Sriwijaya Air Prelim Highlights Thrust Imbalance


Sriwijaya Air Flight 182 experienced a thrust imbalance prior to going down in the Java Sea on Jan. 9, 2021, according to the preliminary accident report issued by Indonesia’s Komite Nasional Keselamatan Transportasi (KNKT/National Transportation Safety Committee) on Wednesday. Data from the Boeing 737-500’s flight data recorder (FDR) showed that the thrust lever for the aircraft’s left engine started reducing while the thrust lever position of the right engine remained the same as the aircraft climbed through 8,150 feet shortly after takeoff from Jakarta’s Soekarno–Hatta International Airport (CGK). The trend continued as the aircraft reached 10,900 feet, at which point the autopilot disengaged and the aircraft rolled left to more than 45 degrees. When the autothrottle disengaged five seconds later, the 737 was at a 10-degree nose-down pitch angle.

The aircraft crashed at 2:40 p.m. local time, approximately four minutes after takeoff. No emergency communications were received. All 62 people onboard were killed.

Maintenance logs show that the aircraft’s autothrottle had been reported as “unserviceable” twice prior to the accident, once on Jan. 3 and once on Jan. 4. Work was done on the system and post-maintenance tests using the built-in test equipment (BITE) came back good each time. At the time of the accident, the captain had logged 17,904 flight hours with 9,023 hours on the 737. The copilot had 5,107 flight hours with 4,957 hours on type.

The search for the cockpit voice recorder is ongoing, as is the overall investigation. KNKT says it intends to focus its investigation on areas including understanding the cause of the split thrust levers, reviewing the history of the autothrottle system serviceability and maintenance records, reviewing pilot performance and training on upset prevention and recovery, and looking at human factors and organizational issues. The U.S. National Transportation Safety Board (NTSB) and the Transport Safety Investigation Bureau (TSIB) of Singapore are participating in the investigation.

KNKT’s complete preliminary aircraft accident investigation report can be viewed at

Kate O'Connor
Kate O’Connor works as AVweb's Editor-in-Chief. She is a private pilot, certificated aircraft dispatcher, and graduate of Embry-Riddle Aeronautical University.

Other AVwebflash Articles


  1. Seriously? The flight crew didn’t notice an UNCOMMANDED movement of a thrust lever?

    Did anybody else notice that all but 150 hours of the first officer’s flight time was listed as “in type?” So, prior to getting a 210-hour ICAO Limited Commercial certificate, the FO already acquired 60 hours in a B-737? Really? Or did this ace start flying right-seat in commercial service as a 150-hour private pilot?

    Call me a bigot, but – short of an anticipated spectacular exhibit of BDS (Boeing Derangement Syndrome) – I suspect that this is another vivid and tragic example of the systemic shortcomings of foreign aircrew.

  2. Reading the KNKT preliminary report’s Findings and Safety Actions I’m disturbed by the proportion of attention given to bureaucratic procedural recommendations versus basic monitoring and airmanship recommendations. In my unscientific estimation that proportion is about 4 to 1. Upset recovery training and ability is also emphasized much more than the basics of upset prevention in the first place. In the report pilots are counseled to practice awareness but emphasis on the basics of awareness and monitoring are terribly lacking proportional to what appears to have been their potential contribution to this accident as described by the preliminary report.

    No system should ever be left unmonitored, but mid ’90s vintage auto throttles were and remain at best so clunky and so behind the fidelity curve in real time that of all systems these should really not be left unmonitored. While the choice to hand fly is an operational choice and not a procedural mandate, consistently doing so all the way to altitude has been invaluable to me in maintaining situational awareness and monitoring. In addition to all the visual and oral cues present in the cockpit, hand flying all the way up to final level-off, even if that is in the thin 40’s, provides valuable tactile information about airplane status, not to mention maintenance of hand flying skills.

    This leads me to the subject of inherent aptitude and societal differences. During my very first pre-flight introduction as a CFI I mentioned to my student that yoke function in an airplane does not directly resemble steering wheel function in a car. My student’s response was “confusing the two won’t be a problem for me. I’ve never driven a car.” Later as I twisted open the winged cam locks on the oil dipstick access hatch he asked me “how did you know which direction, left or right, to turn the screws?”.

    For those of us who have grown up in the mechanized industrial world this kind of stuff is not learned. It is sub-consciously acquired almost from birth and therefore virtually intuitive. For people who have grown up not having tinkered with Dad’s lawn mower, the family car, Grandpa’s tractor, this kind of thing must be learned. My first ever student grew up in such a society and more importantly in a family who could afford to and did hire out anything that required the use and maintenance of machinery.

    In my humble opinion this is not about superiority versus inferiority. It is about a kind of primal life existence lending itself better to some activities than others. Operating aircraft is one of those activities. I do not begrudge any country or society the privilege of nationally operated air travel. But I do believe it is incumbent on the world’s societies to recognize and mitigate the risks inherent in flying airplanes without this phenomenon of virtual inherent aptitude which is acquired by practically having been born operating mechanical things.

    Thankfully modern aircraft are mostly forgiving enough without even more of these ghastly accidents having taken place, but the Boeings and Airbuses of the world can only do so much. Somehow the slack needs to be picked up through more quantity and higher quality of training, experience, discipline and professionalism. Instead, we see lower levels of training and experience in the operations of these airlines. I’m not sure that this is universally recognized. KNKT’s preliminary report disappointingly majoring on bureaucratic procedures mitigation rather than on the airmanship basics of monitoring and prevention makes me wonder all that much more if it is universally recognized.

    • > This leads me to the subject of inherent aptitude and societal differences.

      It is something to be conscious of, and to train for.

