Narrative:

We were near the end of completing rvsm trailing cone engineering flight testing/data collection at 5000 ft, and the low level fuel annunciators began to flicker. The boost pumps were turned on per the abnormal procedures. We completed the last condition. Initially we thought we had an annunciator problem with approximately 1600 pounds of fuel showing on the gauges, and because of the excellent reliability I have experienced on citation jets fuel quantity system. Also, the flight test engineer discussed having this discrepancy on the previous 2 flts. I was on vacation for 2 weeks prior to this flight. Takeoff fuel was indicating 2100 pounds. Then the annunciators came on steady (+/-185 15 pounds, per design). I did not like the condition of the aircraft, and decided to stop testing and return to ict (35 NM northwest of ict). Wichita approach control assigned heading 100 degrees and 4500 ft, vector for sequencing to runway 19L. En route to ict, the left engine flamed out. Wichita approach control was advised. I was indicating approximately 195 KTS. I was preparing myself and the aircraft for the right engine to flame out, and it did approximately 30 seconds after the left. An emergency was declared. A speed of approximately 180 KTS was used to glide. The K-96 highway appeared to be the best place to land the aircraft considering wind and the fact that the fields had been soaked with heavy rains for the past 2 days. I aligned the aircraft with the roadway and tried to figure a way to fit into the flow of traffic. I lowered the landing gear, and it came down and locked, with the windmilling engines driving the hydraulic pumps. Zero flaps was utilized to try and touch down beyond a set of power lines, and before another set of power lines. As the landing point was drawing near, I realized that I was going to land on a vehicle. This was the only traffic that I had concern for. The roadway ahead of this vehicle was clear. I flew the aircraft over the median to avoid the vehicle and figuring that the operator would see me and slow or stop. I reasoned, once I passed the vehicle, she would stop. Once past that vehicle, I maneuvered the aircraft to align with the roadway. I touched down at stick shaker, with a lateral acceleration to the right. I got the aircraft lateral acceleration stopped, but I was off the roadway to the right side. I maneuvered the aircraft back up and left onto the roadway. There was a culvert and a bridge approaching very quickly, directly ahead of the aircraft, while I was off the roadway. I struck three side road delineator markers (reflectors) during the time I was off the roadway. I cleared the obstructions on the culvert/bridge area, centered on the roadway, and slowly brought the aircraft to a stop. To avoid being hit from behind, I slowly brought the aircraft to a stop under normal braking. From the initial touchdown to the stop was approximately 1 mi. I inadvertently got the right main landing gear off the roadway, trying to keep from blocking the entire sebound lanes of K-96. The fuel quantity indicated about 500-550 pounds in each fuel tank, 1000-1100 pounds total fuel at rest on the highway. Minimal damage was sustained to the aircraft nose radome and right wing leading edge, from the road delineators. A small amount of dirt/mud was thrown into the right engine inlet. The fuel gauging system was inaccurate. The system was an experimental fuel gauging system, without the means to annunciate system diagnostic faults. A production aircraft would have annunciated the faults. 3 fuel quantity probes were found to be faulted, post incident investigation. I was not briefed of the fuel anomaly, nor was the conflicting annunciations written up by the previous crews for 2 flts. I would be more than happy to discuss this, for the purposes of enhancing aircraft safety. One conclusion is that you must trust the worst indication/annunciator, when conflicting information is being presented in the cockpit.

Related reports:

Google
 

Original NASA ASRS Text

Title: FLC OF CITATION 525 ON A TEST FLT HAS LOW LEVEL FUEL ANNUNCIATORS FLICKER. HEADING BACK TO THE ARPT, THE R ENG FLAMES OUT, THEN THE L. OFF ARPT LNDG IS MADE ON A HWY.

