Air India 787 crashes on take off 2

[LittleInch]

A full 787-8 has crashed shortly after take off in ahmedabad.

Basically barely got off the ground then look like its trying to land in this video.

Air India flight to London Gatwick crashes in Ahmedabad with more than 240 people onboard – latest updates

Plane for flight AI171 was carrying 242 people including 169 Indians and 53 Britons, says Air India

www.theguardian.com

Specualtion that they pulled flaps up instead of gear up and basically didn't have enough lift so it looks like a gentle stall right into a built up area.

Looks to be flaps up, slats/ nose flaps down and gear down which is very odd.

 

[Garbage1]

So power goes down to CCS, and everything shuts down.

The Boeing 787 Dreamliner employs a revolutionary architecture called the Common Core System (CCS). This centralized system replaces the traditional approach of dedicated electronics for each aircraft function. Within the CCS, two critical components are the Common Computing Resource (CCR) cabinets.

• Each CCR cabinet houses eight GPMs. These modules are the workhorses of the CCS, acting as independent computing platforms.
• Examples of functions hosted by GPMs include:
  • Remote Power Distribution System (RPDS)
  • Generator/Bus Power Control Unit (GCU/BPCU)
  • Landing Gear Indication and Control
  • Thrust Management Function
  • Flight Management Function

Boeing 787 Common Core System (CCS) : A Technical Breakdown - Online Aviation Training

The Boeing 787 Dreamliner employs a revolutionary architecture called the Common Core System (CCS). This centralized system replaces the traditional approach of dedicated electronics for each aircraft function. Within the CCS, two critical components are the Common Computing Resource (CCR)...

oat.aero

 

[TugboatEng]

A diesel engine can set the pump to zero delivery even while the engine is running such as during coasting. Maybe this isn't such a good idea for a turbine engine. I wonder how minimum fuel is set vs no fuel.

Starving a pump for fuel isn't such a wise idea so I imagine the normal shutdown is performed with the FMU and not via supply valves.

We're having trouble with the marine engines where any fault with the engine control module power results in an immediate shutdown with no stored fault even.

Diaphragm pumps in automotive fuel systems typically have a pair of rubber flaps that serve as suction and discharge check valves. In the flat head Ford days these would have been nitrile rubber and short lived if exposed to the temperatures atop the engine. They would behave very similarly to a diaphragm pump pumping vapor.

 

[Garbage1]

After reviewing a service bulletin on the FMU, I agree all fuel flow metering is performed in the FMU, but ultimately it appears EEC and valve control fuel flow.

Training Doc on FMU. Have not read yet but looks interesting.

https://ia600108.us.archive.org/16/items/b787-meav-type-1-training-manual/ATA%2073%20Engine%20-%20Fuel%20and%20Control.pdf

 

[TugboatEng]

Right away, this part of the diagram shows a spring shut valve held open by a normally closed switch. A loss of electrical power would cause an engine shutdown if this is true. It also shows two pressure sources for a shut off?



This may not be the only way the FMU can cause a shutdown, either. If the metering valves can move to a zero position they can also cause a shutdown.

Also note that the engine has a permanent magnet alternator producing 28V DC to run the FMU. That might rule this out as a source of dual engine shutdown.

 

[LionelHutz]

I can't see the plane being built so a power loss kills the engines when the engines are built to continue functioning through most plane electrical system issues. Also, it's be dumb to build a plane that kills the engines between a power loss and the RAT deploying.

 

[Garbage1]

Back to this analysis, which would affect both engines at same time and cause equal dual engine loss of thrust.

In my mind possible that some sort of complete electrical failure caused the in-tank AC fuel pumps to quit. This would cause a RAT drop even before the engines spooled down completely and be triggered by loss of AC generation or loss of hydraulic pressure.

The engines have gear driven fuel pumps that would still work. But without tank pumps running, gear driven fuel pump may not have gone to full flow. Not vapor lock, technically, but dissolved gasses may have limited gear pump flow. NPSH thingy...

Or loss of power combined with software who-knows-what may have caused engines to ramp down. No time to recover.

I fear that the FDR's may not have that much info. Not so much due to loss of power (I think they have internal batts), but the power loss may have caused other devices feeding inputs to the FDR to go silent. So the operating FDR may have little to record.

Could be a complicated investigation. I would figure by now they could have downloaded the things and at least had some very preliminary findings. I would expect those to be published quickly to help the safety of the 787 fleet. The silence to this point is very loud....

 

[LittleInch]

I'm having difficulty seeing why would you get such a complete sudden electrical failure just at the point of take off?

not saying it's not possible, but there really does need to be a triggering event in my mind. And something not being caught by monitoring or maintenance.

I realise we're all struggling in the dark here, but I think we always need to consider why something failed and not what might have happened if it did.

Also it sounds like the FDR has not yet been sent to the US for interrogation, but is still in India.

 

[MechinNC]

Gonna speculate a bit (it is simply how my mind works!!):

With RAT deploying so soon, its signal to deploy was likely not due to dual engine complete failure. The engines seemed to be making some thrust as you could hear them running at low power (???) when the RAT was heard. Certainly NOT at full power! But still running, they should have been providing AC power and some hydraulics. At low N1 would hydraulic from engine driven pumps be enough to fully retract the landing gear? Maybe not, do not know. But hydraulic is normally powered by AC pumps, which may have been out. RAT can not provide enough to retract the gear, but it seems that process started.

So the RAT deploy timing seems to fit with a complete loss of AC power.

The engine control system runs off a gear driven alternator on the engine, so should have remained operational with loss of AC power.

The in-tank AC fuel pumps presumably quit, but the gear driven pump on engines should have (???) provided sufficient flow. That is presently an unknown to me. But seems like a robust design would require this.