      One theory about Japan’s failure to rebuild their naval air force in WW2 after Midway was that the pilot candidate population lived in an early agrarian country. They were at a disadvantage to the millions of American youths at the time, who mostly had high school education and came from farms with tractors and other equipment. As a result, the US had unlimited suitable pilot candidates and Japan ran out. Indonesia is also 95% rice farming today, similar to Japan at the time.

      Chuck Yeager was the pinnacle – he worked for his dad on oil-pumping equipment and was a GI airplane mechanic, so had a significant advantage when operating the complex X-1 rocket systems – and survived where half of his colleagues didn’t.

      (My theory is that the 29 or so pilots shot down at Pearl Harbor was a much greater setback for Japan than realized, the 50-100 at Coral Sea a disaster, and the 200 or so at Midway insurmountable. The induction classes were only around 100 pilot candidates each, with 10-15 washing out from martial discipline. So each of those battles wiped out whole graduating classes. By contrast, the entire US-Canada border is littered with training air bases left over from WW2.)

      Regarding Indonesia specifically, they have a long history of problems with airline maintenance and parts, and returning airplanes to service that crash a few days later. Maybe we need to look at why that is. Their regulator is pretty aggressive, but something’s still slipping by. They will shut down airlines that can’t operate in a professional manner, as shown by them closing half their low-cost operators a while ago, so it’s not directly an issue of corruption or inability. Yet here we are.

  3. You hit the nail on the head John. The older I get the more I see this within our own country. Fortunately for me (by choice) I am not retired. I say fortunately because work keeps me engaged not only with my profession, (carpentry contractor) but, with the rest of the world and what is going on around me. The “primal life experience” you describe is the base line for learning. Without it nothing makes any sense, or, is nothing but distorted reality. You can’t read about throwing a ball and know what it really is i.e. (physics) without physically throwing it first. You can’t read about weather and fully understand it from the comfort of a classroom unless you first live in it, feel it, know what it is to really fear it. Understanding flight is a whole lot easier after you have spent hours, days, years watching seagulls and other birds fly in various weather conditions, or, maneuver uplifting on bluffs. Then and only then can you read about it in a book, computer and better understand the real world. I cringe when I see my grand kids computer try to explain something the author has never experienced. It is such a perversion of reality.

    This is why I spend every chance I get with my grandson outside bike riding, hiking, sailing, fishing, sled riding etc. For Christmas I got him a HD electric balance bike and a small set of tools appropriate for maintaining his bike. Guess what we’re going to be doing. There will be plenty of time for books and computers. Right now he needs to touch things, feel things, experience sensations only possible by engaging in real world activities. Not in some fantasy world drummed up on a computer. Books and computers will be and have been introduced as they become appropriate. Everything in its proper sequence.

  4. Although it’s too early to determine cockpit actions/reactions leading up to this crash, I’ve noticed over the years an increasing tendency to grab the QRH instead of grabbing the controls, when the airplane enters the “What’s it doing?” mode. Recovery of the CVR will go a long way in determining the situation in the cockpit; however, it may never be recovered. I’m with YARS in that I can’t explain how the flight crew failed to recognize an asymmetrical thrust condition; i.e., physical positioning of the thrust levers, engine instruments etc.

  5. In many developing countries where I have traveled on business, there is a tendency to teach “just follow the rules” and things will be okay. Unfortunately the rules are often developed by government officials, many of whom know little about the subject at hand. They are usually individuals from the higher society levels that got their appointments through whom they knew rather than what they knew. As John and Tom said, they may be well educated people, but sadly lacking in actual experience or ever having performed the tasks for which they are now writing rules. Judging from the fact that the aircraft in question had been written up on previous flights for the autothrottle malfunction, it would appear that the maintenance personnel were as limited in their experience as was the flight crew. It’s kind of like reading those wonderful “In Case Of Difficulty” sections in our appliance owner’s guides. “If it is doing this, then check that”. When your symptoms don’t match any of the “doing this” categories, then what do you do? Admitting you don’t know what to do is not the preferred option. The end result is that airlines are operated by pilots and mechanics who lack the background to fully understand what to do when things go wrong. They aren’t ignorant or uncaring people, they just have not gained a high level of knowledge or experience to fill their positions as we would like. And, passing more rules (the government’s preferred approach) isn’t going to solve the problem.

  6. I follow Indonesian airline accidents pretty closely.

    So far it sounds like the typical Indo accident chain that starts with a Mx issue. A similar one was the 2007 AdamAir 737 loss en route over the ocean, related to recurring IRS nav equipment issues. That ended up closing the airline (I know the owner.) And the 2018 Lion Air 737 MAX accident that started with a bad sensor.

    Based on what I’ve read over the years, I think Indonesia needs to:

    – each airline should have a committee of Mx, pilots and mgmt. to review significant or recurring Mx issues before return to flight. That would have caught both the AdamAir and Sriwijaya accidents beforehand.

    – line pilots need to know about Mx issues, and review the related emergency procedures.

    – there has to be a mechanism for pilots to ground the plane, whether that’s a senior pilot or line pilot

    – CRM retraining for coordination during problems. In both the Sriwijaya and AdamAir accidents, likely both pilots were trouble-shooting, with nobody actually flying the plane. Because of the challenging weather conditions there, esp. haze and mountains, somebody always needs to be in alert flying mode, regardless of autopilot.

    Indonesia is a challenging country to fly in because it’s at the end of the earth, so longer to get parts (forcing parts cannibalization), has torrential rains and daily haze, is mountainous, and English is not the native language. I think they need to look at operating procedures that go beyond “business as usual” that work in more forgiving envionments, but just doesn’t work in Indonesia.

    Also, as found out in the AdamAir accident, the SAR equipment and funding is not available to do timely air crash investigations, so it’s doubly-important to prevent them. (The black box batteries expired before a recovery within a month could even be organized.)