Moderators: jsumali2, richierich, ua900, PanAm_DC10, hOMSaR

 
User avatar
PixelFlight
Posts: 1030
Joined: Thu Nov 08, 2018 11:09 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Tue Apr 09, 2019 5:37 pm

zeke wrote:
PixelFlight wrote:
The OFM is an other box that get data from the DFDAU. The OMF don't feed AoA value to the FDR.


Which is a totally different box with a lot more capability than the NG (2 orders of magnitude). The MAX Enhanced DFDAU can be retrofitted to the NG, but the NG equipment cannot be used on the MAX. The ONS QAR stores 75 hours of QAR data.

Fact is you have not been able to demonstrate where the MAX data in the FDR is coming from. The MAX and NG are not the same.

The addition of an OMF with QAR data don't demonstrate either that the DFDAU get the data differently on the MAX.
:stirthepot: 737-8 MAX: "For all speeds higher than 220 Kts and trim set at a value of 2.5 units, the difficulity level of turning the manual trim wheel was level A (trim wheel not movable)." :stirthepot:
 
User avatar
zeke
Posts: 15746
Joined: Thu Dec 14, 2006 1:42 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Tue Apr 09, 2019 5:49 pm

PixelFlight wrote:
And he is the _ONLY_ difference between the two cutout switches.


No it is not, I have already clearly explained the A and B. The A and B are the primary functions of those switches, opening the relay is the secondary function.

If for whatever reason that relay did not open, the primary function of those switches will have stopped the stab trim from moving.

PixelFlight wrote:
The addition of an OMF with QAR data don't demonstrate either that the DFDAU get the data differently on the MAX.


And yet the manufacturers says the MAX has 100 times more data than the NG. Looks like they will need to rewrite their manuals yet again based upon a.net posts.
“Don't be a show-off. Never be too proud to turn back. There are old pilots and bold pilots, but no old, bold pilots.” E. Hamilton Lee, 1949
 
hivue
Posts: 2125
Joined: Tue Feb 26, 2013 2:26 am

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Tue Apr 09, 2019 6:07 pm

jollo wrote:
[*]did some engineer raise concerns about this scenario creating a potentially catastrophic scenario on a single sensor failure event, mitigated only by pilot reactions? Probably, but we'll never know if he was silenced by management decision or if he was convinced by a risk analysis. We do know that nobody blew the whistle.[/list]


This is the other shoe I'm waiting to hear drop. I'm sure there are plenty of journalists out there trying to pry a leaked email email or something like that out of somebody. We'll see if anything turns up.

Great (speculative) summary. Thanks.
"You're sitting. In a chair. In the SKY!!" ~ Louis C.K.
 
User avatar
PixelFlight
Posts: 1030
Joined: Thu Nov 08, 2018 11:09 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Tue Apr 09, 2019 6:24 pm

zeke wrote:
PixelFlight wrote:
And he is the _ONLY_ difference between the two cutout switches.


No it is not, I have already clearly explained the A and B. The A and B are the primary functions of those switches, opening the relay is the secondary function.

If for whatever reason that relay did not open, the primary function of those switches will have stopped the stab trim from moving.

Switches are physical devices that only follow the physic's laws, not the letters on a schematic. There is physically no such thing as "primary functions" or "secondary function". Either of the switches cut the power to the electric stab trim motor, end of the story. This is confirmed by the ET302 preliminary report by the way. A quadruple failure mode (2 switches + 1 relay + 1 reason to cutoff) is so remotely improbable that completely out of the design goal.
zeke wrote:
PixelFlight wrote:
The addition of an OMF with QAR data don't demonstrate either that the DFDAU get the data differently on the MAX.


And yet the manufacturers says the MAX has 100 times more data than the NG. Looks like they will need to rewrite their manuals yet again based upon a.net posts.

The amount of data (probably only for the OME and not for the FDR) don't say anything about a change in the path of the AoA value down to the FDR. That precise path don't pass through the OME, but through the DFDAU.
:stirthepot: 737-8 MAX: "For all speeds higher than 220 Kts and trim set at a value of 2.5 units, the difficulity level of turning the manual trim wheel was level A (trim wheel not movable)." :stirthepot:
 
User avatar
PixelFlight
Posts: 1030
Joined: Thu Nov 08, 2018 11:09 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Tue Apr 09, 2019 6:33 pm

hivue wrote:
jollo wrote:
[*]did some engineer raise concerns about this scenario creating a potentially catastrophic scenario on a single sensor failure event, mitigated only by pilot reactions? Probably, but we'll never know if he was silenced by management decision or if he was convinced by a risk analysis. We do know that nobody blew the whistle.[/list]


This is the other shoe I'm waiting to hear drop. I'm sure there are plenty of journalists out there trying to pry a leaked email email or something like that out of somebody. We'll see if anything turns up.

