Narrative:

Our flight was normal until about 50 NM from the top of descent. My first officer and I were setting up for the RNAV (rnp) Z runway 10L when I noticed my inboard du (display unit) and the upper du blink. I then noticed a dsply source 1 annunciation in the bottom left corner of my outboard du. I verbalized this to the first officer and had him get out the QRH. He found the dsply source checklist and proceeded to run it. About this time the annunciation went away. We reviewed the checklist and concurred that no further action was required. Because of the momentary failure we discussed the legality of executing a rnp approach and decided that we could do so. We briefed the arrival and the approach; and were just about to run the approach descent checklist when the dsply source 1 annunciation returned. Again we got out the QRH and started the checklist. The light again went out after about 60 seconds. As we were now past the TOD and had not received descent clearance we asked for a lower altitude and got the boards out. This was required to regain the path while honoring the 280 knot transition airspeed restriction that is published on the HHOOD3 RNAV arrival. About this point; the flight attendants called up wanting the seat belt sign on for light turbulence. We complied and were grateful for their call as cockpit workload had suddenly gotten very high and it might have been missed.the first officer and I again discussed the wisdom of doing the rvav rnp as the dsply source 1 annunciation intermittently was illuminated for a total of 5 or 6 times. We consulted QRH. Under the section 'malfunction or required equipment' we were confused by the use of the word 'or' in the title. We decided it should read 'of'. We were also confused by the terminology used in the body of the text where it reads 'not authorized for single or dual failure of any equipment item'. We discussed this point and decided it meant any required equipment item as listed but were not completely sure of this interpretation. Looking at another page did not help us decide if an intermittent dsply source 1 annotation would be disqualifying for an rnp approach so I made the command decision to apply a very strict reading of QRH. I directed the first officer to set up and brief the columbia visual backed up ILS 10L. He set up and quickly briefed the approach. About this time the dsply source 1 annunciation came on for several minutes so we ran the QRH checklist in its entirety and then came back together and verified that we had covered all bases with respect to the failure. To say that our RNAV descent was busy would be a massive understatement. The HHOOD3 has several required speed changes and multiple crossing restrictions. Dealing with an equipment malfunction; running a QRH; interpreting poorly written rnp guidance and briefing multiple approaches while trying to regain the path after a late descent clearance taxed us to the max. As far as I can tell; we flew the lateral; vertical and speed profile without error but this was very difficult to do considering the workload. I used the vsd mode on my primary flight display and that was tremendously helpful in maintaining my overall situational awareness. (Very few of my fos use this tool and they should as it gives instantaneous situational awareness of vertical path.) it helped greatly that I had a very capable first officer on this leg. Using all of our CRM tools the two of us managed the threat and got everything done (including the much delayed approach descent checklist) by about FL200. Past blrun on the HHOOD3 the dsply source 1 annunciation illuminated again and showed us something completely new. The FMC CDU scratchpad displayed discontinuity and I believe the aircraft went into cws pitch and roll mode. I can't say that I saw cws annunciated but as the autopilot did not disconnect and the flight path did not change it seems logical that we defaulted into cws. I was very confused by this new failure mode and double clutched the waypoint under 1L. This madeblrun the active waypoint. This was very wrong as we were well past blrun and descending to cross ssdee. I selected ssdee to the scratchpad and moved to 1L. I then confirmed it with the first officer and executed it. By this point I had had enough. Cockpit workload was way too high; and our situational awareness had suddenly become way too low. Most concerning was that for an unknown reason the FMC had shown us a discontinuity and apparently resequenced itself to a waypoint we had already passed. I directed the first officer to tell approach that we were unable the RNAV arrival and that we needed a vector. She gave us a 270 vector with no altitude assigned. We then asked her for an altitude assignment and she realized her mistake and cleared us to descend and maintain 5000 feet.as the flying pilot; I selected lvl chg and heading sel and complied with our clearances as we were given vectors to a short visual approach. We successfully managed this new challenge by using CRM to recognize and verbalize the threat of a high energy approach. Once we realized we were getting the slam dunk; we used timely speedbrakes; an early gear extension and flaps 30 outside the marker to get back on the vertical profile and meet the stabilized approach criteria. The rest of the flight was uneventful.I believe that working together as a crew; we handled everything correctly and within required navigational standards. I am submitting this report mainly for tracking purposes. This is the second time I have had to deal with a degraded FMS on an RNAV arrival within the last 30 days. The first time; the FMC on the flying pilot's side failed; the autopilot disconnected; and went into cws pitch and roll while descending into [a different airport]. That was also a very challenging scenario that I wish I had filed a report for but I did not. Regardless; even a momentary loss of flight path data on an RNAV arrival and has very significant implications for the safe conduct of the flight and it might be an excellent training scenario. It's even more critical if you are planning to use an rnp approach and then suddenly can't. My final concern is that as more and more airports are transitioning to rnp approaches (in particular in the caribbean); even a minor FMC failure might mean that you can't land at your intended destination. I'm not sure that the company's current fuel policy addresses this issue and can envision a scenario where a crew finds itself challenged with few options and not much gas. Regardless; two failures that degrade my ability to fly an RNAV arrival and/or shoot an rnp approach in less than 30 days has gotten my attention and I hope it gets yours.

Related reports:

Google
 

Original NASA ASRS Text

Title: B737 Captain reported multiple FMS malfunctions on the HHOOD3 Arrival and RNAV (RNP) Z Runway 10L to PDX. Captain reported a visual landing.

