Narrative:

I was conducting an IFR cross country training flight on an IFR flight plan. The aircraft had an IFR GPS; a #1 navigation/communication of a KX155 with GS interfaced to a HSI; and a #2 navigation; a narco VOR/localizer/CDI; and a century 2000 autoplt coupled to the #1 navigation. Upon arriving in the vicinity of ZZZ at 5000 ft; the ATIS indicated conditions below minimums; RVR 1800 and 100 ft ceiling. A low layer of morning ground fog had formed in the vicinity. ATC requested intentions; we replied we wished to hold as conditions were improving. They provided vectors instead of a hold and requested we climb to 6000 ft to stay clear of other traffic. After about 15 mins; ATC broadcast that conditions at ZZZ had improved to RVR 4000 and 200 ft ceiling. We requested an approach in sequence. We were vectored to the final approach course. As we approached the ILS; we were requested to descend to 5000 ft; then 3500 ft in rapid sequence. We had decided to use the autoplt for the approach and had it engaged in heading mode with a descent. We verified the ILS 33 identify and that the HSI did not have any flags visible. While in the descent; we were given the final turn on vector to join the localizer and to maintain 2400 ft until established. I believe we were approximately 6 miles from the outer. The student increased the rate of descent. After our turn to the intercept heading; the approach mode of the autoplt was engaged and we verified that it annunciated approach and GS coupling. The autoplt maintained the rate of descent (estimated 1000 FPM) and intercepted the localizer course. The GS indicator on the HSI indicated we were on the GS (needle centered; no flags). The landing gear was lowered and the autoplt continued to descend at the 1000 FPM rate and busted through the 2400 ft intercept altitude. I estimate we were still three miles from the outer marker. I realized something was wrong as the GS continued to indicate we were on glide path; but we should have had a fly up indication at this point. I disengaged the autoplt and applied full power; but by this point we had descended to 1900 ft. At almost the same time; ATC transmitted a low altitude alert to us and told us to climb to 2400 ft. I requested the pilot to pitch up and initiate a climb. Because the gear was still down and the pitch insufficiently high; the climb rate was too low. ATC asked if we were climbing. I took the controls; pitched up and raised the gear and climbed to 2400 ft. I advised ATC we were missing the approach and we had an equipment failure in the ILS. All during the miss; the GS flag did not appear and the GS needle indicated we were on GS. After evaluating our options; we decided to return to our home base; which was VFR conditions. We verified the same ILS indications at our home base and we conducted a localizer only approach. This time the intercept was not from above with a high descent rate and we leveled at the intercept altitude before engaging the autoplt. Once again; the GS indications were no flags and GS indicating on glide path. When we engaged the approach function of the autoplt; it just maintained altitude; as that was what the GS indications were indicating. We disengaged the autoplt and landed normally. After on the ground; we discussed with the avionics shop and concluded the KX155 navigation #1 was at fault as both the HSI and autoplt's actions concurred with one another. The KX155 was removed and reseated. It then began to give GS indications other than on glide path. We flew once again to verify that all the systems were operating correctly and there were no problems indicated. Lessons learned. A rapid descent that had us join the final approach course from above contributed to my initial confusion. The GS failure was difficult to detect as we had verified the ILS identify; verified the GS indication without a flag; engaged the autoplt; and got correct annunciation for the autoplt tracking the ILS. Situational awareness should have been better and I should have realized sooner that the GS indications were wrong; as I should have expected a fly up. The high descent rate with the autoplt coupled should have been anotherindicator. I should have initiated the go around and missed approach sooner. Fortunately; the conditions were still VFR on top of the overcast; and ATC did a good job of warning us. If not; we could easily have flown into the ground. I also observe that there is not a good way of verifying that the GS function is working nor is there an equivalent requirement to have it frequently checked as there is for a VOR receiver. If we had a second ILS receiver on board; that might have indicated a discrepancy sooner. From this point on; I will at least verify GS operation while on the ground at an airport that has an ILS installed; at least to the extent that the GS needle will move from the center position (usually a fly up) as part of our taxi/runup. I think consideration should be given to the FAA having a requirement that an ILS operation be operationally checked on some periodic basis; and logged similar to VOR checks. Callback conversation with reporter revealed the following information: reporter stated that they were in a descent on the autoplt and continued to descend because of the centered GS needle and no 'off flags'. Reporter said they could have continued to fly into the ground if they were not warned by the controller. Reporter stated that the KX155 navigation/communication unit has a printed circuit board with etched gold connectors. Reporter advised that the avionics tech who reseated the unit said that there are two pins which control the GS up/down indication and that these pins were apparently not seated correctly during the flight in question. Reporter advised that there is no self test function on the KX155 that would allow the pilot to check that the off flag is functioning.