Great (speculative) summary. Thanks.

Don't forget the professional investigators mandated by the justice.
:stirthepot: 737-8 MAX: "For all speeds higher than 220 Kts and trim set at a value of 2.5 units, the difficulity level of turning the manual trim wheel was level A (trim wheel not movable)." :stirthepot:
 
User avatar
zeke
Posts: 15746
Joined: Thu Dec 14, 2006 1:42 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Tue Apr 09, 2019 6:56 pm

PixelFlight wrote:
Switches are physical devices that only follow the physic's laws, not the letters on a schematic. There is physically no such thing as "primary functions" or "secondary function". Either of the switches cut the power to the electric stab trim motor, end of the story. This is confirmed by the ET302 preliminary report by the way. A quadruple failure mode (2 switches + 1 relay + 1 reason to cutoff) is so remotely improbable that completely out of the design goal.


The more you post the clearer it is becoming evident to everyone that you do not know what you are taking about. Think of the stab trim motor as being a simple load like light globe in a house. The primary way to turn a light globe off is by actioning the switch in the same room associated with the particular light circuit, the secondary method would be to trip ground fault circuit interrupter (GFCI) for all the light circuits on your main switchboard. The primary means for removing power from the motor is A, that is where the motor is controlled on or off by the pilot, FCC, or autopilot. The secondary method is to open the relay, the pilot, FCC, or autopilot do not use that relay as the primary means to control power to the motor.

PixelFlight wrote:
The amount of data (probably only for the OME and not for the FDR) don't say anything about a change in the path of the AoA value down to the FDR. That precise path don't pass through the OME, but through the DFDAU.


It is clear again you do not know what you are talking about. Up thread you said the FDR got the AOA values from the SYMD, I challenged you on that stating the MAX was different. Now you are saying it comes from the DFDAU, but that is a NG part, the MAX has the Enhanced DFDAU that processes 100 time more data than the NG DFDAU.

Until we know exactly where the AOA values on the FDR are actually being recorded from, we will not know what other failure modes could be in play that would permit a serviceable AOA probe output to be transformed by a significant offset to start this chain of events resulting in an unsafe condition.
“Don't be a show-off. Never be too proud to turn back. There are old pilots and bold pilots, but no old, bold pilots.” E. Hamilton Lee, 1949
 
User avatar
PixelFlight
Posts: 1030
Joined: Thu Nov 08, 2018 11:09 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Tue Apr 09, 2019 7:19 pm

zeke wrote:
PixelFlight wrote:
Switches are physical devices that only follow the physic's laws, not the letters on a schematic. There is physically no such thing as "primary functions" or "secondary function". Either of the switches cut the power to the electric stab trim motor, end of the story. This is confirmed by the ET302 preliminary report by the way. A quadruple failure mode (2 switches + 1 relay + 1 reason to cutoff) is so remotely improbable that completely out of the design goal.


The more you post the clearer it is becoming evident to everyone that you do not know what you are taking about. Think of the stab trim motor as being a simple load like light globe in a house. The primary way to turn a light globe off is by actioning the switch in the same room associated with the particular light circuit, the secondary method would be to trip ground fault circuit interrupter (GFCI) for all the light circuits on your main switchboard. The primary means for removing power from the motor is A, that is where the motor is controlled on or off by the pilot, FCC, or autopilot. The secondary method is to open the relay, the pilot, FCC, or autopilot do not use that relay as the primary means to control power to the motor.

Sorry, but you will not change the laws of physic. Remember that the context of this few messages is about the "STAB TRIM CUTOUT SW". In this context. yes the ELECTRIC STAB TRIM MOTOR assembly is just a load on the triple phase 115V bus after the R64 relay. Sorry again, but there is no way for the FCC or the autopilot to remove the 115V bus power to the ELECTRIC STAB TRIM MOTOR assembly. This is just exactly what the schematic say. You can create a real circuit or use a simulation if you want to verify.
zeke wrote:
PixelFlight wrote:
The amount of data (probably only for the OME and not for the FDR) don't say anything about a change in the path of the AoA value down to the FDR. That precise path don't pass through the OME, but through the DFDAU.


It is clear again you do not know what you are talking about. Up thread you said the FDR got the AOA values from the SYMD, I challenged you on that stating the MAX was different. Now you are saying it comes from the DFDAU, but that is a NG part, the MAX has the Enhanced DFDAU that processes 100 time more data than the NG DFDAU.

Until we know exactly where the AOA values on the FDR are actually being recorded from, we will not know what other failure modes could be in play that would permit a serviceable AOA probe output to be transformed by a significant offset to start this chain of events resulting in an unsafe condition.

