Diamond you are supposed to pull out the paper owner’s manual and leaf through it to calculate the percent power
• all of the fancy screens give you no vertical guidance on approaches; this doesn’t really matter for an ILS, but for a non-precision approach it would be so useful to have a text readout “you’ve passed the FOBAR waypoint, now you can descend to 2000′ MSL”
• approach plates are not available for the MFD, so you need to buy paper books
• I didn’t see any Victor airways on the MFD, so you need to buy paper en-route charts
• (the combination of the preceding three factors makes the entire G1000 not much more useful than a handheld GPS in a 1965 Cessna 172, though it is undeniably slicker)
• the inset map on the PFD is stuck at the same orientation as the big map on the MFD; this prevents you from having one map “north up” (for communication with ATC when they ask for your position relative to other stuff) and one map “track up” (for figuring out where stuff is relative to where you are going); an Avidyne/Garmin 430 system would typically have four moving maps (waypoints inset into the HSI on the PFD, fixed at Track Up; big MFD map, smaller maps on each of the two 430s, all three of the latter configurable track up or north up)
• Diamond is still selling the King KAP140 autopilot, which is not very well integrated with the G1000. The G1000 has an altitude bug, for example, whose setting is not communicated to the A/P. There are three places to set the altimeter (PFD, backup steam gauge altimeter, A/P). The actual A/P settings are not displayed on the PFD. You actually can’t tell if the A/P is engaged or not unless you look down at the physical unit (compare to the Cirrus/Avidyne/S-Tec where the S-Tec takes its vertical speed and altitude goals from the PFD and the A/P settings are repeated right above the attitude indicator) — unlike the Cirrus, the DA40 is an easy plane to hand-fly and it is very suitable for a trainer where you don’t want the student using the A/P all the time
• the user interface overall is very complex and involves a lot of cursor-knob-enter or cursor-knob-cursor action; with the Avidyne/Garmin 430 combo at least the miserable and confusing part of flying is limited to stuff you do on the 430; the Avidyne soft keys and knobs are very simple
• soft keys (also a plus, above), because some things are available after you bring up the Menu and others only on the soft keys, so now you have an extra place to remember to look
• the wind vector; on the Avidyne this shows up next to the HSI so you can see how it relates to where you are trying to go and it reads out in magnetic direction and speed so you can give Flightwatch a professional-quality pilot report; on the Garmin it is stuck into the inset or MFD map and the direction is not available as a number
• the en-route safe altitude is not displayable for the area that you’re in, but only the maximum ESA for the entire route you’ve said that you’re going to fly; this could be unsettling for a long IFR flight because the G1000 will show that you need to be thousands of feet higher than you are to avoid terrain
• the documentation is poor and reads as though it was written by someone who had never flown an airplane; a lot of material in the thick pilot’s guide is tautological and uninformative, e.g., “you use this to switch the CDI intercept from auto to manual” with no explanation of the consequence of either setting (one solution is to go through the King Schools training system on DVD-ROM, which I tried and found much more useful than the Garmin manual)

 

Things that might be broken or I couldn’t figure out:

 

 

• the COM 1/2 split switch had no effect at all (this is supposed to put the copilot on COM2 and the pilot on COM1 for transmit/receive)
• the side tone was almost inaudible to me when the pilot (Tom) was transmitting to ATC

 

At the end of the day, I was surprised that Garmin had not taken better advantage of the two years it had to study all the things that are good and bad about the Avidyne.

 

Oh yes… time to short Boeing stock (and buy ERJ?). I went up to Canada on one of the newest Embraer 175 regional jets with more than 100 seats (but no middle seats; 2×2 all the way down through Coach). It seems like a great airplane, just as good as Boeing 737 for most passengers, but presumably for a fraction of the cost to buy and operate.

Full post, including comments

Here is a report on the construction of an oil-cooled silent personal computer: http://www.tomshardware.com/2006/01/09/strip_out_the_fans/index.html

There are a few limitations, e.g., “When exchanging components, all of the oil might have to be drained and the hardware cleaned.”  So perhaps it would be better to order a Dell XPS with the new Dell 30″ LCD monitor to replace my four-year-old desktop Windows machine (I had planned to wait for Longhorn/Vista, but it seems as though that could be a long ways off).

I’m off to London, Ontario today to pick up a new Diamond Star DA40 with G1000 glass cockpit.  The owner, a newly minted private pilot, and I will fly it back to Boston tomorrow.  Friday morning, I leave for Kona, Hawaii to finish my helicopter instructor rating at Mauna Loa Helicopters.