Narrative: Our flight was normal until about 50 NM from the top of descent. My FO and I were setting up for the RNAV (RNP) Z RWY 10L when I noticed my inboard DU (Display Unit) and the upper DU blink. I then noticed a DSPLY SOURCE 1 annunciation in the bottom left corner of my outboard DU. I verbalized this to the FO and had him get out the QRH. He found the DSPLY SOURCE Checklist and proceeded to run it. About this time the annunciation went away. We reviewed the checklist and concurred that no further action was required. Because of the momentary failure we discussed the legality of executing a RNP approach and decided that we could do so. We briefed the arrival and the approach; and were just about to run the approach Descent Checklist when the DSPLY SOURCE 1 annunciation returned. Again we got out the QRH and started the checklist. The light again went out after about 60 seconds. As we were now past the TOD and had not received descent clearance we asked for a lower altitude and got the boards out. This was required to regain the path while honoring the 280 knot transition airspeed restriction that is published on the HHOOD3 RNAV arrival. About this point; the Flight Attendants called up wanting the seat belt sign on for light turbulence. We complied and were grateful for their call as cockpit workload had suddenly gotten very high and it might have been missed.The FO and I again discussed the wisdom of doing the RVAV RNP as the DSPLY SOURCE 1 annunciation intermittently was illuminated for a total of 5 or 6 times. We consulted QRH. Under the section 'Malfunction OR Required Equipment' we were confused by the use of the word 'OR' in the title. We decided it should read 'of'. We were also confused by the terminology used in the body of the text where it reads 'not authorized for single or dual failure of any equipment item'. We discussed this point and decided it meant any required equipment item as listed but were not completely sure of this interpretation. Looking at another page did not help us decide if an intermittent DSPLY SOURCE 1 annotation would be disqualifying for an RNP approach so I made the command decision to apply a very strict reading of QRH. I directed the FO to set up and brief the Columbia Visual backed up ILS 10L. He set up and quickly briefed the approach. About this time the DSPLY SOURCE 1 annunciation came on for several minutes so we ran the QRH checklist in its entirety and then came back together and verified that we had covered all bases with respect to the failure. To say that our RNAV descent was busy would be a massive understatement. The HHOOD3 has several required speed changes and multiple crossing restrictions. Dealing with an equipment malfunction; running a QRH; interpreting poorly written RNP guidance and briefing multiple approaches while trying to regain the path after a late descent clearance taxed us to the max. As far as I can tell; we flew the lateral; vertical and speed profile without error but this was very difficult to do considering the workload. I used the VSD mode on my Primary Flight Display and that was tremendously helpful in maintaining my overall situational awareness. (Very few of my FOs use this tool and they should as it gives instantaneous situational awareness of vertical path.) It helped greatly that I had a very capable FO on this leg. Using all of our CRM tools the two of us managed the threat and got everything done (including the much delayed approach descent checklist) by about FL200. Past BLRUN on the HHOOD3 the DSPLY SOURCE 1 annunciation illuminated again and showed us something completely new. The FMC CDU scratchpad displayed DISCONTINUITY and I believe the aircraft went into CWS Pitch and Roll mode. I can't say that I saw CWS annunciated but as the autopilot did not disconnect and the flight path did not change it seems logical that we defaulted into CWS. I was very confused by this new failure mode and double clutched the waypoint under 1L. This madeBLRUN the active waypoint. This was very wrong as we were well past BLRUN and descending to cross SSDEE. I selected SSDEE to the scratchpad and moved to 1L. I then confirmed it with the FO and executed it. By this point I had had enough. Cockpit workload was way too high; and our situational awareness had suddenly become way too low. Most concerning was that for an unknown reason the FMC had shown us a discontinuity and apparently resequenced itself to a waypoint we had already passed. I directed the FO to tell approach that we were unable the RNAV arrival and that we needed a vector. She gave us a 270 vector with no altitude assigned. We then asked her for an altitude assignment and she realized her mistake and cleared us to descend and maintain 5000 feet.As the flying pilot; I selected LVL CHG and HDG SEL and complied with our clearances as we were given vectors to a short visual approach. We successfully managed this new challenge by using CRM to recognize and verbalize the threat of a high energy approach. Once we realized we were getting the slam dunk; we used timely speedbrakes; an early gear extension and flaps 30 outside the marker to get back on the vertical profile and meet the stabilized approach criteria. The rest of the flight was uneventful.I believe that working together as a crew; we handled everything correctly and within required navigational standards. I am submitting this report mainly for tracking purposes. This is the second time I have had to deal with a degraded FMS on an RNAV arrival within the last 30 days. The first time; the FMC on the flying pilot's side failed; the autopilot disconnected; and went into CWS Pitch and Roll while descending into [a different airport]. That was also a very challenging scenario that I wish I had filed a report for but I did not. Regardless; even a momentary loss of flight path data on an RNAV arrival and has very significant implications for the safe conduct of the flight and it might be an excellent training scenario. It's even more critical if you are planning to use an RNP approach and then suddenly can't. My final concern is that as more and more airports are transitioning to RNP approaches (in particular in the Caribbean); even a minor FMC failure might mean that you can't land at your intended destination. I'm not sure that the company's current fuel policy addresses this issue and can envision a scenario where a crew finds itself challenged with few options and not much gas. Regardless; two failures that degrade my ability to fly an RNAV arrival and/or shoot an RNP Approach in less than 30 days has gotten my attention and I hope it gets yours.

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.