Narrative:

Incident 1: these were two separate events the first involved a B767 and another air carrier. The final had gotten slowed way down (about 150 KTS at 15 mile final). The B767 was overtaking the other air carrier both established on the prm soia [simultaneous offset instrument approaches]. I turned the B767 off of the course to build spacing and when I got him back towards the final I cleared him just prior to being established on the prm localizer. This was considered an error because both aircraft must be established prior to the 28R aircraft being cleared. Just some insight into what was going on in the control room: prior to this incident I was ensuring every aircraft was established (and that [runway] 28L traffic was also established) before clearing; however I was getting a lot of north downwind traffic [down the bay] which is unusual for this operation so I was told that as long as the [runway] 28L aircraft was established I can clear my aircraft as this would help get the aircraft down prior to being established because many of them were left high to stay on top of the [runway] 28L traffic. This was my first ever session working soia at foster and we have not run soia much at all this year; my initial memory of the rules was that both aircraft had to be established however considering my inexperience I was very open to taking advice; I obviously should not have been. I spoke with my area manager about this already but every month we have to do soia monitor refresher training and because the soia procedures can be somewhat complex I suggested that we put together a list of best practices/hard rules for the soia operation to be handed out at monitor refresher training. The fact that we have to do monitor refresher training every month but there is no soia finals refresher training is not very wise in my opinion.incident 2: aircraft X and Y; both A320's. Aircraft X was on the north downwind at 210 KTS IAS; aircraft Y was established on the [runway] 28L prm final already at final approach speed. There was also an air carrier about a 15 mile final already slowed to 150 KTS; in my opinion it would have been tougher to stay behind the other air carrier who was already slowed therefore I decided to catch aircraft Y. I kept X above Y until X was on a heading to intercept; about a 1/4 mile prior to X being established I cleared him for the approach (see incident 1). Approximately 5 seconds prior to clearing aircraft X the woodside controller had to break aircraft Y out because of a wake turbulence situation with the aircraft ahead. In my opinion there was absolutely no way for me to keep separation because the woodside controller must turn and climb above 3;000 (really 4;000 ft because there is a 3;700 ft MVA just south of the final) according to the soia breakout procedures and aircraft X is on a heading to join so separation was lost instantly as aircraft Y climbed. Even if I would have broken aircraft X out the instructions call for a very slight right turn and climb therefore both aircraft would have been climbing and there is no way I would be able to ensure 1;000 ft separation since one aircraft may have climbed faster than the other. Since woodside was already turning out to the left and climbing it seemed safest to me to keep my aircraft going in on the approach. Many controllers were unaware (myself included) of the fact that when one of the aircraft in the soia pair is broken out due to anything other than a ntz violation the other must also be broken out as this is opposite of the breakout procedures for a ntz violation were the violator is returned to course to continue on the approach.

Google
 

Original NASA ASRS Text

Title: NCT Approach Controller describes two incidents of loss of separation while conducting PRM SOIA approaches at SFO.

Narrative: Incident 1: These were two separate events the first involved a B767 and another air carrier. The final had gotten slowed way down (about 150 KTS at 15 mile final). The B767 was overtaking the other air carrier both established on the PRM SOIA [Simultaneous Offset Instrument Approaches]. I turned the B767 off of the course to build spacing and when I got him back towards the final I cleared him just prior to being established on the PRM localizer. This was considered an error because both aircraft must be established prior to the 28R aircraft being cleared. Just some insight into what was going on in the control room: prior to this incident I was ensuring every aircraft was established (and that [Runway] 28L traffic was also established) before clearing; however I was getting a lot of north downwind traffic [down the bay] which is unusual for this operation so I was told that as long as the [Runway] 28L aircraft was established I can clear my aircraft as this would help get the aircraft down prior to being established because many of them were left high to stay on top of the [Runway] 28L traffic. This was my first ever session working SOIA at Foster and we have not run SOIA much at all this year; my initial memory of the rules was that both aircraft had to be established however considering my inexperience I was very open to taking advice; I obviously should not have been. I spoke with my Area Manager about this already but every month we have to do SOIA monitor refresher training and because the SOIA procedures can be somewhat complex I suggested that we put together a list of best practices/hard rules for the SOIA operation to be handed out at monitor refresher training. The fact that we have to do monitor refresher training every month but there is no SOIA finals refresher training is not very wise in my opinion.Incident 2: Aircraft X and Y; both A320's. Aircraft X was on the north downwind at 210 KTS IAS; Aircraft Y was established on the [Runway] 28L PRM final already at final approach speed. There was also an air carrier about a 15 mile final already slowed to 150 KTS; in my opinion it would have been tougher to stay behind the other air carrier who was already slowed therefore I decided to catch Aircraft Y. I kept X above Y until X was on a heading to intercept; about a 1/4 mile prior to X being established I cleared him for the approach (see incident 1). Approximately 5 seconds prior to clearing Aircraft X the Woodside Controller had to break Aircraft Y out because of a wake turbulence situation with the aircraft ahead. In my opinion there was absolutely no way for me to keep separation because the Woodside Controller MUST turn and CLIMB above 3;000 (really 4;000 FT because there is a 3;700 FT MVA just south of the final) according to the SOIA breakout procedures and Aircraft X is on a heading to join so separation was lost instantly as Aircraft Y climbed. Even if I would have broken Aircraft X out the instructions call for a very slight right turn and climb therefore both aircraft would have been climbing and there is no way I would be able to ensure 1;000 FT separation since one aircraft may have climbed faster than the other. Since Woodside was already turning out to the left and climbing it seemed safest to me to keep my aircraft going in on the approach. Many controllers were unaware (myself included) of the fact that when one of the aircraft in the SOIA pair is broken out due to anything other than a NTZ violation the other must also be broken out as this is opposite of the breakout procedures for a NTZ violation were the violator is returned to course to continue on the approach.

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.