https://avherald.com/h?article=4d97ca46&opt=0 The Aviation Herald Last Update: Saturday, Sep 4th 2021 20:09Z 27108 Articles available Events from Mar 23rd 1994 to Sep 4th 2021 www.avherald.com Incidents and News in Aviation Next Sort list List Earlier Next Later List by: by currently Filter: Crashes Accidents Incidents News Reports [ ] search Article Article Occurrence sorted by On On On On On date Update --------------------------------------------------------------------- [INS::INS] [INS::INS] The Aviation Herald Apps Android and iOS Incident: China Airlines A333 at Taipei on Jun 14th 2020, all primary computers, reversers and autobrakes failed on touchdown AVHAPP on Android and iOS Support The Aviation Herald By Simon Hradecky, created Friday, Jul 3rd 2020 14:39Z, last updated Friday, Sep 3rd 2021 14:01Z one time Support The Aviation Herald A China Airlines Airbus A330-300, registration B-18302 performing flight CI-202 from Shanghai Pudong (China) to Taipei Songshan Monthly support (Taiwan) with 87 passengers and 11 crew, landed on Songshan's wet 1 EUR/month runway 10, when upon touchdown all three primary flight computers, Monthly Support The Aviation Herald thrust reversers and autobrake systems failed affecting the stopping distance of the aircraft. The crew applied maximum manual braking and Interview: managed to stop the aircraft 10 meters/33 feet ahead of the runway The human factor named "Simon end (runway length 2600 meters/8530 feet). The aircraft was towed to Hradecky" and the team of man and the apron. machine The aircraft remained on the ground until Jun 23rd 2020, then positioned to Taipei's Taoyuan International Airport and is on the ground there since. Taiwan's CAA reported the root cause of the occurrence is still under investigation and advises all A330 operators to take countermeasures: - prior to dispatch consider possible deceleration deficiencies with the conditions mentioned above on wet runways. - required landing distance shall be predetermined for a wet runway, if the distance is a concern consider an alternate aerodrome - operators should enhance crew awareness of wet runway operations, if automatic braking is out of function promptly change to the alternate system or apply manual braking On Sep 3rd 2021 Taiwan's ASC released their final report in Chinese and their English Executive Summary concluding the probable causes of the incident were: Findings related to probable causes 1. The three flight control primary computers (FCPCs) of the occurrence aircraft became inoperative almost at the same time during touchdown. The root cause was determined to be an undue triggering of the rudder order COM/MON monitoring concomitantly in the 3 FCPC. At the time of the aircraft lateral control flight law switching to lateral ground law at touch down, the combination of a high COM/MON channels asynchronism and the pilot pedal inputs resulted in the rudder order difference between the two channels to exceed the monitoring threshold. The FCPC1 failed first. 2. After the FCPC1 failure, the master control of flight control system was handed over to FCPC2 and FCPC3 in sequence whose asynchronism were also high at that moment; thus eventually all three FCPCs became inoperative. As a consequence of the three FCPCs loss, the thrust reversers, the ground spoilers, and the autobrake system were lost, resulting in an increased landing distance for the aircraft. Findings related to risk 1. During landing, flight controls reconfigured from normal law to direct law after all three flight control primary computers (FCPCs) became inoperative. While all aircraft primary control surfaces were still controllable, the deceleration devices including ground spoilers, thrust reversers, and autobrake were lost, the deceleration of aircraft was relied on manual brake by the pilots. 2. Given all three flight control primary computers (FCPCs) failed seconds after touchdown, should other factors (long flare, runway state, !) have affected the landing distance, the aircraft could have overrun the runway even if the pilots had immediately applied maximum manual brake after realizing the autobrake had failed. Other findings 1. The occurrence flight crew were properly certificated and qualified in accordance with the requirements of the Civil Aviation Authority of Taiwan. Records of training and checks have no anomaly related to this occurrence operation. The rest and activities of flight crew 72 hours before the occurrence were normal. No evidence indicated any pre-existing medical conditions or alcohol that might have adversely affected the flight crew!