Narrative: WE WERE NEAR THE END OF COMPLETING RVSM TRAILING CONE ENGINEERING FLT TESTING/DATA COLLECTION AT 5000 FT, AND THE LOW LEVEL FUEL ANNUNCIATORS BEGAN TO FLICKER. THE BOOST PUMPS WERE TURNED ON PER THE ABNORMAL PROCS. WE COMPLETED THE LAST CONDITION. INITIALLY WE THOUGHT WE HAD AN ANNUNCIATOR PROB WITH APPROX 1600 LBS OF FUEL SHOWING ON THE GAUGES, AND BECAUSE OF THE EXCELLENT RELIABILITY I HAVE EXPERIENCED ON CITATION JETS FUEL QUANTITY SYS. ALSO, THE FLT TEST ENGINEER DISCUSSED HAVING THIS DISCREPANCY ON THE PREVIOUS 2 FLTS. I WAS ON VACATION FOR 2 WKS PRIOR TO THIS FLT. TKOF FUEL WAS INDICATING 2100 LBS. THEN THE ANNUNCIATORS CAME ON STEADY (+/-185 15 LBS, PER DESIGN). I DID NOT LIKE THE CONDITION OF THE ACFT, AND DECIDED TO STOP TESTING AND RETURN TO ICT (35 NM NW OF ICT). WICHITA APCH CTL ASSIGNED HDG 100 DEGS AND 4500 FT, VECTOR FOR SEQUENCING TO RWY 19L. ENRTE TO ICT, THE L ENG FLAMED OUT. WICHITA APCH CTL WAS ADVISED. I WAS INDICATING APPROX 195 KTS. I WAS PREPARING MYSELF AND THE ACFT FOR THE R ENG TO FLAME OUT, AND IT DID APPROX 30 SECONDS AFTER THE L. AN EMER WAS DECLARED. A SPD OF APPROX 180 KTS WAS USED TO GLIDE. THE K-96 HWY APPEARED TO BE THE BEST PLACE TO LAND THE ACFT CONSIDERING WIND AND THE FACT THAT THE FIELDS HAD BEEN SOAKED WITH HVY RAINS FOR THE PAST 2 DAYS. I ALIGNED THE ACFT WITH THE ROADWAY AND TRIED TO FIGURE A WAY TO FIT INTO THE FLOW OF TFC. I LOWERED THE LNDG GEAR, AND IT CAME DOWN AND LOCKED, WITH THE WINDMILLING ENGS DRIVING THE HYD PUMPS. ZERO FLAPS WAS UTILIZED TO TRY AND TOUCH DOWN BEYOND A SET OF PWR LINES, AND BEFORE ANOTHER SET OF PWR LINES. AS THE LNDG POINT WAS DRAWING NEAR, I REALIZED THAT I WAS GOING TO LAND ON A VEHICLE. THIS WAS THE ONLY TFC THAT I HAD CONCERN FOR. THE ROADWAY AHEAD OF THIS VEHICLE WAS CLR. I FLEW THE ACFT OVER THE MEDIAN TO AVOID THE VEHICLE AND FIGURING THAT THE OPERATOR WOULD SEE ME AND SLOW OR STOP. I REASONED, ONCE I PASSED THE VEHICLE, SHE WOULD STOP. ONCE PAST THAT VEHICLE, I MANEUVERED THE ACFT TO ALIGN WITH THE ROADWAY. I TOUCHED DOWN AT STICK SHAKER, WITH A LATERAL ACCELERATION TO THE R. I GOT THE ACFT LATERAL ACCELERATION STOPPED, BUT I WAS OFF THE ROADWAY TO THE R SIDE. I MANEUVERED THE ACFT BACK UP AND L ONTO THE ROADWAY. THERE WAS A CULVERT AND A BRIDGE APCHING VERY QUICKLY, DIRECTLY AHEAD OF THE ACFT, WHILE I WAS OFF THE ROADWAY. I STRUCK THREE SIDE ROAD DELINEATOR MARKERS (REFLECTORS) DURING THE TIME I WAS OFF THE ROADWAY. I CLRED THE OBSTRUCTIONS ON THE CULVERT/BRIDGE AREA, CTRED ON THE ROADWAY, AND SLOWLY BROUGHT THE ACFT TO A STOP. TO AVOID BEING HIT FROM BEHIND, I SLOWLY BROUGHT THE ACFT TO A STOP UNDER NORMAL BRAKING. FROM THE INITIAL TOUCHDOWN TO THE STOP WAS APPROX 1 MI. I INADVERTENTLY GOT THE R MAIN LNDG GEAR OFF THE ROADWAY, TRYING TO KEEP FROM BLOCKING THE ENTIRE SEBOUND LANES OF K-96. THE FUEL QUANTITY INDICATED ABOUT 500-550 LBS IN EACH FUEL TANK, 1000-1100 LBS TOTAL FUEL AT REST ON THE HWY. MINIMAL DAMAGE WAS SUSTAINED TO THE ACFT NOSE RADOME AND R WING LEADING EDGE, FROM THE ROAD DELINEATORS. A SMALL AMOUNT OF DIRT/MUD WAS THROWN INTO THE R ENG INLET. THE FUEL GAUGING SYS WAS INACCURATE. THE SYS WAS AN EXPERIMENTAL FUEL GAUGING SYS, WITHOUT THE MEANS TO ANNUNCIATE SYS DIAGNOSTIC FAULTS. A PRODUCTION ACFT WOULD HAVE ANNUNCIATED THE FAULTS. 3 FUEL QUANTITY PROBES WERE FOUND TO BE FAULTED, POST INCIDENT INVESTIGATION. I WAS NOT BRIEFED OF THE FUEL ANOMALY, NOR WAS THE CONFLICTING ANNUNCIATIONS WRITTEN UP BY THE PREVIOUS CREWS FOR 2 FLTS. I WOULD BE MORE THAN HAPPY TO DISCUSS THIS, FOR THE PURPOSES OF ENHANCING ACFT SAFETY. ONE CONCLUSION IS THAT YOU MUST TRUST THE WORST INDICATION/ANNUNCIATOR, WHEN CONFLICTING INFO IS BEING PRESENTED IN THE COCKPIT.

Data retrieved from NASA's ASRS site as of July 2007 and automatically converted to unabbreviated mixed upper/lowercase text. This report is for informational purposes with no guarantee of accuracy. See NASA's ASRS site for official report.