Well, here's the story:
First the Moral: Be very careful on the first flight after extensive maintenance has been performed.
Info: Most jets have excellent fuel quantity indication systems. It is common that most pilots of that type equipment use the aircraft indicating system for fuel planning purposes and for refueling purposes. It is not the same as our light-planes that have "toy" gauges that can never be trusted.
Of course, Trust.... is developed as the consequence of accurate, repetitive, reproduceable, never-erring results... also called "operating experience." I thought I'd accurately and correctly determined how much fuel was on this flight. I was wrong.
The story:
I was chief pilot of a group of banks that had two Hawkers (HS-125) and one of them was due an inspection that called for it's fuel-quantity amplifiers to be overhauled. (These were black-boxes that took info from the fuel sending units in the wing and communicated that info to the fuel quantity gauges in the cockpit.)
Since we had a trip coming up in two days, I asked the mx facility to get some "loaner" or "rental" units as temporary replacements, to be used until our own freshly-overhauled units returned from the overhaul facility. That was done and I was notified my airplane was ready to be picked up. The mx facility was in Dallas, as was the departure point for my trip to Phoenix Sky Harbor Airport (PHX). The flight would require 5,000 lbs (2 hrs) of fuel and we wanted 2,000 lbs (45 mins) of reserve fuel, so the total fuel load was to be 7,000 lbs. We had left that fuel order with the mx facility, and when we arrived to pick up the airplane for the trip the fuel gauges displayed 3500 lbs per side.
I asked the mx facility if the fuel gauges had been calibrated to the rental/loaner amplifiers and the shop foreman said "Well, airframe-shop installed the amplifiers but avionics-shop calibrated them, according to the work order. But let's go talk to the avionics technician who did the actual calibration work." I said, "Great!" and so we did.
The avionics technician replied that yes, he'd calibrated the gauges. I asked how he accomplished that task, and he replied that he'd noticed when the rental/loaner amplifiers had arrived that they read excessively high when the tanks were actually empty, so ...(the fuel tanks had been emptied during the work order, and since they were dry tanks it was a perfect opportunity to determine exactly how much fuel would be in the tanks by this excersize).... what the technician did was to adjust the amplifiers such that the fuel gauges read "zero" with empty tanks.
He then stated that he'd added 15 gallons of jet fuel to each tank and noted the cockpit gauges rose to read 100 lbs. He added another 15 gallons and noted the gauges read 200 lbs. And so on, and on, etc etc. (This is a commonly accepted method of calculating jet fuel. 15 gals equals 100 lbs. 150 gals equals 1000 lbs. 1500 gals equals 10,000 lbs, etc etc.) Therefore when the total fuel load in each wing had come up to 525 gallons each, the gauges had displayed 3500 lbs....and therefore the gauges and amplifiers were determined to be exactly correct! Both wings together therefore were determined to hold the 7,000 lbs of fuel we needed for our trip.
"Good job! Thanks!", I replied. And we taxiied across the airport and picked up our passenger and departed for PHX.
Two hours later we were over a clear-blue PHX on right downwind to runway 8.
It was a beautiful scene. There we were pointed West, on high, right downwind with the airport under our right wing and we had a Douglas DC-9 in front of us on right base in sight. We were to follow him on a visual to the airport, when the final controller asked us if we could take a short, ten-minute-hold for an incoming emergency aircraft. We looked at our fuel gauges and noted we had 1000 lbs in each wing...almost an hour's worth of fuel in VFR weather. Ten minutes holding would be no problem at all!
Both the DC-9 and ourselves agreed to take the VFR hold, and we both simply orbited while a Mooney came in with a false warning indication and landed safely.
The DC-9 was then cleared for visual approach and we were cleared to follow the DC-9. Our fuel gauges at that moment indicated 750 lbs of fuel each wing (still about 45 mins at low power settings.)
The DC-9 turned final and we turned base leg. The DC-9 landed and we turned onto a 4 mile final. The DC-9 missed his high-speed turn-off and we were on 2 mile final and our right fuel-low-pressure warning suddenly illuminated. "Hey, Look at this, Jerry. We just lost a boost pump", I said as I simultaneously opened up the fuel cross-feed valve. Opening the cross-feed valve allowed the still-operating left boost pump to supply fuel pressure to both engines and extinguish the low-pressure warning. We were over the approach lights and life was good, if that DC-9 will get off the runway ahead of us before we cross the threshold.
"Be prepared to go-around", said the tower....but just at that moment the DC-9 was exiting the runway at the next high-speed exit so the tower cancelled the warning and cleared us to land, just as we were crossing the runway numbers.
We touched down and Jerry started the Auxilliary Power Unit (APU), which is a small jet-engine in the fuselage that supplies air conditioning and electrical power completely seperate from the main engines. The APU on a Hawker gets its fuel from the left main wing tank. It burns about 10-12 gph, while the engines
idle at 600 lbs/hr (90 gph).
