January 21, 2006

Flying Glass (Part 2)

A grey morning, layered overcasts from about 1,000 feet up through maybe 5,000 or 6,000 feet, clear above 6,000, and a forecast for slow clearing across the region. Icing levels about 6,000', but escape routes above and below. Not a bad day to continue my education with the G1000 in N14008, one of the California Airways glass cockpit 172SP's. John arranged it with me yesterday, and although I vowed to get it together this morning, due to other commitments I'm already late and feeling under-prepared by the time I reach Hayward airport (KHWD, my main home base). And, once I get there, instead of filing, pre-flighting, studying the charts and POH, etc., I end up discussing the club with Keith (the owner) and the G1000 with Mal R. (one of Cal Air's instructors), and not noticing the time going by. Oh well.

John calls and asks if it's OK to bring another of his students, Andy K. (whom I've met before a couple of times at the Other Club) as a back-seat observer. Fine by me, and John and Andy turn up as I'm finally getting ready to pre-flight. Andy's wearing shorts, which initially horrifies me — it's California-cold and grey outside — but then I think "Well, he's from Russia..." and it all makes more sense. It is, after all, somewhere above 12C outside already, despite the winter grey.

And so for the first complication: when you fly a 172 with just two people, as long as neither of them are especially overweight and you don't have a lot of luggage, you pretty much can't bust gross weight or centre-of-gravity limits even with full fuel loads, and you basically don't spend too much time obsessing over weight and balance issues. But there are three of us this time, and although none of us is particularly large — I'm 180 cm tall and under the average weight for the FAA standard person (who apparently weighs 170lbs in Nineteenth Century measurements), and John and Andy are both taller than me and similarly proportioned — we have to do a weight and balance check. After going through the agony of trying to remember how to do one without a spreadsheet, it confirms what I suspect: we'd be 100lbs over gross and close to the aft CG limit with Andy along and with full fuel. We could dump fuel, but it seems easier to dump Andy instead :-). So we're left pondering the poor lifting capacity of the 172SP with full 56 gallon tanks — back at the Other Club we kept 2SP (the 172SP there) fueled up to the collars instead of fully-fueled for exactly this reason, as that at least allowed three full-size adults and baggage to depart legally. It seems pathetic that you can't get such a relatively-powerful plane off the ground (legally, at any rate) in such circumstance, but there it is. Something else to ponder if I want to drag this plane back to Santa Monica or up to Portland some day — get the previous pilot to refuel only up to the tabs. I'm not sure what the hell I'd do if I turned up one morning with passengers and discovered I had to drain 20 gallons of fuel to make gross….

After all this we're already late and I'm feeling even more under-prepared, and since it's likely to be IMC for a large part of our flight to and from Stockton (at least I've had the wit to get a decent DUATS briefing for the flight), I tell John we shouldn't bust a gut trying to do the G1000 IFR signoff today — I'd be happy with a quick IFR / IMC flight out to Stockton (KSCK, one of my favourite Central Valley IFR practice destinations) and a couple of approaches — whatever we do will be a good lesson, and I'm resigned to completing the G1000 IFR sign-off sometime later. After all, I already have the VFR version. This sounds like a plan, and we wander out to the apron. John does some of the pre-flight while I file for Stockton (OAK V244 ECA direct, the same whether you depart Oakland or Hayward, a very familiar route).

Once on board we follow the startup checklist and get things started — and, once again, this is a lot more work than with a conventional set of instruments. I know this is probably obvious to most people, but the G1000 is a complex system, not just a disparate collection of mechanical instruments, and unless you start thinking of it as an integrated system, and interacting with it accordingly, it'll never make sense. This time round the workload's somewhat easier, and if you concentrate on the PFD (primary flight display) and the engine gauges on the MFD (multi-function display), ignoring all the extras on the MFD (the maps, the terrain options, the XM radio, the checklists, etc.), it's all fairly comprehensible, if a little complex the first few times. When we're ready to taxi, I call Hayward Deliverance, who gives me a distracted sort of "that's nice, dear" response and tells me to contact Ground without giving me a clearance. We taxi off to the runup area, finally getting the clearance (the usual "runway heading; passing through 400' left turn heading 160; radar vectors for OAK VOR; V244 ECA direct", of which only "left turn 160" and "ECA direct" are likely to be given by NorCal once you're in the air… ) as we get to the runup. Programming a clearance into the G1000 is pretty much identical to the GPS 530 procedure, and nothing about it is surprising or difficult (just tricky at times). We spend a few minutes in the runup area going over the various ways to superimpose the three different NAV sources on the HSI, which for someone raised on separate OBS's (etc), is just magic (this magic turns out to be a significant part of the flight a little later).

