Narrative:

Departed ZZZ in IMC for VFR on-top. Tracked runway heading; engaged the autoplt; noted the autoplt was inoperative. Turned autoplt off and continued climb. Began on-course turn and noticed directional gyro was sluggish. Stayed in turn and climb as I contacted approach. Noted at this point that directional gyro was frozen. Approach asked verification of departure procedure. I simultaneously checked compass heading and realized that I was off course. As I checked compass heading; I broke out of the clouds; canceled IFR and verified that directional gyro was frozen. Recycled autoplt one more time and verified that autoplt was inoperative; checked compass heading; reset directional gyro and directional gyro came back on-line. Continued VFR flight frequently checking directional gyro against compass heading. Completed trip without further directional gyro problems. Autoplt remained inoperative. The frozen directional gyro problem appears to be associated with the inoperative autoplt (directional gyro has heading bug coupled to the autoplt). An appointment has been set with an avionics shop to have the autoplt/directional gyro problem corrected. Had the fog layer been thicker or a less experienced pilot at the controls; this could have led to a disastrous conclusion. It appears that a failsafe disconnect between the autoplt and directional gyro should be in place in the event of autoplt failure. Callback conversation with reporter revealed the following information: reporter stated the piper arrow came with a 'piper autopilot 3-B' and a mitchel 52d54 directional gyro (dg). Reporter also stated that it wasn't until after breaking out of the clouds on top of the fog; did he realize the initial attempt to turn 90 degrees from a 110 to 180 degree heading; actually ended up as a 180 degree turn to 270. In what appeared to be a short time it took to realize the dg was not only sluggish; but frozen; he was amazed he had turned 180 degrees. The mitchel dg has been replaced with an edo-aire type dg unit. Reporter concluded by stating had he crashed; most likely the accident would have been considered pilot error; when; in fact; the autopilot was inoperative and the dg apparently did not; or could not; uncouple from the failed autopilot.

Related reports:

Google
 

Original NASA ASRS Text

Title: AFTER DEPARTURE IN IMC CONDITIONS; PILOT FLYING A PIPER ARROW REALIZES THE AUTOPILOT IS INOPERATIVE AND THE DIRECTIONAL GYRO (DG) WAS FROZEN.

Narrative: DEPARTED ZZZ IN IMC FOR VFR ON-TOP. TRACKED RWY HDG; ENGAGED THE AUTOPLT; NOTED THE AUTOPLT WAS INOP. TURNED AUTOPLT OFF AND CONTINUED CLB. BEGAN ON-COURSE TURN AND NOTICED DIRECTIONAL GYRO WAS SLUGGISH. STAYED IN TURN AND CLB AS I CONTACTED APCH. NOTED AT THIS POINT THAT DIRECTIONAL GYRO WAS FROZEN. APCH ASKED VERIFICATION OF DEP PROC. I SIMULTANEOUSLY CHKED COMPASS HDG AND REALIZED THAT I WAS OFF COURSE. AS I CHKED COMPASS HDG; I BROKE OUT OF THE CLOUDS; CANCELED IFR AND VERIFIED THAT DIRECTIONAL GYRO WAS FROZEN. RECYCLED AUTOPLT ONE MORE TIME AND VERIFIED THAT AUTOPLT WAS INOP; CHKED COMPASS HDG; RESET DIRECTIONAL GYRO AND DIRECTIONAL GYRO CAME BACK ON-LINE. CONTINUED VFR FLT FREQUENTLY CHKING DIRECTIONAL GYRO AGAINST COMPASS HDG. COMPLETED TRIP WITHOUT FURTHER DIRECTIONAL GYRO PROBS. AUTOPLT REMAINED INOP. THE FROZEN DIRECTIONAL GYRO PROB APPEARS TO BE ASSOCIATED WITH THE INOP AUTOPLT (DIRECTIONAL GYRO HAS HDG BUG COUPLED TO THE AUTOPLT). AN APPOINTMENT HAS BEEN SET WITH AN AVIONICS SHOP TO HAVE THE AUTOPLT/DIRECTIONAL GYRO PROB CORRECTED. HAD THE FOG LAYER BEEN THICKER OR A LESS EXPERIENCED PLT AT THE CTLS; THIS COULD HAVE LED TO A DISASTROUS CONCLUSION. IT APPEARS THAT A FAILSAFE DISCONNECT BTWN THE AUTOPLT AND DIRECTIONAL GYRO SHOULD BE IN PLACE IN THE EVENT OF AUTOPLT FAILURE. CALLBACK CONVERSATION WITH RPTR REVEALED THE FOLLOWING INFO: REPORTER STATED THE PIPER ARROW CAME WITH A 'PIPER AUTOPILOT 3-B' AND A MITCHEL 52D54 DIRECTIONAL GYRO (DG). REPORTER ALSO STATED THAT IT WASN'T UNTIL AFTER BREAKING OUT OF THE CLOUDS ON TOP OF THE FOG; DID HE REALIZE THE INITIAL ATTEMPT TO TURN 90 DEGREES FROM A 110 TO 180 DEGREE HEADING; ACTUALLY ENDED UP AS A 180 DEGREE TURN TO 270. IN WHAT APPEARED TO BE A SHORT TIME IT TOOK TO REALIZE THE DG WAS NOT ONLY SLUGGISH; BUT FROZEN; HE WAS AMAZED HE HAD TURNED 180 DEGREES. THE MITCHEL DG HAS BEEN REPLACED WITH AN EDO-AIRE TYPE DG UNIT. REPORTER CONCLUDED BY STATING HAD HE CRASHED; MOST LIKELY THE ACCIDENT WOULD HAVE BEEN CONSIDERED PILOT ERROR; WHEN; IN FACT; THE AUTOPILOT WAS INOPERATIVE AND THE DG APPARENTLY DID NOT; OR COULD NOT; UNCOUPLE FROM THE FAILED AUTOPILOT.

Data retrieved from NASA's ASRS site as of May 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.