-s performance during the occurrence flight. 2. During the approach, flare, landing, and roll out until aircraft came to a full stop, the actions performed by the flight crew complied with stable approach and manual landing Standard Operation Procedures (SOP) prescribed in Flight Crew Operating Manual (FCOM). 3. During the landing roll, the crew kept good interaction and high situation awareness based on pilot-flying!-s response to decelerating the aircraft and pilot-monitoring's call out of relevant abnormal system status. 4. With three FCPCs inoperative, actual remaining runway distance (30 feet margin) of the occurrence flight was shorter than the calculated value (172 feet margin), possibly due to tailwinds, runway conditions, and manual braking as these factors might increase the braking distance. 5. Ground spoilers function requires at least one functional FCPC, arming autobrake requires at least two functional FCPCs, deployment of thrust reversers require unlock signal from either FCPC1 or FCPC3. As a consequence of the three FCPCs loss, the non-release of the independent locking system prevented the reversers!- deployment, the ground spoilers were cancelled and autobrake system was lost. 6. Shop finding of FCPC1 indicated that the unit is no fault found (NFF). The built-in test (BITE) shows SAOPS"Sp"|cification Assist"|e par OrdinateurPS(c)fault at the time of the triple FCPC fault. The SAO fault corresponds to the fault was trigged during COM/MON monitoring rather than the fault of computer hardware. 7. Following the occurrence, Airbus reviewed its in-service experience, and confirmed that no other triple PRIM fault at touchdown event had been reported on A330/A340 aircraft family since entry into service. The A330/A340 fleet fitted with electrical rudder has accumulated 8.7 millions of Flight Cycles and 44.3 millions of Flight Hours (in-service data from April 2020). 8. The runway surface friction, longitudinal slope, transverse slope, and longitudinal slope changes of the Songshan Airport runway 10 complied with relevant standards. 9. The deceleration performance of the occurrence flight between 6,600 feet and 7,300 feet from the threshold of runway 10 deteriorated. It may be due to paint marking and rubber deposit on the touchdown zone of runway 28. 10. The occurrence flight first touchdown and second touchdown were about 1,500 feet and 1,800 feet respectively with respect to the runway threshold. The touchdown points were both located at runway touchdown zone. 11. After the flight crew applied manual braking, the overall deceleration performance was between !degmedium!+- and !deggood!+- level consistent with the reported wet condition of the runway, which should be able to rule out the effect of hydroplaning. The ASC summarized the sequence of events: The aircraft took off at 1625 with the captain as the pilot flying (PF) and the co-pilot as the pilot monitoring (PM). During descent, the flight crew received the information L from automatic terminal information system (ATIS), and was instructed to use the instrument landing system (ILS) of runway 10. After calculating the landing performance, the Electronic Flight Bag (EFB) showed that there would be 362 feet runway distance remaining if autobrake was set at low. During the descent and approach phases, no abnormalities were found. About six minutes before landing, the PF was alerted about the rain near Songshan Airport, he asked the PM to review the latest weather information, and learned that the tail wind blew from 280 degrees at 6 knots with a light thunder shower rain. Runway 10 was in use. At 1743:21, during the final approach, the airplane was at the barometric altitude of about 3,008 feet, the flight crew was informed by the tower that the visibility at the airport had dropped to 2,500 meters. At 1743:51, the airplane was at the barometric altitude of about 2,480 feet, the tower issued a landing permission: "dynasty 202 runway 10 wind 250 degrees 9 knots caution tail winds clear to land." About fifteen seconds later, the PF reminded the PM, "call out when spoilers deploy so I can tell if the main gear has touched down." At 1744:37, the airplane was at the barometric altitude of about 1,832 feet, the flight crew conducted landing checklist and the PF asked the PM to set the autobrake from low to medium due to the weather change at Songshan Airport. At 1745:41, the airplane was at the radio altitude of about 996 feet, the tower reminded that wind speed was 10 knots. Six seconds later, the radio altitude was 919 feet, and the PF said "the wipers can be faster, it's okay", the PM answered "it!