The Captain on my flight to DFW last night informed me that he’d just finished a drug screening, and his first attempt to produce a sample of sufficient quantity had failed, and he’d had to sit around drinking water until he could produce another. As a result, he was well hydrated and was likely to need the lavatory on our flight to Dallas. He was a former Dash 8 pilot, and was brand new to the jet, fresh off of IOE.

On the taxi, we got excited because it looked like we were going to be #4 or so for takeoff. Which is when ground announced that they were switching runways, and we would have to taxi all the way around the airport. JFK has two taxiways around the perimeter of the airport, one runs clockwise and the other counterclockwise. They make a complete circle. We started at about 9 o’clock position, and taxied to the 7 o’clock position… clockwise. Because of the long taxi we shut down the APU and taxied single-engine in order to save fuel.

When it came time to start the engine, the captain called for the “cleared to push/start below the line number one,” which I dutifully ran.

Me: Parking Brake
Captain: Off
Me: Fuel Pumps both on. Engine start, spinning one.

I started the clock for the engine start, selected the #2 engine start switch. Some pneumatic valves opened and closed and the engine core started spinning. I waited for it to get to 20% RPM, verified the temperature was below 120ºC, and moved the left thrust lever up to idle. The engine fired off and the starter cut out. The temperature kept rising though, and got over 700ºC before settling back down. I was afraid I was going to have to abort the start, but fortunately it stayed just shy of the limits. I remarked to the captain that was the hottest start I’d ever seen, and reflected that we must have more of a tailwind than it looked like. He just said, “huh,” not being all that familiar with the plane and not to mention that the captains almost never start the engines.

It wasn’t until today that it suddenly hit me what happened. To start the engines, we need a source of pressurized air to get them spinning. This can come from the APU (auxillary power unit), an external air compressor connected to the plane, the other engine, or if we’re in the air, just the free air stream for the movement of the aircraft. 99% of the time, we start the engines from the APU. But in this case, we’d shut down the APU which meant a crossbleed start using the compressed air from the other engine. My plane is very smart, and when you push the start button it figures out automatically which valves need to be open and closed to start the engine based on where the pressurized air is coming from. The only catch with a crossbleed start is that the engines need 42psi to start, and at idle they only produce about 32 psi. In short, this means that you need to advance the thrust lever on the running engine before pushing the start button on a crossbleed start. Neither I nor the Captain remembered to do this. Me, because I’m so used to starting from the APU and the Captain, because he was so new. So, now I know you can do a crossbleed start with one engine at idle in the CRJ-900 but it’s pretty darned likely that you’ll get a hot start. It’s a good thing we didn’t have to abort the start, because then we would have had to explain why we’re stupid.

The taxi was long, and the Captain really had to pee. After awhile, he gave up and ran to the lavatory while we were stopped. Of course the moment he left the flight deck, ground control wanted the plane to be moved. Since the steering wheel is on his side, I had to tell them that we needed a few minutes.

We eventually managed to get airborne. Right after takeoff I noticed a very loud whining noise. As we got faster and higher, the noise increased in pitch. After getting the gear up and flaps up, the aircraft configured for climb, and then pointing in the right direction I remarked to the captain that the noise was not normal, and I didn’t like it. He agreed that it seemed strange, and suggested I call maintenance. I checked with the flight attendants to see if they could hear anything. They couldn’t, which didn’t surprise me as it seemed to be coming from in front of us. I radioed ARINC, the company that provides air to ground telephone patches for us and asked to be connected to our dispatcher. A short time later, I was connected to the dispatcher who sounded like he was talking to me from the bottom of a well 100 miles away with a kid whistling in the background. I shudder to think what my employer probably pays for such a link. I explained our situation and asked to be patched through to maintenance control. He informed me that he had no way for me to be connected to maintenance. I’m not sure why not; the telephone patch can connect us to any telephone number anywhere. The dispatcher relayed our problem to maintenance and said he’d get back to us in a minute.

About then, I’d decided that the noise is likely a small pressurization leak. It sounds just like a leaky door seal that was all too common while I was a flight attendant. Such leaks generally increase in pitch as you climb and for whatever reason seal themselves and become quiet after the aircraft has reached its normal pressure differential. I took the book of approach plates that was sitting on top of my oxygen masks and put it in my bag in case I needed the mask. I figured that since the noise wasn’t getting worse, and was actually getting quieter it probably wasn’t a big deal.

The dispatcher came back and said that maintenance thinks we probably have an open com panel on the outside of the aircraft. There’s a little hatch that allows the ground crew to plug up a headset and talk to us. Unfortunately, the ground crew in JFK is too lazy to plug up a headset and so we have to rely on hand signals instead. So it didn’t make sense that they’d leave that door open since there was no reason for them to open it in the first place. I’ve heard of that door being left open though and supposedly you hear it banging around, and I didn’t hear it banging at all.

Dispatch asked if we wanted to divert or not. Since that’s a Captain decision, I went ahead and took over the flight controls and the job of talking to ATC while the captain talked to dispatch. They eventually decided to divert to Dulles, but this involved a lengthy discussion. The big issue is that we were overweight for landing, and it seemed to be prudent to burn off fuel until we were down to landing weight before landing in Dulles. Because DFW is our longest flight, we’d taken off with nearly full fuel. Anyway the back and forth between the Captain and dispatcher took a good long while, probably more than 20 minutes. During that time I had to fly single pilot which meant things were busy for me, especially since I had to enter a hold and explain our situation to ATC all by myself. I got us in a hold, and after it turned out we were going to divert to Dulles I got a new clearance for Dulles. My captain programmed the FMS while I flew, and fortunately I caught him entering DCA instead of IAD as our new destination into the computer. That would have been bad.

I was flying with the autopilot in heading mode which just means that it will point the plane in whichever direction I tell it to. I had it pointing to the first fix on our new route. Unfortunately I forgot that it wasn’t in navigation mode, which is an easy thing to do, because I’d pointed it in the right direction so it’s not obvious that the plane doesn’t know it’s not going in the right direction. I noticed it after we got about 10 miles past our fix and the plane hadn’t turned to follow the route like it was supposed to. I got it turning in the right direction about the time ATC asked me what the heck was up. Since she knew we were busy she mostly seemed amused rather than upset which is good. They asked us repeatedly if we were declaring an emergency and we reiterated that we were not. In the whole thing, I was very impressed with how much ATC worked with us and how helpful they were, even without us declaring an emergency.

We eventually got the plane back on the ground in DC. Mechanics met us at the gate, and sure enough there was a panel open, but not the one we thought it was. It was for the connection for external AC power, and the ramp crew in JFK had failed to close the door before the aircraft departed. Such a simple mistake cost tends of thousands of dollars. And it could have been worse: had the door come detached from the aircraft and been ingested by an engine (not a terribly unlikely scenario), then we could have had a serious emergency. While everyone makes mistakes (especially me!) it drives me crazy that there are so many people working on the ramp at airports around the country that don’t seem to take their jobs at all seriously.