Narrative:

At XA00 it was dark with approximately 1200 ft ceilings and 3 mi visibility. We were vectored to, and cleared for, the ILS runway 12R at msp. As per normal procedure, this was done on autoplt (coupled approach). Once established on the localizer and GS and inside the FAF, the GS became very erratic. It first went full scale down, which caused the autoplt to 'degrade' and the autoplt caution light illuminated. I disconnected the autoplt. The GS then showed 2 dots up. Then I noted the localizer showed full scale deflection. I executed a go around, since we were in and out of clouds and did not have the airport in sight. There was a NOTAM indicating the GS may be unreliable if an aircraft was on a newly constructed taxiway in front of the GS antenna. Protection for ILS would not be available unless WX was 800 ft 2 mi or less. So, the erratic GS was not a complete surprise. However, ATC was clearing aircraft for the ILS. (The NOTAM was in our WX packet, not on ATIS.) with automated aircraft such as B757 and A320, the crew must decide which approach will be flown well in advance because FMC programming and procedures are different for localizer only (wind dive and drive or VNAV) versus ILS. If you assume the GS will work, you are committed to the ILS or go around. It is unsafe to have an unreliable GS in IFR conditions. This is a continuing problem for this approach -- not an isolated incident. ATC stands by their 800 ft 2 mi 'legality.'

Related reports:

Google
 

Original NASA ASRS Text

Title: A B757 CREW PERFORMS A GAR WHEN RECEIVING FAULTY INDICATIONS ON THEIR ILS RECEIVER, NEGATING THE APCH AT MSP, MN.

Narrative: AT XA00 IT WAS DARK WITH APPROX 1200 FT CEILINGS AND 3 MI VISIBILITY. WE WERE VECTORED TO, AND CLRED FOR, THE ILS RWY 12R AT MSP. AS PER NORMAL PROC, THIS WAS DONE ON AUTOPLT (COUPLED APCH). ONCE ESTABLISHED ON THE LOC AND GS AND INSIDE THE FAF, THE GS BECAME VERY ERRATIC. IT FIRST WENT FULL SCALE DOWN, WHICH CAUSED THE AUTOPLT TO 'DEGRADE' AND THE AUTOPLT CAUTION LIGHT ILLUMINATED. I DISCONNECTED THE AUTOPLT. THE GS THEN SHOWED 2 DOTS UP. THEN I NOTED THE LOC SHOWED FULL SCALE DEFLECTION. I EXECUTED A GAR, SINCE WE WERE IN AND OUT OF CLOUDS AND DID NOT HAVE THE ARPT IN SIGHT. THERE WAS A NOTAM INDICATING THE GS MAY BE UNRELIABLE IF AN ACFT WAS ON A NEWLY CONSTRUCTED TXWY IN FRONT OF THE GS ANTENNA. PROTECTION FOR ILS WOULD NOT BE AVAILABLE UNLESS WX WAS 800 FT 2 MI OR LESS. SO, THE ERRATIC GS WAS NOT A COMPLETE SURPRISE. HOWEVER, ATC WAS CLRING ACFT FOR THE ILS. (THE NOTAM WAS IN OUR WX PACKET, NOT ON ATIS.) WITH AUTOMATED ACFT SUCH AS B757 AND A320, THE CREW MUST DECIDE WHICH APCH WILL BE FLOWN WELL IN ADVANCE BECAUSE FMC PROGRAMMING AND PROCS ARE DIFFERENT FOR LOC ONLY (WIND DIVE AND DRIVE OR VNAV) VERSUS ILS. IF YOU ASSUME THE GS WILL WORK, YOU ARE COMMITTED TO THE ILS OR GAR. IT IS UNSAFE TO HAVE AN UNRELIABLE GS IN IFR CONDITIONS. THIS IS A CONTINUING PROB FOR THIS APCH -- NOT AN ISOLATED INCIDENT. ATC STANDS BY THEIR 800 FT 2 MI 'LEGALITY.'

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.