-s at the fastest speed already". At 1745:58, the airplane was at the radio altitude of about 773 feet, the PM reported !degapproach lights ahead!+- and the PF then disengaged the autopilot to continue the approach. At 1746:41, the airplane was at the radio altitude of about 136 feet, the PM reminded the PF !degcenter line!+- to maintain on track. Seven seconds after the aircraft passed the radio altitude of 60 feet, at 1746:54, the aircraft touched down at between 1,500 and 2,000 feet from runway 10 threshold with pitch up about 4.2 degrees, roll to the right at about 1.1 degrees, and its magnetic heading at about 94 degrees. The ground speed was 147 knots (indicated airspeed was 135.5 knots) and the maximum vertical acceleration was 1.28g's. The slats/flaps configuration was FULL. Ground spoilers (i.e. spoilers 2 to 6) started to deploy. One second after the touchdown, the PM immediately called out !degspoilers!+-, while the left and right main gear shifted between air mode and ground mode for about 0.75 seconds and 0.5 seconds respectively. Three seconds after the main gear touched down, autobrake system fault was recorded on FDR. One second later, PRIM1/ PRIM2/PRIM3 faults were recorded at the same time and the spoilers retracted, as the ground spoiler function was lost. The PM called out !degreverse!+- and the nose gear touched ground at the same time. After that, the nose gear flipped between air mode and ground mode for nearly 7 seconds. At 1746:59, the PF asked twice if !degautobrake is on!+- (the ground speed was 141 knots). The PM answered, !degautobrake is not on!+-. About five seconds later (1747:04), the PF called out !degmanual brake!+-, and applied full brake pedal. The normal brake hydraulic pressure value was 448 psi and longitudinal acceleration rate value was about -0.1g, indicating a deceleration. At 1747:07, the PM called out "reverse no green", about one second later, the PF requested to his first officer !degquickly help me brake help me brake!+- (the ground speed was 127 knots at this time), from then on, both pilots applied full pedals on the brakes manually, the normal brake pressure was up to 576 psi and the longitudinal acceleration rate was -0.14g 's. Until 1747:36 when the aircraft came to a full stop at 30 ft before the end of the runway, the aircraft brake pressure and longitudinal acceleration rate fluctuated from 128 psi to 2,560 psi and from -0.05g's to -0.47g's respectively during this period. Metars: RCSS 141100Z VRB02KT 8000 -SHRA FEW006 FEW016CB SCT025 BKN045 26/26 Q1013 RETSRA NOSIG RMK CB S A2992 RA AMT 17.6MM= RCSS 141044Z VRB02KT 6000 -SHRA FEW008 FEW016CB BKN025 BKN045 26/26 Q1012 RETSRA NOSIG RMK CB S A2991= RCSS 141036Z VRB02KT 3000 TSRA FEW008 FEW010CB BKN016 BKN035 26/26 Q1012 BECMG 5000 NSW RMK TS S AND N MOV E A2991= RCSS 141030Z VRB03KT 1200 R10/1300U +TSRA FEW006 FEW008CB BKN014 BKN025 26/26 Q1012 TEMPO 3000 RMK TS OVHD A2991= RCSS 141009Z 23006KT 200V270 0800 R10/1500D +TSRA FEW004 FEW008CB BKN010 OVC025 26/26 Q1013 TEMPO 3000 RMK TS OVHD A2992= RCSS 141000Z 25008G19KT 200V280 1000 R10/1300D +TSRA FEW006 SCT010CB BKN012 BKN025 27/26 Q1012 TEMPO 3000 RMK TS OVHD A2991 RA AMT 18.2MM= RCSS 140949Z 25008G19KT 210V270 1000 R10/1500D TSRA FEW006 FEW010CB BKN012 BKN030 27/26 Q1012 TEMPO 3000 RMK TS OVHD A2990= RCSS 140945Z 26008KT 220V280 1200 R10/1800D -TSRA FEW006 FEW012CB BKN014 BKN030 28/27 Q1012 TEMPO 3000 RMK TS S STNR A2989= RCSS 140941Z 26007KT 2500 -TSRA FEW008 FEW014CB BKN016 BKN040 28/27 Q1012 BECMG 1500 TSRA RMK TS S STNR A2989= RCSS 140937Z 28006KT 250V320 4000 -TSRA FEW008 FEW016CB BKN018 BKN040 28/27 Q1012 BECMG 3000 TSRA RMK TS S STNR A2989= RCSS 140930Z 29005KT 260V320 7000 -TSRA FEW008 FEW016CB BKN020 BKN045 29/27 Q1012 BECMG 3000 TSRA RMK TS S STNR A2989= RCSS 140907Z 26004KT 230V290 7000 -TSRA FEW012 FEW016CB BKN025 BKN045 29/27 Q1011 BECMG 3000 TSRA RMK TS OVHD A2988= RCSS 140900Z 28003KT 260V320 7000 -TSRA FEW014 FEW018CB BKN025 BKN060 29/27 Q1011 BECMG 3000 TSRA RMK TS SW MOV N A2987 RA AMT 2.2MM= RCSS 140857Z 28003KT 260V320 7000 -TSRA FEW014 FEW018CB BKN025 BKN060 29/27 Q1011 BECMG 3000 TSRA RMK TS SW MOV N A2987= RCSS 140839Z VRB03KT 6000 VCTS -SHRA FEW014 FEW018CB BKN020 