The big question what is between the power setting levers in the cockpit and the on-engine controls. Could some sort of software command a roll back??

The longer the delay to where we hear from the investigators reading the boxes, the more my eyebrows raise...

 

[Alistair_Heaton]

I still recon something commanded the emergency valves to shut. Or there was an electrical issue which had the same effect.

If the emergency fuel valves are commanded shut this also shuts the hydraulic as well.

To my knowledge we don't have capability of emergency retraction of gear. We do have emergency extension.

I believe In Tank AC pumps have been not certifiable for decades. There will be external backup which will then power a motive flow pump internal to the tank. Most of the pumping for burn use is by suction from mechanical from the accessory gear box.

 

[LittleInch]

The loss of power also started the spool up of the APU which could account for some of the noise heard. Even flight idle would still power all the electrics which doesn't match with the RAT and the APU seemingly firing up.

The 787 is a very electrically driven airplane so hydraulic system is powered by electric driven pumps. The RAT just supplies basic flight instruments and flight control surface hydraulics, but not enough to drive the gear up hence it seems to have started going up then stopped.

I've been on the sudden and incorrect activation of the fire switches which chops everything to the engine or those fuel cut off switches below the throttles. Its very difficult to see how either of the pilots would do that at that height without having a death wish (which can now never be discounted).

But how the wiring system works and what might happen if the switchboard just exploded, who knows?

 

[Alistair_Heaton]

I can't see how they can do it symmetric even if they had a death wish.

What happened with your inadvertent hard cut? Only ever seen it in the SIM for jet. Turboprop Jetstream we had to do it occasionally if there were overtemp issues starting the Garrett engines.

 

[Alistair_Heaton]

Grounding the fleet with such a track record on the first event would be a bit over reach.

Interestingly though there hasn't been the usual finger pointing at the pilots or airline in question.

It's been remarkably balanced due to the lack of information.

 

[Garbage1]

Sept 6, 2016 Alert Service Bulletin B787-81205-SB240027-00 ELECTRICAL POWER -Electrical Power Generation and Start System - Variable Frequency Starter Generator Terminal Lugs Installation Inspection and Change.

Where VFSG Power Feeder Cables were found to contain terminal lugs incorrectly installed common to terminal blocks located in the wing front spar.

If the inspection/change in this service bulletin is not done, incorrectly installed terminal lugs may contact adjacent structure and get damaged. The damaged terminal lugs could cause electrical arcing in a flammable leakage zone which will result in an electrical short and potential loss of several functions essential for safe flight.

One engine just swapped, thus opportunity exists for service issue? From images provided in Service Bulletin I hang this on Design for creating the opportunity. If left side bolt down was lower, and matched right side hold down bolt height, there would NOT be an opportunity to make this mistake. One of those 'Latent' Defects that takes years to potentially could have caused loss of thrust at absolutely the worst time....

 

[Garbage1]

Looking at the block diagrams below, I don't see any gear driven power generation off engines other than the VFSG, with the 28Vdc FMU and EEC unit power being derived from the VSFGs. BTBs are designed normally open and are automatically closed when there is a loss of power from one of the VSFGs. So if one VSFG is shorted, automatically connecting other busses would seem to be a cascading short.... Further that affect on generator would cause sensor based automatically disengagement of gear drive between engine and VSFG, and/or connection of BTB.

Bus Power Control Unit (BPCU) is the control center of BTBs and it realizes the power transfer functions by switching on/off the breakers including BTBs, Auxiliary Power Breakers (APBs), and RAT Control Breaker (RCB). In a normal flight, all the BTBs are off and the system operates in a 4-line isolated configuration. When an accidental event results in a VFSG failure, BPCU will switch on the related BTB to transfer associated loads to the available generators. If 3 or more VFSGs are out of service, BPCU will switch on APBs to put the ASGs into operation, satisfying the load demand on the aircraft. In the worst case, all VGSGs and ASGs break down, BPCU will switch on RCB, using the RAT generator as the last power source.

 

[Brian Malone]

This video from Hoover (pilot debrief) touches on an idea  LittleInch posited back in posts #34 & 36, that the fuel cutoffs may have been inadvertently activated when attempting to adjust for a failing engine just at the V1 point.

 

[TugboatEng]

The SR71 flight manual says that the windscreen your head impacts after an engine failure is opposite of the engine that failed.

What kind of training/instruction do pilots receive to determine what engine to shut down. Why are the in such a hurry to shut the engine down ?

My first question is thinking aloud and I don't really expect an answer.

However, the second question is a problem that occurs in my fleet. We don't *currently* have any data logging which complicates troubleshooting. When there is a fault, the crews response is always to immediately reset everything to get the equipment back online as fast as possible. Well, that usually works, much to my chagrin. Now I have to troubleshoot a problem that has no symptoms. I've been trying to explain to the crews to not try to clear the fault until they have time to understand the conditions of the fault.

 

[Garbage1]

If this image is from a 787-8, and it is this easy to cut fuel off to both engines, then that is scary........ I find it hard to imagine an unprotected toggle switch design like this to cut fuel off. But if one engine failed, and you hit good engine or both toggle switches, that would explain total loss of electrical power, and RAT deployment. He also explains how quickly 787 would react to engine failure, such that no visible yaw. However, it does appear plane drifted left some for a moment.

 

[Roga50]

The Airbus design is almost identical to that. But you can't just move those switches. You have to pull up on them to move them.

 

[Garbage1]

Ok, that explains the mushroom top on toggle switches. No way to accidential bump fuel off. I disagree with some of Pilot Debrief's thoughts, and conclusions, but welcomed another perspective.

 

[TugboatEng]

Why is fuel cut so important that it needs to be the front and center most switch?