Google
 

Original NASA ASRS Text

Title: BE33 FOLLOWS ON GLIDE PATH INDICATION 500 FT BELOW GS INTERCEPT ALT WITH NO INSTRUMENT OFF FLAGS VISIBLE ON A COUPLED AUTOPLT APCH.

Narrative: I WAS CONDUCTING AN IFR CROSS COUNTRY TRAINING FLT ON AN IFR FLT PLAN. THE ACFT HAD AN IFR GPS; A #1 NAV/COM OF A KX155 WITH GS INTERFACED TO A HSI; AND A #2 NAV; A NARCO VOR/LOCALIZER/CDI; AND A CENTURY 2000 AUTOPLT COUPLED TO THE #1 NAV. UPON ARRIVING IN THE VICINITY OF ZZZ AT 5000 FT; THE ATIS INDICATED CONDITIONS BELOW MINIMUMS; RVR 1800 AND 100 FT CEILING. A LOW LAYER OF MORNING GND FOG HAD FORMED IN THE VICINITY. ATC REQUESTED INTENTIONS; WE REPLIED WE WISHED TO HOLD AS CONDITIONS WERE IMPROVING. THEY PROVIDED VECTORS INSTEAD OF A HOLD AND REQUESTED WE CLB TO 6000 FT TO STAY CLR OF OTHER TFC. AFTER ABOUT 15 MINS; ATC BROADCAST THAT CONDITIONS AT ZZZ HAD IMPROVED TO RVR 4000 AND 200 FT CEILING. WE REQUESTED AN APCH IN SEQUENCE. WE WERE VECTORED TO THE FINAL APCH COURSE. AS WE APCHED THE ILS; WE WERE REQUESTED TO DSND TO 5000 FT; THEN 3500 FT IN RAPID SEQUENCE. WE HAD DECIDED TO USE THE AUTOPLT FOR THE APCH AND HAD IT ENGAGED IN HDG MODE WITH A DSCNT. WE VERIFIED THE ILS 33 IDENT AND THAT THE HSI DID NOT HAVE ANY FLAGS VISIBLE. WHILE IN THE DSCNT; WE WERE GIVEN THE FINAL TURN ON VECTOR TO JOIN THE LOCALIZER AND TO MAINTAIN 2400 FT UNTIL ESTABLISHED. I BELIEVE WE WERE APPROX 6 MILES FROM THE OUTER. THE STUDENT INCREASED THE RATE OF DSCNT. AFTER OUR TURN TO THE INTERCEPT HDG; THE APCH MODE OF THE AUTOPLT WAS ENGAGED AND WE VERIFIED THAT IT ANNUNCIATED APCH AND GS COUPLING. THE AUTOPLT MAINTAINED THE RATE OF DSCNT (ESTIMATED 1000 FPM) AND INTERCEPTED THE LOCALIZER COURSE. THE GS INDICATOR ON THE HSI INDICATED WE WERE ON THE GS (NEEDLE CENTERED; NO FLAGS). THE LNDG GEAR WAS LOWERED AND THE AUTOPLT CONTINUED TO DSND AT THE 1000 FPM RATE AND BUSTED THROUGH THE 2400 FT INTERCEPT ALT. I ESTIMATE WE WERE STILL THREE MILES FROM THE OUTER MARKER. I REALIZED SOMETHING WAS WRONG AS THE GS CONTINUED TO INDICATE WE WERE ON GLIDE PATH; BUT WE SHOULD HAVE HAD A FLY UP INDICATION AT THIS POINT. I DISENGAGED THE AUTOPLT AND APPLIED FULL POWER; BUT BY THIS POINT WE HAD DSNDED TO 1900 FT. AT ALMOST THE SAME TIME; ATC TRANSMITTED A LOW ALT ALERT TO US AND TOLD US TO CLB TO 2400 FT. I REQUESTED THE PLT TO PITCH UP AND INITIATE A CLB. BECAUSE THE GEAR WAS STILL DOWN AND THE PITCH INSUFFICIENTLY HIGH; THE CLB RATE WAS TOO LOW. ATC ASKED IF WE WERE CLBING. I TOOK THE CTLS; PITCHED UP AND RAISED THE GEAR AND CLBED TO 2400 FT. I ADVISED ATC WE WERE MISSING THE APCH AND WE HAD AN EQUIP FAILURE IN THE ILS. ALL DURING THE MISS; THE GS FLAG DID NOT APPEAR AND THE GS NEEDLE INDICATED WE WERE ON GS. AFTER EVALUATING OUR OPTIONS; WE DECIDED TO RETURN TO OUR HOME BASE; WHICH WAS VFR CONDITIONS. WE VERIFIED THE SAME ILS INDICATIONS AT OUR HOME BASE AND WE CONDUCTED A LOCALIZER ONLY APCH. THIS TIME THE INTERCEPT WAS NOT FROM ABOVE WITH A HIGH DSCNT RATE AND