BKN050 29 /27 Q1011 BECMG 3000 TSRA RMK TS S MOV NE A2986= RCSS 140830Z 15003KT 090V210 7000 -SHRA FEW014 FEW018CB BKN025 BKN050 29/27 Q1011 RETS RESHRA NOSIG RMK CB S-SW A2987= RCSS 140812Z VRB02KT 6000 SHRA FEW014 FEW016CB BKN020 BKN035 29/27 Q1011 RETS NOSIG RMK CB S A2986= RCSS 140800Z 28004KT 250V320 7000 VCTS -SHRA FEW014 FEW018CB BKN025 BKN040 30/27 Q1011 NOSIG RMK TS SE MOV NE A2986 RA AMT 0.2MM= RCSS 140755Z 29004KT 260V320 7000 VCTS -SHRA FEW016 FEW020CB BKN025 BKN040 30/27 Q1011 NOSIG RMK TS SE MOV NE A2986= RCSS 140747Z 30005KT 270V330 8000 VCTS FEW018 FEW020CB BKN028 BKN040 31/27 Q1011 NOSIG RMK TS SE MOV NE A2986= RCSS 140730Z 30004KT 270V340 8000 VCSH FEW018 BKN030 BKN050 30/27 Q1011 NOSIG RMK A2986= RCSS 140721Z 30005KT 270V330 8000 VCSH FEW018 BKN030 BKN045 30/27 Q1011 NOSIG RMK A2986= RCSS 140700Z 29004KT 240V330 7000 -SHRA FEW018 BKN030 BKN045 29/27 Q1011 NOSIG RMK A2986 RA AMT 4.6MM= --------------------------------------------------------------------- Reader Comments: (the comments posted below do not reflect the view of The Aviation Herald but represent the view of the various posters) --------------------------------------------------------------------- Aircraft By baba budangiri on Sunday, Sep 5th 2021 19:21Z Not really familiar with the airport, but I understand it is constrained by neighbor community (extremely close to Taipei city) and terrain. So the landing direction might be influenced by this, especially no one knows they are going to lose thrust reverser upon touchdown. --------------------------------------------------------------------- @ ocelot By Lee on Saturday, Sep 4th 2021 17:24Z Yep, bang on. Interesting specific statement from AB: "7. Following the occurrence, Airbus reviewed its in-service experience, and confirmed that no other triple PRIM fault ***at touchdown event*** had been reported on A330/A340 aircraft family since entry into service. The A330/A340 fleet fitted with electrical rudder has accumulated 8.7 millions of Flight Cycles and 44.3 millions of Flight Hours" How many triple PRIM faults have taken place overall? Way to overlook the obvious question, AB. Triple PRIM faults is something that should never even be a possibility, esp in the air. --------------------------------------------------------------------- software again By ocelot on Friday, Sep 3rd 2021 19:38Z This is an excellent example of what I was talking about regarding software the other day -- a combination of circumstances created a special case that nobody had thought about which turned out to not work as desired. Also I notice that Captain Crunch wrote a year ago about a different event: "The monitoring algorithm was comparing the timing of the orders (COM vs MON) ..." and "I wonder if the same kind of logic may reside on A330?" ...apparently so, it seems :-( --------------------------------------------------------------------- @Captain Crunch - operational / investigative context has not advanced much since previous generation of (non-FBW) a/c? By (anonymous) on Sunday, Jul 12th 2020 09:33Z What I mean is that we still depend mainly on FDR / CVR for avionics incident analysis (plus physical print-outs upon landing, it seems). Taking ACARs as an example, it seems that a lot of real-time information is captured, transmitted and analysed by engine manufacturers. Shouldn't this apply equally also to avionics? e.g. the Germanwings disaster where it was discovered that the autopilot went into two modes simultaneously, when this should not have been possible. And also the strange wandering of altitude target settings (across multiple flights), compounded by inability to differentiate between autopilot and FCU manipulation of altitude target. Without the loss of the a/c, none of this would have been discovered. Yet engine health / efficiency is monitored in real time? Surely this has to change.. --------------------------------------------------------------------- @Captain Crunch By (anonymous) on Sunday, Jul 12th 2020 09:24Z Thanks again for pointing the MYX-9001 event out - amazing for myself to learn of this one after the PIA accident having occurred in terms of engines contacting the runway and then failing as a result. It sounds like they were quite fortunate to make the circuit and survive. Not only that, but their horz stab trim position just happened to allow the aircraft to lift off the runway with pilot inputs not having any effect, both during the first attempt to lift off, and again after