Full post, including comments

According to this Avweb piece, Continental has agreed to pay $20 million to the children of some people killed on a 1977 twin-engine Cessna 421.  The settlement came out of a North Carolina lawsuit (where former Democratic VP candidate John Edwards made his fortune suing ob-gyns).  To judge from the Avweb article, the evidence against Continental was slim.  The NTSB report, as usual, blames the pilot.  In this case it seems that the pilot shut down the working engine and failed to feather the prop on the dead engine.  So instead of one strong engine and one low-drag feathered engine, he had one medium-drag idling engine producing almost no power and one high-drag dead engine (though actually the NTSB says that they couldn’t find anything that would have prevented the right engine from operating).  The maintenance on the engines does not sound as though it was done very carefully; the NTSB reported 124 hours since the last oil change (recommended interval is 50 hours).  So… even if the engine really did quit, it might have been due to inadequate maintenance and not Continental’s negligence.  And regardless of the reason for an engine failure, a multi-engine rated pilot is supposed to be able to manage the failure (though of course many are not successful and “the second engine carries you to the scene of the accident”).

What I find interesting about this case is the comparison of the $20 million number to the approximately 1000 new airplane engines that a company like Continental is able to sell every year for certified airplanes (i.e., not kits).  There are about 2500 piston-engine planes sold every year.  About half of these are sold with Continental engines, the other half with Lycoming.  An airplane engine costs around $20,000 new.  So basically this $20 million settlement is equal to nearly all of the revenue that Continental could hope to earn in one year from selling airplane engines and perhaps 5-10 years of profits.

Full post, including comments

We’re looking for mentors to work with student teams in 6.171, Software Engineering for Internet Applications, from February 7, 2006 until May 18.  If you’re a working software engineer, ideally in the Boston area, please visit http://philip.greenspun.com/teaching/6171/mentorship-program and apply!

There is still time to propose a client project for the course as well.  This is especially useful for smaller businesses and non-profit organizations.

MIT students:  Just a few days left in which to apply for the course.

Full post, including comments

My friend Julian and I decided to do a little instrument training flight yesterday in our Cirrus SR20.  The ceilings were perfect for practicing instrument approaches:  a layer of clouds from about 1000′ above the ground to 3000′ above sea level.  Wintertime instrument flying, however, requires being careful about icing.  It was below freezing on the ground.  The weather briefer told Julian that it was a beautiful day for practicing instrument approaches.  There were no airmets out for icing.  There were no pilot reports of icing.  There was supposedly an inversion with warmer air aloft.

After an hour on its electric block heater, we pulled the airplane out of the hangar and noticed some red brake fluid seeping out from an inspection panel on the inboard right wing.  Fortunately, I had some good karma built up with the mechanics at East Coast Aero Club and Rob Brigham zipped over to lie down on the ground, remove the panel, and tighten a hydraulic fitting.  Given the spate of Cirrus pilots who’ve managed to set their planes on fire with the brakes, and the fact that we’d never seen a leaking airplane braking system, we were a bit concerned about this.  But we pushed on the brakes and Rob didn’t see any more fluid coming out.

We took off and did an uneventful ILS 11 approach into Worcester, going missed at 200′ above the ground and climbing back up into the clouds for Lawrence.  We were directed to climb up to 5000′.  With the extra delay of getting the brakes tightened, it was beginning to get darker and colder.  We stepped down from 5000′ to 2000′ for the approach into Lawrence.  Julian was doing a bad job of holding altitude.  The plane should hold 105 knots and altitude with about 50 percent power.  He kept cranking up the power to almost 75 percent.  The plane began to vibrate a bit.  I had spent the first part of the flight checking the wings for ice every few minutes, but had become complacent.  We looked at the wings:  about 3/8″ of ice.  Then the windshield began to ice up.  “Your airplane,” said Julian, who had been flying up until now.  I was in the right (copilot’s) seat, looking sideways at the instruments on the left side.  I asked Julian to engage the alternate air intake so that the engine wasn’t trying to suck air through a potentially iced-over filtered inlet.  Our pitot heat was already on, which prevents the airspeed indicator and altimeter from becoming useless and freaking out pilots.