WE LEVELED AT THE INTERCEPT ALT BEFORE ENGAGING THE AUTOPLT. ONCE AGAIN; THE GS INDICATIONS WERE NO FLAGS AND GS INDICATING ON GLIDE PATH. WHEN WE ENGAGED THE APCH FUNCTION OF THE AUTOPLT; IT JUST MAINTAINED ALT; AS THAT WAS WHAT THE GS INDICATIONS WERE INDICATING. WE DISENGAGED THE AUTOPLT AND LANDED NORMALLY. AFTER ON THE GND; WE DISCUSSED WITH THE AVIONICS SHOP AND CONCLUDED THE KX155 NAV #1 WAS AT FAULT AS BOTH THE HSI AND AUTOPLT'S ACTIONS CONCURRED WITH ONE ANOTHER. THE KX155 WAS REMOVED AND RESEATED. IT THEN BEGAN TO GIVE GS INDICATIONS OTHER THAN ON GLIDE PATH. WE FLEW ONCE AGAIN TO VERIFY THAT ALL THE SYSTEMS WERE OPERATING CORRECTLY AND THERE WERE NO PROBS INDICATED. LESSONS LEARNED. A RAPID DSCNT THAT HAD US JOIN THE FINAL APCH COURSE FROM ABOVE CONTRIBUTED TO MY INITIAL CONFUSION. THE GS FAILURE WAS DIFFICULT TO DETECT AS WE HAD VERIFIED THE ILS IDENT; VERIFIED THE GS INDICATION WITHOUT A FLAG; ENGAGED THE AUTOPLT; AND GOT CORRECT ANNUNCIATION FOR THE AUTOPLT TRACKING THE ILS. SITUATIONAL AWARENESS SHOULD HAVE BEEN BETTER AND I SHOULD HAVE REALIZED SOONER THAT THE GS INDICATIONS WERE WRONG; AS I SHOULD HAVE EXPECTED A FLY UP. THE HIGH DSCNT RATE WITH THE AUTOPLT COUPLED SHOULD HAVE BEEN ANOTHERINDICATOR. I SHOULD HAVE INITIATED THE GO AROUND AND MISSED APCH SOONER. FORTUNATELY; THE CONDITIONS WERE STILL VFR ON TOP OF THE OVERCAST; AND ATC DID A GOOD JOB OF WARNING US. IF NOT; WE COULD EASILY HAVE FLOWN INTO THE GND. I ALSO OBSERVE THAT THERE IS NOT A GOOD WAY OF VERIFYING THAT THE GS FUNCTION IS WORKING NOR IS THERE AN EQUIVALENT REQUIREMENT TO HAVE IT FREQUENTLY CHKED AS THERE IS FOR A VOR RECEIVER. IF WE HAD A SECOND ILS RECEIVER ON BOARD; THAT MIGHT HAVE INDICATED A DISCREPANCY SOONER. FROM THIS POINT ON; I WILL AT LEAST VERIFY GS OPERATION WHILE ON THE GND AT AN ARPT THAT HAS AN ILS INSTALLED; AT LEAST TO THE EXTENT THAT THE GS NEEDLE WILL MOVE FROM THE CTR POSITION (USUALLY A FLY UP) AS PART OF OUR TAXI/RUNUP. I THINK CONSIDERATION SHOULD BE GIVEN TO THE FAA HAVING A REQUIREMENT THAT AN ILS OPERATION BE OPERATIONALLY CHKED ON SOME PERIODIC BASIS; AND LOGGED SIMILAR TO VOR CHECKS. CALLBACK CONVERSATION WITH RPTR REVEALED THE FOLLOWING INFO: RPTR STATED THAT THEY WERE IN A DSCNT ON THE AUTOPLT AND CONTINUED TO DSND BECAUSE OF THE CENTERED GS NEEDLE AND NO 'OFF FLAGS'. RPTR SAID THEY COULD HAVE CONTINUED TO FLY INTO THE GND IF THEY WERE NOT WARNED BY THE CTLR. RPTR STATED THAT THE KX155 NAV/COM UNIT HAS A PRINTED CIRCUIT BOARD WITH ETCHED GOLD CONNECTORS. RPTR ADVISED THAT THE AVIONICS TECH WHO RESEATED THE UNIT SAID THAT THERE ARE TWO PINS WHICH CTL THE GS UP/DOWN INDICATION AND THAT THESE PINS WERE APPARENTLY NOT SEATED CORRECTLY DURING THE FLT IN QUESTION. RPTR ADVISED THAT THERE IS NO SELF TEST FUNCTION ON THE KX155 THAT WOULD ALLOW THE PLT TO CHK THAT THE OFF FLAG IS FUNCTIONING.

Data retrieved from NASA's ASRS site as of January 2009 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.