Narrative:

I had departed on an IFR flight plan to mvy. At the time of departure the metar for mvy was good visibility and 300 sct. The total flight time was about 40 minutes. When I was about 20 minutes from mvy the ceiling and visibility had dropped considerably. I don't remember what the first ATIS report was but there were multiple ATIS updates during the remaining flight time in which the ceiling and/or visibility changed considerably. The second to last update I received from the tower or TRACON (I don't remember which) was an RVR of 2400 and ceiling of 200. I considered aborting the approach and returning to ash but an update shortly thereafter was an RVR of 3000 feet. The aircraft I was flying is equipped with a FD (KFC200) and the RVR when using a FD or ap for this field is 1800 feet. I coupled the FD onto the approach and all indicators on the FD was good ('coupled' light illuminated). While almost at DH the FD misbehaved; it started drifting the plane to the right; at this point we were very close to DH. The reason I am reporting this is that at this time I do not know if the a/P uncoupled because I may have accidently hit the a/P disconnect instead of the ptt. I honestly don't know and this is a problem. Because we were so close to the DH; I at first chased the needle back to center but it overshot two dots to the left. I made the decision to do the missed approach when my passenger stated 'there's the runway' I looked up and I was about 50 feet to the left of centerline at about 200-400 feet AGL. I successfully moved the plane slightly to the right and we had an uneventful landing. Tower asked me where I broke out and I at first said 'just about at DH; maybe a little higher' however it was clear that as we taxied down the runway that the ceiling was lower or getting lower very quickly. It may have been only the RVR that was decreasing as we taxied down the runway. My thoughts at this point are not clear. However; on the approach behind us was another small GA aircraft who landed successfully so I am thinking it was RVR further down the runway that was decreasing and not ceiling. I called (by telephone) the tower and tried to further explain; as best I could; what I experienced on the approach. I have yet to go back up in the plane to see if the FD had failed in the approach mode. It was fine during the return trip in enroute navigation and heading mode.I don't believe any fars were violated during this approach however it was 'close' because of the failure of the FD and the rapidly changing conditions on the field at the time. When I called the tower the controller explained this is very common with 'bubbles' (I think that is the term he used) of fog that rolls in off the ocean. There is no taf for mvy; only for ack and hya and duats defers to hya which did not predict a weather situation such as this.

Google
 

Original NASA ASRS Text

Title: A twin engine pilot flying a coupled approach to MVY; recovered and landed at minimums after the autopilot was either inadvertently disconnected or failed.

Narrative: I had departed on an IFR flight plan to MVY. At the time of departure the METAR for MVY was good visibility and 300 sct. The total flight time was about 40 minutes. When I was about 20 minutes from MVY the ceiling and visibility had dropped considerably. I don't remember what the first ATIS report was but there were multiple ATIS updates during the remaining flight time in which the ceiling and/or visibility changed considerably. The second to last update I received from the tower or TRACON (I don't remember which) was an RVR of 2400 and ceiling of 200. I considered aborting the approach and returning to ASH but an update shortly thereafter was an RVR of 3000 feet. The aircraft I was flying is equipped with a FD (KFC200) and the RVR when using a FD or AP for this field is 1800 feet. I coupled the FD onto the approach and all indicators on the FD was good ('Coupled' light illuminated). While almost at DH the FD misbehaved; it started drifting the plane to the right; at this point we were very close to DH. The reason I am reporting this is that at this time I do not know if the A/P uncoupled because I may have accidently hit the A/P disconnect instead of the PTT. I honestly don't know and this is a problem. Because we were so close to the DH; I at first chased the needle back to center but it overshot two dots to the left. I made the decision to do the missed approach when my passenger stated 'There's the runway' I looked up and I was about 50 feet to the left of centerline at about 200-400 feet AGL. I successfully moved the plane slightly to the right and we had an uneventful landing. Tower asked me where I broke out and I at first said 'just about at DH; maybe a little higher' however it was clear that as we taxied down the runway that the ceiling was lower or getting lower very quickly. It may have been only the RVR that was decreasing as we taxied down the runway. My thoughts at this point are not clear. However; on the approach behind us was another small GA aircraft who landed successfully so I am thinking it was RVR further down the runway that was decreasing and not ceiling. I called (by telephone) the tower and tried to further explain; as best I could; what I experienced on the approach. I have yet to go back up in the plane to see if the FD had failed in the approach mode. It was fine during the return trip in enroute NAV and HDG mode.I don't believe any FARs were violated during this approach however it was 'close' because of the failure of the FD and the rapidly changing conditions on the field at the time. When I called the tower the controller explained this is very common with 'bubbles' (I think that is the term he used) of fog that rolls in off the ocean. There is no TAF for MVY; only for ACK and HYA and DUATS defers to HYA which did not predict a weather situation such as this.

Data retrieved from NASA's ASRS site 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.