As Usual, Nothing Goes Horribly Wrong...

A short lesson in IFR GPS usage with the Garmin GPS 530 in 4JG over San Pablo Bay: GPS terminal and enroute procedures, holds, the RNAV (GPS) RWY 6 approach with a circle-to-land on runway 18R at Napa (KAPC) and a shot at the missed, and the GPS RWY 27R approach back into Oakland for a change. As usual, nothing goes horribly wrong, but I'm not real pleased about my ability to cope with ad hoc holds (John got me to hold at REBAS using the GPS, with only a minute or so's notice, which I flew poorly) and other precision airwork. I've spent a few hours over the last week playing with Garmin's excellent GPS 530 simulator on my laptop, which was great preparation for the approaches themselves, so at least I'm mostly ahead of the plane and instruments for that part of the lesson. Pity about the flying, though.

* * *

I've been a fan of GPS for flying, and for things like backpacking, sailing, cross-country skiing, cycling, etc., for years (I have at least 5 handheld GPS units somewhere around here), and this lesson helps reinforce a few prejudices. GPS as a concept, and GPS enroute and approach procedures and the associated instruments and displays (at least as experienced with the 530), just strike me as natural. There's nothing conceptually or procedurally difficult about using GPS for IFR, and there are huge advantages -- the help with things like missed approach procedures or holds, the heading advice, the moving map for situational awareness, etc. These things make IFR flying a lot less hit-or-miss, and a lot more forgiving (which some would probably say makes GPS a Bad Thing, but never mind). I think I'd agree with John: if you own a small GA plane and you fly it single-pilot in IMC, it's difficult to justify not having something like this (well, he puts it more forcefully). And the flip side of all this is that GPS approaches can be designed and specified for a whole bunch of smaller airports that wouldn't otherwise have any sort of approach.

But the way most IFR GPS interfaces are currently implemented is a human factors mess, and the 530 is no exception, at least for me. I'm an experienced computer (and GPS) user and programmer, and I have trouble navigating my way around the menus and screens. There's just no affordance in the design -- there's no simple mapping from the world out there or your mental model of it or IFR flying to the 530's interface, as there arguably is for simple VOR navigation instruments, for example. And the complexity of choices on offer at any point in a sequence of entries can be overwhelming -- there doesn't seem to be any way to intelligently minimise the choices available based on locality, likely use, phase of flight, etc. Part of the problem is obviously just the sheer complexity of what the 530 can do (so much more than a simple OBS or HSI), but part of it also just poor interface design. A classic example is the need to hit a separate "entry" button after entering a waypoint name (or whatever) with the cursor knob; it's more natural to just press the same knob in (as I kept doing), but that action brings up a new menu and erases what you've just entered (leading to a whole string of "D'Oh!"s 3,000' over the Bay...). As John says, this is a unit that just cries out for an "undo" or "cancel" or "back" key. And things only get worse if you have to change your clearance mid-air -- navigating a series of screens, buttons, and menus to amend a stored flightplan or cope with vectors to an approach can become a serious distraction. In my case it took a minute or so to enter some new information, during which time I lost heading and altitude, both quite seriously. It's obviously an acquired skill to keep the plane steady without a copilot or autopilot while doing all the bit-twiddling; it's also obvious that it'll take me a fair while to learn that skill. It's obviously worth it....

A less serious problem is that the beautiful moving map display -- especially with the superimposed CDI and heading indicator -- is so damn seductive, so eye-catching, that it's way too easy to watch it instead of the coupled HSI or OBS. But the 530's display has a noticeable lag to it, and it's just not as simple to read and react to as the HSI; plus it keeps you away from your normal instrument scan, which could have serious consequences if (say) you missed the altitude problem you're developing as you're mesmerised by the GPS. The trick here is probably to do what John recommends -- put the unit display onto the (boring) flightplan page and concentrate entirely on the coupled HSI, glancing over every now and then to check progress, read any advisories, etc. from the 530's screen. This may take some getting used to...

But overall, it's hard to imagine doing serious IFR work in the future without something like the 530. Just watching it sequence through the approach waypoints while telling me the upcoming desired track and when to turn (based on its groundspeed calculations), or the way you just hit OBS on the missed and it automatically points you to the next waypoint in the missed procedure, or having it tell you the hold entry procedure on the upcoming hold -- well, all that's just magic. When it's retrofitted with WAAS, the whole precision GPS approach thing suddenly starts looking real.

* * *

4JG's a 172 with a 180 HP engine (as opposed to the more typical 150 HP), and the difference makes for extreme climb performance and great sustained cruise speeds (for a 172, at least). It also proves to be very stable in pitch, airspeed, and heading at slower speeds (below about 100KIAS), which is great in the approaches and when going missed. But I have trouble with heading and altitude stability much above that, and the entire leg home from the Napa missed approach hold to vectors for the Oakland approach I keep hunting around the heading in ways I don't seem to have to do in the other planes. I always seem to be a few degrees of heading, scalloping slowly along the course at 125 KIAS. Oh well -- I was never more than a dot or two out, but it would drive the average passenger crazy after a while, I suspect.