Narrative:

This was copilot's leg, sfo-smo. After the before takeoff briefing, copilot performed normal takeoff from sfo runway 1L. Offshore SID prescribes tracking sfo 351 degree right for 4 mi, left turn to 200 degrees to intercept pt reyes 151 degree right. After entering clouds at 2800' and 4 mi, copilot started left turn to 200 degrees. ATC then assigned 210 degrees. At about 200 degrees, copilot steepened bank to more than 45 degrees. I noticed this after he passed through 210 degrees and told him to turn right and reestablish 210 degree heading. The bank was eventually reduced to approximately 25 degree left bank, but he still was turning away from assigned 210 degree heading. I looked at his horizon and saw a right bank while my horizon indicated left turn/bank and heading about 130 degrees. I took controls and leveled wings by my horizon (still IMC). At this time, ATC noticed improper heading and gave us right turn to west to reestablish us on SID. While copilot was turning left prior to my taking controllers, I asked for right turn to 210 degrees at least 4 times and he replied, 'I am turning right.' I should have taken controls sooner, but it took me some time to identify/rectify error, as I was performing after takeoff, climb checklist and setting up next navigation segment. Copilot could not or did not recognize conflicting instrument information, horizon indicating right turn while dg indicating left turn. A new copilot's horizon has been installed and copilot will receive partial panel/instrument failure recognition training. I will not allow sic's to fly IMC legs any longer. Callback conversation with reporter revealed the following: callback made to counsel reporter re: CRM techniques. Usual scenario in CRM involves a 2 request environment. That is, after 2 requests to the other crew member and the situation doesn't change, assume command at that point. The company that reporter flies for had 2 aircraft and several capts and first officer's. They just recently went to a flight safety organization for CRM training, although first officer's have been excluded from this type of training in the past. Sterile cockpit and checklist use under 10000' were talked about and reporter had not really been exposed to the sterile cockpit philosophy nor the CRM training as he has only been with the company for 6 months. The first officer was low time, under 1000 hours and had a time problem with instrument scanning. Tended to blame scan failures on the initial instrument failure instead. Reporter very receptive to counseling and better crew coordination techniques.

Related reports:

Google
 

Original NASA ASRS Text

Title: HEADING TRACK DEVIATION WHILE ON SID DEP AFTER ARTIFICIAL HORIZON FAILS ON FO SIDE.

Narrative: THIS WAS COPLT'S LEG, SFO-SMO. AFTER THE BEFORE TKOF BRIEFING, COPLT PERFORMED NORMAL TKOF FROM SFO RWY 1L. OFFSHORE SID PRESCRIBES TRACKING SFO 351 DEG R FOR 4 MI, LEFT TURN TO 200 DEGS TO INTERCEPT PT REYES 151 DEG R. AFTER ENTERING CLOUDS AT 2800' AND 4 MI, COPLT STARTED L TURN TO 200 DEGS. ATC THEN ASSIGNED 210 DEGS. AT ABOUT 200 DEGS, COPLT STEEPENED BANK TO MORE THAN 45 DEGS. I NOTICED THIS AFTER HE PASSED THROUGH 210 DEGS AND TOLD HIM TO TURN R AND REESTABLISH 210 DEG HDG. THE BANK WAS EVENTUALLY REDUCED TO APPROX 25 DEG LEFT BANK, BUT HE STILL WAS TURNING AWAY FROM ASSIGNED 210 DEG HDG. I LOOKED AT HIS HORIZON AND SAW A R BANK WHILE MY HORIZON INDICATED L TURN/BANK AND HDG ABOUT 130 DEGS. I TOOK CTLS AND LEVELED WINGS BY MY HORIZON (STILL IMC). AT THIS TIME, ATC NOTICED IMPROPER HDG AND GAVE US R TURN TO W TO REESTABLISH US ON SID. WHILE COPLT WAS TURNING L PRIOR TO MY TAKING CTLRS, I ASKED FOR R TURN TO 210 DEGS AT LEAST 4 TIMES AND HE REPLIED, 'I AM TURNING R.' I SHOULD HAVE TAKEN CTLS SOONER, BUT IT TOOK ME SOME TIME TO IDENT/RECTIFY ERROR, AS I WAS PERFORMING AFTER TKOF, CLB CHKLIST AND SETTING UP NEXT NAV SEGMENT. COPLT COULD NOT OR DID NOT RECOGNIZE CONFLICTING INSTRUMENT INFO, HORIZON INDICATING R TURN WHILE DG INDICATING L TURN. A NEW COPLT'S HORIZON HAS BEEN INSTALLED AND COPLT WILL RECEIVE PARTIAL PANEL/INST FAILURE RECOGNITION TRAINING. I WILL NOT ALLOW SIC'S TO FLY IMC LEGS ANY LONGER. CALLBACK CONVERSATION WITH RPTR REVEALED THE FOLLOWING: CALLBACK MADE TO COUNSEL RPTR RE: CRM TECHNIQUES. USUAL SCENARIO IN CRM INVOLVES A 2 REQUEST ENVIRONMENT. THAT IS, AFTER 2 REQUESTS TO THE OTHER CREW MEMBER AND THE SIT DOESN'T CHANGE, ASSUME COMMAND AT THAT POINT. THE COMPANY THAT RPTR FLIES FOR HAD 2 ACFT AND SEVERAL CAPTS AND F/O'S. THEY JUST RECENTLY WENT TO A FLT SAFETY ORGANIZATION FOR CRM TRAINING, ALTHOUGH F/O'S HAVE BEEN EXCLUDED FROM THIS TYPE OF TRAINING IN THE PAST. STERILE COCKPIT AND CHKLIST USE UNDER 10000' WERE TALKED ABOUT AND RPTR HAD NOT REALLY BEEN EXPOSED TO THE STERILE COCKPIT PHILOSOPHY NOR THE CRM TRAINING AS HE HAS ONLY BEEN WITH THE COMPANY FOR 6 MONTHS. THE F/O WAS LOW TIME, UNDER 1000 HRS AND HAD A TIME PROB WITH INSTRUMENT SCANNING. TENDED TO BLAME SCAN FAILURES ON THE INITIAL INSTRUMENT FAILURE INSTEAD. RPTR VERY RECEPTIVE TO COUNSELING AND BETTER CREW COORD TECHNIQUES.

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.