I call tower from the runup area, expecting to be told we'll be waiting fifteen minutes for release, but we're told to position and hold on 28L, then almost immediately cleared for takeoff. The departure is pretty standard and I hand-fly until we get into the clouds and NorCal starts giving me the usual vectors and altitude assignments on the Hayward Runaround. Time to let George do the the work. George stays engaged pretty much the rest of the flight until the last few hundred feet at Stockton (with exceptions noted below), which suits me just fine: I like being able to use it as a command instrument, dialing in a new altitude or heading. The only criticism I'd have of the G1000 / KAP 140 combination in this setup is the total lack of altitude coupling between the two (except for the ILS glideslope coupling), which seems a real missed opportunity — it'd be nice to just dial in a new altitude and vertical speed on the G1000 when coupled and have the KAP 140 respond accordingly. Not a big deal, just one of those minor irritations (as someone who spent most of his instrument training in aircraft without GPS, let alone a glass cockpit or an autopilot, I guess I still feel astonished that I have access to any of this technology at all, let alone the ability to complain about some its imperfections…).

We enter IMC about 1000 feet and spend most of the rest of the flight out to Stockton in and out of IMC. The flight out is routine except for two things: at some point after a bunch of vectors we're given a clearance direct to Manteca VOR (ECA) and we set the GPS and autopilot accordingly, but after a few minutes it's obvious the autopilot's developing a mind of its own, and we're increasingly off course. In a classic case of getting distracted by the unimportant, I start trying to debug what the hell went wrong, rather than just turning George off and starting again with a suitable (manual) heading, and using the autopilot in heading mode doing the various corrections with the heading bug until we're back on course and I can set it up properly again. D'Oh! It's the whole (Dis)Grace Under Pressure thing all over again, but with more sophisticated equipment this time. A very useful lesson, needed several more times today as George doesn't always do what I think it will (usually because I don't get the right key sequence). The good news is that (so far) I'm always noticing the problems; the bad news is I try to debug them in real time instead of reverting to something more basic then setting things up again at my leisure.

The only other bit of real interest on the way out was the DME arc John made me do to the final course for the VOR 29R approach into Stockton (in and out of IMC). He'd explained what he'd be asking me to do when we were still in the runup area, so it wasn't a surprise, and NorCal was OK with the request. The surprise was just how much easier it turns out to be to do a DME arc with the RMI-like superimposed display using the HSI with the NAV1 pointer pointing to ECA VOR over the left wing. Sweet! And really a lot more intuitive and easy to visualise than the old OBS-twisting / DME-watching method. I don't think I was more than .2 NM off on the whole arc, a much better result than the last time I did one for real. The following VOR approach is pretty standard, breaking out at about 800 feet (I think), with a quick touch and go and return to NorCal approach on the missed.

We're getting late, so I ask NorCal for a pop-up clearance back to Hayward, which comes a minute or so later, and we head back to Hayward, this time mostly above the clouds for the enroute segments. I get to spend some time playing with the MFD and learning a bit more about how to use it effectively beyond the usual engine instruments and moving map pages (the default versions of which are useful enough in themselves anyway). The LOC/DME 28L approach back into Hayward is routine, breaking out fairly high, and in fact by the time we land the sun's shining over most of the Bay north of about Hayward. We land roughly, really quite badly — last week's landing practice doesn't seem to have had as good an effect as I'd hoped…. Hmmm.

* * *

Once again, a really enjoyable and worthwhile lesson, despite and because of my screwups. No, I still don't have the G1000 IFR signoff, but it's all coming together well, and the next time I fly with John we basically don't have to run through much more than a decent circling approach and a few holds. Cool! And, once again, I find myself actually enjoying even the bits I flew terribly — always a good sign that you're learning. And I learned a lot today, especially about keeping cool under pressure and learning to fall back to something simpler without letting things get out of hand. Yes, I already knew a lot of this, but it's something that needs to be hammered home in a realistic environment (e.g. IMC in a new plane and using new instruments) to be really effective.


Sam Weigel said...

Good point about reverting to a more basic mode whenever things start to go wrong. It's a basic tenet of the automation policy at my airline and many others. Automation is wonderful when everything is going well, but there's a point at which you need to turn it off and revert to the familiar until you get things figured out.

Hamish said...

I've noticed something weird about the reversion thing -- when I'm flying on my own I'm much more robust about things like that (I have a healthy distrust of the AP even though I really like using it), but when there's another pilot or instructor in the right seat, I tend to get distracted by asking them for help. On my own in the situation I had this flight I would probably have just immediately disengaged the autopilot and hand-flown the thing for a while when I noticed the course deviation (an easy option in the enroute segments), then after a few minutes carefully set the AP up again and engaged it. In fact, that's exactly what's happened several times in the past few months. With John around, though, I guess I just used him as a crutch and forgot the basics. Urgh...