We reported the ice to the Lawrence Tower, but there was really nothing to do differently because we’d already been cleared for the approach and were planning to descend as soon as we intercepted the glide slope.  One thing that I remembered about icing is not to use the flaps, which increase the risk of a tailplane stall and a steep pitch down of the nose.  We didn’t know what our stall speed would be with the new wing shape, so we kept our speed up at 100 knots until just short of the runway, then slowed down to 85 knots for a no-flap not-particularly-great landing.  Two things were in our favor during this landing:  (1) the ceilings remained high (1000′), so we didn’t build up a lot of ice in the final minute or two and we didn’t have to fly the approach down to minimums (200′ above the ground), and (2) the runway at Lawrence is 5000′ long, so you can have lots of extra airspeed and still stop well before reaching the end.

It was getting on toward 5:00 pm in Lawrence, but the guys at Eagle East Aviation were there.  They had done some maintenance work on my old Diamond Star, and I would always stop in whenever I was in Lawrence.  Tim Campbell, one of the owners, cleared some planes out of his maintenance hangar and helped us pull the Cirrus in over the icy ramp.  Then he cranked up one of the big overhead heaters and we went into the office to hang out and wait for the ice to melt off the wing leading edges and prop.  My primary instructor, Hal Spector, happened to be there.  When you do something unnecessary and stupid, why is it that your teacher must always be present?

After 30 minutes, we had the ice off, the ceilings were still more than 1000′, the visibility was more than 10 miles, and there was no precipitation.  Despite the darkness, we decided to go VFR under the clouds rather than IFR through the clouds back to Bedford (a 10-minute flight).  Lawrence was calling itself IFR, so we needed a “special VFR” clearance to get out (at night, this is available only to instrument-rated pilots with instrument-equipped planes).  I don’t like to scud-run at night, but we wanted to get home (a dangerous tendency in itself) and we knew that we didn’t like the clouds.  Visibility underneath the cloud layer was good and the trip back to Bedford was uneventful.  We put the plane away, checked for more brake fluid leaking (none found), and stopped at Jet Aviation to have some chocolate with the gals working the desk.

My previous encounters with ice were all in situations where it was warm at lower altitudes and those were altitudes where it was legal and safe to fly instruments, i.e., I could have at any time descended and melted the ice off the plane.  In this case, we were at all times close to airports with instrument approaches, but the surface temperature was below freezing and we would be forced to fly the plane all the way to the ground with whatever ice it had picked up.

So… what did we learn?  Don’t fly a feeble non-deiced airplane through a cloud in the winter, even if the briefers say that no ice or precip is forecast, unless there are a few thousand feet of warm air over the ground.

Full post, including comments

This Boston Globe article says that 1 in 20 Boston households have more than $1 million in assets (excluding their house or condo).  Big question:  How to convince these folks that they need helicopter ratings?

Full post, including comments

“You will never know what it’s like to work on a farm until your hands are raw, just so people can have fresh marijuana.” — http://www.newyorker.com/shouts/content/articles/060109sh_shouts

Full post, including comments

The latest issue of AVweb opens with a story about how airplanes like my Cirrus SR20 are catching on fire, with the fires starting inside the wheel pants due to overheated brakes:

there are no design or equipment faults at work, and for Cirrus this leaves only “operator error” as the cause. Unlike many aircraft pilots may be transitioning from, Cirruses have a free castering nosewheel and are steered only with differential braking, plus some positive or negative contribution from the rudder

The Diamond Katanas in which I trained and the Diamond DA40 that I formerly owned both had the free castering nosewheel and as far as I know, no pilot ever managed to set one of those on fire.  Cirrus chose to put the parking brake lever way down underneath the panel where it isn’t in one’s line of sight.  That might be a factor (the Diamond parking brake is right up by the throttle).  Otherwise, it is tough to say.  To folks learning to fly the Cirrus:  remember to push the rudder all the way in before resorting to a light tap on the brakes.  If you feel that you must tap both brakes to slow down when taxiing, it means you’ve got too much throttle in.

Full post, including comments

Current reading:  The Great Bridge by David McCullough, one of my favorite historians.  I’m loving the book so far, especially as it takes us back to an age when engineers were heroes and engineering fascinated the public.

Travel plans:  Flying down to Washington, D.C. (GAI) tomorrow in the Cirrus SR20.  Doing full-down autorotations on Friday at Advanced Helicopters in Frederick, MD (FDK).  Hanging out with friends and family.  Returning to Boston, weather permitting, on Saturday, Dec 31.

Full post, including comments

Friends and readers keep asking me “What is the best digital camera to buy”?  By this, they mean a compact or point-and-shoot style digital camera.  So I have drafted an article simply titled “The Best Digital Cameras“.  Comments would be appreciated.

Full post, including comments