having sunk back down onto engines (and power re-applied) - unbelievable. Thanks a lot for the breakdown on the WOW sensors and effect on ELACs and FCCs. I think I can recall reading other reports here where not having spoilers armed led to computer problems, once again re WOW sensors, but not as severe as this. It seems unfathomable that a design flaw of this severity could exist in modern FBW systems. I definitely share your concern about whether perhaps the same logic is present in the A330.. --------------------------------------------------------------------- @ Anonomous Network talk, part 2 By Captain Crunch on Sunday, Jul 12th 2020 00:47Z cnt'd from part 1: I think the FCC measures the time of the bounce airborne, based on when the data packet arrives at the FCC in the avionics bay. The WOW= Compressed-packet must have arrived late at SEC-COM because it claimed the math produced a greater than one-second bounce. (Maybe only arriving a nanosecond late, but that changed the value.) The monitoring algorithm was comparing the timing of the orders (COM vs MON) when it always concluded the elevator was a runaway. Flight Orders were more aggressive than Ground Orders in control-surface-travel-movement-rate (according to the wave amplitude.) But the four comparator boards didn't know this, and dumbly decided that the elevator was a runaway. This exact scenario happened to ELAC-2, SEC-1, and SEC-2, leaving dead elevators. As the very-good accident investigators said: It was a design flaw. Unreliable Redundancy? I wonder if the same kind of logic may reside on A330? --------------------------------------------------------------------- @ anonymous regarding Network packets, Part 1 By Captain Crunch on Sunday, Jul 12th 2020 00:44Z Great posts. With the MYX-9001 hull-loss, they didn't arm the Ground Spoilers and bounced. There were actually two separate malfunctions. The mechanical PTA micro-switch (Pitch Trim Actuator) "flapping" you speak of caused ELAC-1 to fault earlier, that's true. But the WOW (weight-on-wheels) sensor sent messages to the FCC's, indicating left strut compression or extension. Identically flawed FCC's came up with different air/ground answers. The SEC-COM (FCC in command) determined that the WOW sensor packets said extension for more than 1 second = Flight Orders. The SEC-MON (monitoring FCC) determined that the very same WOW sensor packets said extension for less than 1 second = Ground Orders. Issuing two different orders (air/grd) caused HAL-9000 to lose his mind and kill all FCCs. cnt'd next post --------------------------------------------------------------------- @Captain Crunch - I didn't know about MYX-9001, I've just read thanks - quite remarkable. By (anonymous) on Thursday, Jul 9th 2020 03:58Z Just to cover off though - actually QF72 was caused by a faulting ADIRU mis-labelling packet data being placed into the ARINC bus ahead of the relevant prim. The ARINC (network) bus itself was doing it's job properly, moving the data between systems. Both channels within the affected prim were getting the same data, and would have produced the same outputs, so there was nothing different for the prim comparison logic to detect. The timing aspect of MYX-9001 sounds like it was more related to 'flapping' microswitch outputs, on the mechanical level, and therefore not actually an ARINC (network) problem either. So I have to say I don't think that there is a common pattern going on between these incidents. All the same, it's the system as a whole that has to function correctly, and yes at that level you could say that 'FBW' is implicated for both. In both cases, as for most complex software systems, the coverage, handling and reporting of fault scenarios could be improved. --------------------------------------------------------------------- @Captain Crunch By (anonymous) on Thursday, Jul 9th 2020 03:52Z Thanks for your response - some good points. Network sequencing (jitter), packets lost/dropped and latency are all well understood, so I'd expect ARINC and broader A330 avionics suite are designed accordingly. Having said that, when building complex networked systems from multiple manufacturers there is always an expectation for surprises. When testing multi-layered complex systems, a decision has to be made about how far is good enough. There is an interesting document floating around somewhere describing this problem with respect to development of the Space Shuttle avionics platform. Then there are