As per our SOP's (Standard Operating Procedures), we shut down the right engine after the APU was up-and-running, and we taxiied in on one engine. We came to the chocks, shut down the left engine, and the pax got off and into a waiting limo and departed for their business meeting. The fuel truck pulled up in anticipation of our placing a fuel order for the return trip. Our fuel gauges still indicated 700 lbs (105 gals) each side for a total of 1400 lbs, so I intended to order about 5,600 lbs (420 gallons per side) for a total return fuel load of 7,000 lbs.
I gave Jerry the gear-locking pin for the right main landing gear, and I walked to nose wheel to install the nose gear pin while he installed the right gear pin. He then met me at the left main landing gear, and while I installed the left gear pin he remarked, "It sure is hot and dry here in Phoenix! That little 10 minute hold at 10,000 feet completely evaporated the condensation under the wings!" (Normally after a flight at high altitude a jet lands and has either a thin coat of ice beneath the wings, or at least a wet lower wing-skin due to condensation from the chilled fuel inside the wing.)
"Yeah, ---These wings don't have any condensation at all!" I responded as I considered the hot, dry, desert air that must have dried the wings during the 10 minute hold we'd accepted on downwind.
As I walked up to the lineman to give him our fuel order for the return trip, .... the APU flamed out and shut down!
"Whoa! What's this!", I thought. Those fresh-overhauled and calibrated cockpit fuel gauges still indicated 700 lbs (more than 100 gallons) of fuel per side!
We decided to have the wings completely topped off with fuel in order to check this situation out. If the wings indeed had 700 lbs of fuel in them, then they should have only taken about 3500 lbs (525 gallons) each. (They only hold 624, and only 612 of that is actually useable.)
But they put in 618 gals in the left and 622 gals in the right wing. That low fuel pressure indication on short final hadn't been a failed boost pump. It had been due to a completely dry and empty fuel tank! We had landed completely out of fuel! When we'd agreed to take a 10 minute hold for someone else's precautionary...we were unknowingly in a state of fuel emergency ourselves!
How did this chain of events happen?
When we investigated, it turned out that the mx facility in Dallas worked three shifts. The day shift removed our fuel amplifiers and sent them for overhaul and installed the loaner/rental amplifiers. The day-shift avionics shop then calibrated those amplifiers, and then went home. The mx supervisor then notified us that our airplane was ready, and we made airline arrangements to arrive the next morning to pick up our airplane and make the trip to PHX.
But in the meantime, the late night express-freight arrived with our original fuel qty amplifiers. The overhaul shop in Florida had gotten them overhauled super-quick (probably nothing more than a bench-check) and express-freighted them back to the mx facility. The midnight shift received those fresh-certified units, removed the calibrated rental/loaner units, and re-installed my original amplifiers. (Which, due to the standards of the overhaul shop were calibrated exactly like the rental units..... they read too high (about 700 lbs) when the tanks were actually empty.)
When I arrived to pick up the airplane, the airframe shop and the avionics technician had no idea the midnight shift had removed the calbrated units and re-installed uncalibrated units.
How was it possible that the fuel load used to calibrate the rental/loaner units wasn't the fuel-load still onboard when I arrived? Because by pure chance, after fueling the aircraft, the shop noticed that some newly installed O-rings were leaking at the inspection covers in the bottom of the wings. So the shop DE-fueled the airplane to replace those O-rings, while the other guys installed my original, un-calibrated fuel amplifiers.
The lineman was then called and was told the pilots wanted 3500 lbs of fuel in each wing. So, not knowing a thing about what had transpired, the lineman used the (un-calibrated) cockpit gauges to pump fuel into the wings...until those gauges indicated 3500 lbs each.
Two hours later I and my crewmember arrived to pick up the airplane.
If that DC-9 had missed that next high-speed exit then I would have had to make a go-around, and both engines would have flamed out about the time that we had reached the departure end of Rwy 8 at PHX.
I'm certain the NTSB report would have read, "Pilot Error - Lack of proper fuel planning." because the aircraft logbooks would have testified that I'd had new, overhauled, freshly-calibrated fuel quantity indicators.
If the lineman had been a bit faster pumping fuel into the wings so that APU had not shut down due to lack of fuel, we might have only ordered 840 gallons total and we'd have possibly run out of fuel either on the way back to Dallas or on a subsequent trip.
Watch out when you get your airplane out of maintenance. And keep track of how much fuel you should burn each flight, compare it to what is loaded onto the airplane during refueling, and BE THERE when they do it!
PS-I'll bet every one of us has, at one time or another, trusted some lineman to refuel our airplanes in our absence. No matter how many times we've allowed that practice and gotten away with it.... it's probably only a matter of time before we are caught with a mis-fueled aircraft... a situation that more likely will be avoided if we properly supervise the activity like we've been trained.
