As I said, it was a freighter trip. I started flying 747s as a freight dog, and finishing that way was okay. Flying freighters is more fun, the actual flying part, than passenger work because you can do things that would be inappropriate with passengers aboard. Things like steep banks when close to the ground, keeping your speed until close in and then killing it by dropping the gear (helluva noise from the gear at high speed, scares passengers sometimes), descents too steep for the pressurization to keep up, banking and pitching quickly (again, scares the passengers), steep deck angles. That kind of stuff.

Anyway, they gave me the flight plan and weather at Sal Island for the trip to Curitiba. The weather for our arrival was forecast to be good, so I loaded minimum fuel. That's standard. It costs fuel to carry fuel. If you load extra fuel over the minimum, it will cost 20% of that extra fuel just to carry it. In other words, if you want 8,000 extra, you have to load 10,000 extra. If you regularly load extra fuel and then don't use it, the Chief Pilot will call you into the office for a little chat. I was always pretty good at keeping fuel costs down.

However, minimum fuel for Curitiba turned out to be a bad idea. Unforecast fog closed the field. ATC notified us when we were about thirty-five minutes out. To further complicate things, there was only a one degree temperature/dew point spread at our alternate, Campinas, which we had just passed over.

What to do? Minimum fuel meant we had enough to get to Curitiba and then to Campinas, and enough holding fuel to hold for half an hour at either of them. Of course, the big question was, when would the fog in Curitiba lift? The typical solution would have been to hold over Curitiba for as long as we would and then proceed to Campinas. However, with only a one degree temperature/dew point spread at Campinas, it might fog in. My way of handling these things in flight is to verbally go over the options and then solicit opinions from the other crew members. In this case both the f.o. and the f.e. indicated a slight preference for proceeding straight to Campinas. However, if we did that, we would be landing at Campinas with quite a bit of fuel. Somebody was bound to ask, with that much fuel, why we didn't try to get in. So, I elected to head for Curitiba, and, if on arrival overhead we didn't have minimums, proceed immediately to Campinas. I didn't want to waste our holding fuel at Curitiba when we might need it at Campinas. Blowing all our holding fuel and then arriving over a fogged-in Campinas would not have been politically correct. Had that happened, we would have had to declare a fuel emergency and land in spite of the field being closed—lots of paperwork.

I made the decision and then told the flight engineer to contact company on HF. He kept trying but never did succeed. We found out later that Tower dispatch, seeing the weather deteriorating at Curitiba, had been trying to contact us and never succeeded. Not surprising, HF communication in that part of the world can be difficult, and our number one HF was very weak, our number two inop—deferred.

The rules on shooting an approach at Curitiba are unusual because it is surrounded by high terrain, the highest a 9,000 foot mountain to the right of the final approach course for the ILS there. Field elevation is 3000 feet. You can't even begin an approach unless you have the required weather minimums, and Curitiba's minimums for Tower are 1200 meters (yes, metric everywhere but in the U.S.) and a 400 foot ceiling (ceilings are still in feet in all countries except the former Soviet Union and a few others). When ATC first notified us that Curitiba was below minimums, the ceiling was 200 feet and visibility 600 meters. Shortly after we started our descent, they told us the ceiling had dropped to 100 feet, visibility to 75 meters. I immediately headed for Campinas. That would put us into Campinas at over 45,000 pounds of fuel (desired normal is 30,000) if we got in there without having to hold, but my judgement was, given the further deterioration at Curitiba, that it wasn't going to come up any time soon, and that going there would be a waste of fuel. That turned out to be a good decision. Curitiba didn't come up until over three hours after we had landed at Campinas.

When we got to Campinas, the weather was still VFR with the sky clear, but ground fog was forming fast. Other aircraft were also inbound, and they threw us into an NDB hold. NDB holds are rarely ever used in the U.S. I couldn't remember the last time I flew an NDB hold. By this time I was hand flying, and the mental arithmetic for other than an NDB hold can be formidable, that for an NDB hold even more so, and the f.o. picked that time to get confused and gave me bad holding entry instructions. I caught the error and we got lined up for the hold. I also told the f.o. to request a visual approach. If ATC granted it, they could work us in between arriving IFR aircraft, and they did both. Approach Control told us to fly a right downwind for runway 09. We terminated the hold entry and headed for the right downwind. Unfortunately, given a right downwind, the left seat pilot loses sight of the airport and has to depend on the right seat pilot for downwind leg distance from the runway and the when to turn base. In the increasing haze, the f.o. lost sight of the runway (he was having a bad day), guessed at both items, and we turned to the final course without having sight of the airport or runway. Technically speaking, when that happened, we should have called the tower, cancelled our visual approach, and gotten back in line for an instrument approach. We didn't, and continued, counting on being able to reacquire the airport and runway before it was too late.

And we did reacquire the airport and runway, but we were very close in and quite high, and the runway was well off to the right. No problem. What followed wouldn't have been pleasant for passengers, but boxes don't complain. Flaps to 30 degrees, normal is 25, and a 45 degree right bank (30 degrees is considered the normal max for large aircraft).

Whenever it appears you will not be able to touch down within the touch down zone (first 3000 feet of the runway), a go-around is mandatory, but that rule is often broken. I was prepared to break the rule, but it wasn't necessary. I touched down, smoothly but not a greaser, around 2500 feet.

Three and one half hours later, we took off for Curitiba. It's only a 45 minute flight, and it would be my last. I decided to hand fly all the way. I was already very tired having been up about twenty-eight hours by then, and hand flying at altitude really soaks up your energy. The airplane is very sensitive. You have to be right on top of it every moment. To make it more interesting, I killed the altitude hold function of the flight director and slewed its pitch bar out of the way. I would take care of the altitude myself, and doing that is almost unheard of in 747s. It's just too much work, and the altitude excursions can be very embarrassing. However, I managed to keep all excursions to less than one hundred feet, but I was really trying. I didn't care how badly I wore myself out. It was my last flight and I was doing it my way.

Curitiba has a very short runway for a 747, only 7200 feet. That short a runway demands that the pilot's first priority is to “get it on.” Normally, you're willing to give up a little runway trying for a good landing. On short runways you have to get it on as soon as possible so the wheels can spin up so the autobrakes can start slowing you, and you have to get the reversers out as soon as possible as well. I did all that and got a greaser to boot. I was a very happy man, and both the f.o. and the f.e. were dutifully impressed and said so. I had done my last leg right.