hardware design problems/faults/environmental problems (as you've mentioned). All of which as we know adds up to the need for redundancy across multiple technology implementations, multiple system paths, and filtering/voting out of faulting systems (which airbus FBW has got). Hence my focus on the filtering/voting mechanism as a critical firewall against the above.. --------------------------------------------------------------------- @ Anon: "we don't understand why this system/component failed in this way" PART 2 By Captain Crunch on Thursday, Jul 9th 2020 00:08Z Without a degree in CS, most readers cannot appreciate that THE TIMING of data transmission packets is everything. If ANYTHING disturbs the timing of the data packet, the comparison logic can get fooled. This is one reason why I abhor Fly-by-wire (FBW.) When new, the FBW system may produce better safety with pilots who can't hand-fly in Direct Law. But during data-corruption: It's Moments-of-Sheer-Terror unless it reboots. It can take decades to fully test newer robust software loads. So probability models are used instead. That's why MYX-9001 A320 "One-Second-Bounce-Bugs" can hide in 100,000 lines of code and show up years later like on the Smartlynks MYX-9001 dead-elevators accident. We are assuming here in this incident, that the SEC's did not fall off the line also. Thanks to Bus drivers for explaining their systems. All my posts are JMHO's. --------------------------------------------------------------------- @ Anon QF72 : "we don't understand why this system/component failed in this way" PART 1 By Captain Crunch on Thursday, Jul 9th 2020 00:05Z Extremely rare, but Flt Soft fail happens. Good obserb that accdnt invstgtrs have great trouble ident why elevators failed. Here's my humble opinion: Bus share one wire for many components to save weight. On CA & Thai accdnts, one theory was that after the uncommanded go-around, everyone made a call on their cellphones that RF-ed the elevators out of control. Theory was dismissed. But seem's like a pattern in bus FBW design. In QF72 and MYX-9001 accidents, erroneous input data (bad AOA spikes and conflicting Air/ Grd orders, respectively) both caused loss of elev cntrl and injuries. The former was because comparison algorithms did not filter out AOA spikes and the later was because THE TIMING of conflicting air and grd orders fooled all four FCC's into thinking all four elev actuators were runaways. In that case, all ELAC primary and SEC backup FCC's dropped off the line since they couldn't resolve conflicting air/grd order phase imbalance. --------------------------------------------------------------------- @ By Joe on Friday, By Great Flying results in another teeth clencher..... on Wednesday, Jul 8th 2020 15:18Z Tower says use RWY 10, we will follow orders and do exactly that. Who cares if there is a gusting tailwind, exceeding limits on a slippery runway, with microburst potential, and expecting autobrakes, autospeedbrakes, autothrottles, etc all doing the job of stopping the aircraft. Only when the hydroplaning, skidding and end of runway looms does the crew finally step in and apply manual pedal braking, maybe max reverse and oh, manual deployment of speedbrakes. A perfect end to a crappy day! Pilots are heroes, ATC gasping with relief, Airbus gets black eye and business as usual. Welcome to China aviation. --------------------------------------------------------------------- @Also busdriver - immense respect for you as a pilot & I know that the A330 avionics engineering is actually very good By (anonymous) on Monday, Jul 6th 2020 06:12Z I'm only commenting regarding the outcome of at least two relevant investigations that have essentially concluded with statements such as "we don't understand why this system/component failed in this way". In the case of QF72, where one prim went off the rails but the other two were good, I can understand why both channels in the faulting prim would have produced identical results, and the comparison logic wouldn't have seen any difference. But why didn't the filtering logic across all fcc's not detect the rogue prim much faster and vote it offline? Shouldn't that have happened at the beginning of the first (unneeded) stall recovery attempt? Wasn't the problem really that a faulting prim was being held online? Unless there was a flaw in the fcc filter specs, even if you could get access to them, how would they help? Without the source code (very tightly controlled?), how