Narrative:

The aircraft entered instrument conditions upon departure from gaithersburg; md; on an IFR flight to lynchburg; va. Shortly after the aircraft came level at 6000 ft MSL; something went wrong with the attitude indicator (ai). The gyro tumbled vertically; like an eyeball rotating upward into a person's head. Based on troubleshooting after the flight; the anomaly apparently was caused by partial loss of pressure in the aircraft's vacuum system. An avionics technician said he found leakage in a hose going to both the attitude indicator and the directional gyro. The loss apparently was not enough to trigger the 'lo vacuum' warning light on the annunciator panel; which did not illuminate. (The light had worked ok in the preflight check before departure.) the directional gyro continued to function normally; despite the problem with the attitude indicator. At the time; the flight was cleared direct to the martinsburg VOR; which would have been a magnetic course of approximately 280 degrees. There are some gaps in my knowledge and understanding of the ensuing developments; but I am able to report the following: I ignored the extreme nose-down indication on the attitude indicator; which was so incongruous as to be an obvious instrument problem. I started comparing the l-r bank indications on the attitude indicator against the turn coordinator; finding that they still seemed to agree. I shifted my focus to the turn coordinator; in an effort to keep the airplane upright; but somehow allowed the plane to drift off course to the right. Just as I noticed that the heading on the directional gyro was some 90 degrees to the right; ATC called and said we were tracking north instead of west. Around the same time; I noticed the altimeter unwinding rapidly. We lost 400 ft before I was able to arrest the descent at 5600 ft. I do not know how I first detected that the aircraft was descending. I don't know if it was the changing altimeter reading or perhaps the sound of the engine. I did not take note of either the airspeed indicator or the vertical speed indicator. I also do not know how I realized that the situation required a correction to the left -- whether that was from reading the attitude indicator or turn coordinator or simply from knowing that the track had deviated to the right. Let me just say that I am grateful for training received from the mooney aircraft safety foundation in the recognition of and recovery from an incipient spiral. I credit this training with enabling me to make the correct control inputs to regain control of the airplane. I did not detect any G forces either during the uncontrolled descent or during the recovery. With the airplane back on course and back at altitude; I covered the attitude indicator with a rubber patch and began to sort out our options. I would like to have engaged the autoplt at this point; but believe the autoplt depends for guidance on the attitude indicator and therefore would have been unavailable. Luckily; we soon broke out of solid IMC and found ourselves between layers. Leaving the frequency; with permission; I called flight service and verified that we were heading toward better conditions. The layers eventually gave way to clear WX; and the flight proceeded uneventfully to the destination.

Google
 

Original NASA ASRS Text

Title: MOONEY M20J PLT HAS LOSS OF THE ATTITUDE INDICATOR (AI) WHILE IN IMC.