could you anticipate or interpret that kind of filter behaviour, regardless of having flown an A330 or not? --------------------------------------------------------------------- @Also busdriver By (anonymous) on Monday, Jul 6th 2020 05:28Z Not only do we (non-pilots) not understand how they work, but neither do the authorities or the manufacturer(s) in the case of a few well-known A330 incidents/accidents. --------------------------------------------------------------------- CAA advises By Xaviation on Monday, Jul 6th 2020 04:56Z The CAA advises the operators, "if automatic braking is out of function promptly change to the alternate system,... ". Really....? The a/c changes automatically to ALTN braking when Normal braking is U/S. Not a pilot action. No auto-brakes on ALTN system, but indeed still anti-skid. Was something lost in translation..? --------------------------------------------------------------------- @Bandar By Laurence Doyle on Sunday, Jul 5th 2020 16:35Z Nothing to do with China, apart from the fact the flight originated in China. China Airlines is the flag carrier of Taiwan. --------------------------------------------------------------------- @BusDriver By Also busdriver on Sunday, Jul 5th 2020 08:40Z Busdriver you are correct, they would always have sidestick control with total prim loss. The A330 is remarkably safe regarding computers and flight controls. I ve seen many negative comments lately regarding computers and flight controls (Airbus), most of them I bet never flew A330/320 and don t understand how they work. --------------------------------------------------------------------- @BusDriver By (anonymous) on Sunday, Jul 5th 2020 08:37Z Thanks for the clarification - quite relieving and I'm glad to learn a bit more. After some further reading it seems that the primary and secondary fcc's are built using different CPUs and are programmed by different teams, so that is good to learn also. That's interesting re the THS, I imagine that pilots are trained to know about that and should have no problem to handle. --------------------------------------------------------------------- @anon 032305072020 By BusDriver on Sunday, Jul 5th 2020 05:57Z Not true, any SEC will provide complete control of all flight controls in direct law even with a total loss of all 3 PRIMS. They are independent systems. With all 3 PRIMS failed then you will lose electrical control of the THS, requiring manual input to the trim Crunch - The sidesticks will be fully operational in the event of a G /A, the aircraft will be in direct law and trimming the THS will have to be done manually on the trim wheel. The 330 has 5 flight control computers, any one of which allows full control. With all 3 PRIMS failed there can be no REV deployment, no autobrake, however manual breaking is available, with antiskid. --------------------------------------------------------------------- I'm totally gobsmacked ! By (anonymous) on Sunday, Jul 5th 2020 03:23Z Unless I'm wrong, without any prims, there is no way to control primary flight surfaces on an A330? Isn't (even) direct law really just a software mode within the prims? All three failed on touchdown - remarkable. Software bug? Network error? Power bus problem? The investigation on this will be interesting to say the least.. --------------------------------------------------------------------- Another Computer-Close-Call By Captain Crunch on Sunday, Jul 5th 2020 02:18Z I wonder if they bounced due to no ground spoilers deployed? Sounds to me like possibly an air-ground sensor malfunction. Stuck in Air mode would prevent ground spoilers deploying, inop the auto brakes and prevent reversers from functioning. Right? Could Air Order/Ground Order disagreements inop all three flight computers like they do in the A320 sometimes? Then THS pitch trim only? Please tell me I'm wrong! Some city airports in the Orient have no 500 ft stopway. I wonder about China Air op specs. Is a 19-knot tailwind GUST not controlling? Just how tall was this overhead Thunderstorm? This really sounds pretty bad to me. Am I to understand that a Go-Around in crap weather might have no flight computers, no side-sticks working? Gulmp. This is an airline my pilot group refused to be paxed on due to wild maneuvers all the time when everything was working. --------------------------------------------------------------------- why runway 10? By menil on Saturday, Jul 4th 2020 13:17Z the reason is (probably) that the wisibility