Narrative: THE ACFT ENTERED INST CONDITIONS UPON DEP FROM GAITHERSBURG; MD; ON AN IFR FLT TO LYNCHBURG; VA. SHORTLY AFTER THE ACFT CAME LEVEL AT 6000 FT MSL; SOMETHING WENT WRONG WITH THE ATTITUDE INDICATOR (AI). THE GYRO TUMBLED VERTICALLY; LIKE AN EYEBALL ROTATING UPWARD INTO A PERSON'S HEAD. BASED ON TROUBLESHOOTING AFTER THE FLT; THE ANOMALY APPARENTLY WAS CAUSED BY PARTIAL LOSS OF PRESSURE IN THE ACFT'S VACUUM SYS. AN AVIONICS TECHNICIAN SAID HE FOUND LEAKAGE IN A HOSE GOING TO BOTH THE ATTITUDE INDICATOR AND THE DIRECTIONAL GYRO. THE LOSS APPARENTLY WAS NOT ENOUGH TO TRIGGER THE 'LO VACUUM' WARNING LIGHT ON THE ANNUNCIATOR PANEL; WHICH DID NOT ILLUMINATE. (THE LIGHT HAD WORKED OK IN THE PREFLT CHK BEFORE DEP.) THE DIRECTIONAL GYRO CONTINUED TO FUNCTION NORMALLY; DESPITE THE PROB WITH THE ATTITUDE INDICATOR. AT THE TIME; THE FLT WAS CLRED DIRECT TO THE MARTINSBURG VOR; WHICH WOULD HAVE BEEN A MAGNETIC COURSE OF APPROX 280 DEGS. THERE ARE SOME GAPS IN MY KNOWLEDGE AND UNDERSTANDING OF THE ENSUING DEVELOPMENTS; BUT I AM ABLE TO RPT THE FOLLOWING: I IGNORED THE EXTREME NOSE-DOWN INDICATION ON THE ATTITUDE INDICATOR; WHICH WAS SO INCONGRUOUS AS TO BE AN OBVIOUS INST PROB. I STARTED COMPARING THE L-R BANK INDICATIONS ON THE ATTITUDE INDICATOR AGAINST THE TURN COORDINATOR; FINDING THAT THEY STILL SEEMED TO AGREE. I SHIFTED MY FOCUS TO THE TURN COORDINATOR; IN AN EFFORT TO KEEP THE AIRPLANE UPRIGHT; BUT SOMEHOW ALLOWED THE PLANE TO DRIFT OFF COURSE TO THE R. JUST AS I NOTICED THAT THE HDG ON THE DIRECTIONAL GYRO WAS SOME 90 DEGS TO THE R; ATC CALLED AND SAID WE WERE TRACKING N INSTEAD OF W. AROUND THE SAME TIME; I NOTICED THE ALTIMETER UNWINDING RAPIDLY. WE LOST 400 FT BEFORE I WAS ABLE TO ARREST THE DSCNT AT 5600 FT. I DO NOT KNOW HOW I FIRST DETECTED THAT THE ACFT WAS DSNDING. I DON'T KNOW IF IT WAS THE CHANGING ALTIMETER READING OR PERHAPS THE SOUND OF THE ENG. I DID NOT TAKE NOTE OF EITHER THE AIRSPD INDICATOR OR THE VERT SPD INDICATOR. I ALSO DO NOT KNOW HOW I REALIZED THAT THE SITUATION REQUIRED A CORRECTION TO THE L -- WHETHER THAT WAS FROM READING THE ATTITUDE INDICATOR OR TURN COORDINATOR OR SIMPLY FROM KNOWING THAT THE TRACK HAD DEVIATED TO THE R. LET ME JUST SAY THAT I AM GRATEFUL FOR TRAINING RECEIVED FROM THE MOONEY ACFT SAFETY FOUNDATION IN THE RECOGNITION OF AND RECOVERY FROM AN INCIPIENT SPIRAL. I CREDIT THIS TRAINING WITH ENABLING ME TO MAKE THE CORRECT CTL INPUTS TO REGAIN CTL OF THE AIRPLANE. I DID NOT DETECT ANY G FORCES EITHER DURING THE UNCTLED DSCNT OR DURING THE RECOVERY. WITH THE AIRPLANE BACK ON COURSE AND BACK AT ALT; I COVERED THE ATTITUDE INDICATOR WITH A RUBBER PATCH AND BEGAN TO SORT OUT OUR OPTIONS. I WOULD LIKE TO HAVE ENGAGED THE AUTOPLT AT THIS POINT; BUT BELIEVE THE AUTOPLT DEPENDS FOR GUIDANCE ON THE ATTITUDE INDICATOR AND THEREFORE WOULD HAVE BEEN UNAVAILABLE. LUCKILY; WE SOON BROKE OUT OF SOLID IMC AND FOUND OURSELVES BTWN LAYERS. LEAVING THE FREQ; WITH PERMISSION; I CALLED FLT SVC AND VERIFIED THAT WE WERE HEADING TOWARD BETTER CONDITIONS. THE LAYERS EVENTUALLY GAVE WAY TO CLR WX; AND THE FLT PROCEEDED UNEVENTFULLY TO THE DEST.

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.