was way below minima for runway 28, that has an LDA/RNP/VOR approaches with required visibility o 3,600 meters while the required visibility for ILS10 is 800/750 RVR. Before the CAA requires to calculate nonsense it is better to take ask the crew some tough questions, not on the malfunction, but on their choice to land on a soked short runway with a considerable tail wind. --------------------------------------------------------------------- @anonymous By airbusdriver on Saturday, Jul 4th 2020 12:09Z If I would reread my posts BEFORE posting, I would see the misprints: ... see that REVERSER deployment... sorry --------------------------------------------------------------------- @anonymous By airbusdriver on Saturday, Jul 4th 2020 12:04Z If you would do just a little research yourself, you would see that spoiler deployment requires its FADEC and PRIM1 or PRIM3. All 3 PRIMs were u/s... --------------------------------------------------------------------- By (anonymous) on Saturday, Jul 4th 2020 07:25Z Yeah but what would cause the thrust reverse to not engage? --------------------------------------------------------------------- Fortunate in unfortunate By James on Saturday, Jul 4th 2020 04:03Z Songshan Airport is middle in the city with residential and commercial buildings all around. Glad that the plane stopped within airport parameters or the consequences could be deadly --------------------------------------------------------------------- @anonymous By Kalle on Friday, Jul 3rd 2020 19:55Z The A330 isn't exactly like the A320. It has 3 PRIM (primary) and 2 SEC (meaning secondary on the A330) flight control computers. According to the FCOM, the PRIMs do - Normal, alternate and direct flight control laws - Speedbrake and ground spoiler control - Protection speec computation - Rudder travel limit --------------------------------------------------------------------- Landing runway By (anonymous) on Friday, Jul 3rd 2020 17:49Z Not really familiar with the airport, but I understand it is constrained by neighbor community (extremely close to Taipei city) and terrain. So the landing direction might be influenced by this, especially no one knows they are going to lose thrust reverser upon touchdown. PS : I'm not sure if 19kts gust was present on the metar received, they changed quite a few during the approach --------------------------------------------------------------------- @ (anonymous) on Friday, Jul 3rd 2020 16:54Z By Simon Hradecky on Friday, Jul 3rd 2020 17:37Z It is my understanding, that all three FCPCs (P1, P2, P3) failed. --------------------------------------------------------------------- By (anonymous) on Friday, Jul 3rd 2020 16:54Z all three primary flight computers What is being referenced here? FACs, SECs, ELACs? At this stage the report from Taiwan s CCA appears to be lacking in technical substance, making it challenging to understand what exactly happened. --------------------------------------------------------------------- Only the most recent 30 comments are shown to reduce server load. Click here to show the remaining comments --------------------------------------------------------------------- Add your comment: (max 1024 characters) Your IP address 68.37.28.109 is being tracked. We reserve the right to remove any comment for any reason. Profanities, personal attacks (against any physical or legal person), discussions about site policies, false statements or similiar as well as copyrighted materials from third parties will not be tolerated, those messages will summarily be deleted. Links and mail addresses are not permitted and will not appear in the display, posts trying to circumvent the restriction of links/mail addresses will be removed. We ask for your valid e-mail address in the email field. This address is only used by us to establish contact with you in case of further questions, it will not be displayed anywhere or be used otherwise. Your [ ] Name: Your [ ] Email: Subject: [ ] Your comment: [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [Post Comment][Return to Home] --------------------------------------------------------------------- Get the news right onto your desktop when they [subscribe] [login] [faq] [contact] [impressum] happen --------------------------------------------------------------------- (c) 2008-2021 by The Aviation Herald, all rights reserved, reprint and republishing prohibited. We use cookies to ensure you get the best experience on our site, learn more.