AoA -> SYMD -> DFDAU -> FDR.
A eDFDAU is still a DFDAU.
:stirthepot: 737-8 MAX: "For all speeds higher than 220 Kts and trim set at a value of 2.5 units, the difficulity level of turning the manual trim wheel was level A (trim wheel not movable)." :stirthepot:
 
smartplane
Posts: 1648
Joined: Fri Aug 03, 2018 9:23 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Tue Apr 09, 2019 7:24 pm

morrisond wrote:
SomebodyInTLS wrote:
XRAYretired wrote:
Well, not quite I think, since single sensor is not a legitimate design approach for a 'hazardous' classified system even if mitigated by trim cut-out for sensor failing high was considered legitimate at the time.


I've thought for a while now that it was like BaconButty described - the original intention was to be an upgrade to STS to maintain the NG "feel" close to stall (not safety-critical). However, flight testing revealed worse than anticipated characteristics close to stall, hence the pumped-up MCAS to act as a safety-critical anti-stall measure. Then management leans on the engineers not to allow any changes to reach the pilots, their training and other documentation... and at that point questions need to be answered about how much was made known to the authorities, how much was overlooked in haste and how much was deliberately repressed.

Part of the "it's not anti-stall", "it's not about safety" BS may be down to it having changed purpose during flight testing and the same questions re. culpability need to be asked about the people spreading this argument as well.



If you read the best decription I've seen on MCAS from 737.org.uk

Which is a press release issued by Boeing. Probably crafted by the same team that came up with the memorable 'safe making safer'.
 
flybucky
Posts: 376
Joined: Tue Apr 17, 2018 7:44 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 12:13 am

zeke wrote:
The switches are not labelled A bad B, they are the placeholders for the other A and B on the motor control and and the FCC input.
The primary has very similar function to the autopilot switch on the NG, it will remove power from the FCCs.
The backup switch does not remove power from the FCCs. Much like the stab cutout on the NG.
The A and B are the primary functions of those switches, opening the relay is the secondary function.
If for whatever reason that relay did not open, the primary function of those switches will have stopped the stab trim from moving.

This is great info, thanks.

One question though:
You said that PRI is similar to the NG Auto Pilot Cutout, and B/U is similar to the NG Main Elect Cutout.
PRI Cutout is on the left, whereas NG Auto Pilot Cutout is on the right.
B/U Cutout is on the right, whereas NG Main Elect Cutout is on the left.

I would have thought it would be easier to keep the left/right the same from NG to MAX. Did they really flip the left/right?
 
9Patch
Posts: 642
Joined: Wed Mar 13, 2019 10:38 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 1:25 am

6 Minutes of Terror: What Passengers and Crew Experienced Aboard Ethiopian Airlines Flight 302

8:40:35 a.m.: “Stab trim cut-out!” the first officer says — that is, the electrical trim system for the stabilizer must be disconnected. “Stab trim cut out!” The captain agrees. They recognize that they are experiencing a runway MCAS, a condition that destroyed Lion Air Flight 610 less than five months before. They flip a pair of switches that disables the trim system, and along with it the MCAS.

8:41:46 a.m.: The pilots have made a mistake that is easy to overlook amid the confusion, but which will have severe consequences. When they shut off the electric trim system to disable MCAS, they also shut off their ability to use the electric switch on the top of their yokes to trim the stabilizer back into a neutral position.

http://nymag.com/intelligencer/2019/04/ ... light.html
 
PacificWest
Posts: 102
Joined: Sun May 06, 2007 3:52 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 4:53 am

jollo wrote:
flybucky wrote:
jollo wrote:
Since getting to the bottom of whether AUTO-PILOT STAB TRIM CUTOFF disables STS on a 737 NG seems to be harder than expected

Peter Lemme said that on pre-MAX 737s, the AUTOPILOT STAB TRIM CUTOFF would disable STS:

The cutout switch function was changed with the 737MAX from all prior 737 models. The legacy switch combination was one switch to cutout electric trim altogether, the other to cutout the autopilot trim commands. MCAS and Speed Trim System are both commands from the “autopilot”. With the legacy switch configuration, the flight crew can disable the autopilot commands and retain electric trim.

https://www.satcom.guru/2019/04/what-ha ... et302.html


Thanks a lot: I'll try to find additional confirmation out of an abundance of caution, but this pretty much clinches the case for me.

IMO, the chain of decision leading to the current MCAS design was something like:
  • aero analysis (or flight testing) highlights non-linear, non-monotonous response to pitch inputs in some flight envelope regions (flaps up, high speed, high AoA, etc.)
  • longitudinal stability augmentation system needed to comply with FAR
  • MCAS out of the KC-767A / KC-46 design is an "off the shelf", proven solution, and is also compatible with 737 "legacy" air data architecture (2 AoA channels, not 3)
  • MCAS is implemented as an "STS extension" run by the FCC; MCAS is disabled when A/P in engaged, so no interference; MCAS operates in rather different flight envelope regions vs. STS, so no interference there either
  • MCAS has stronger authority than STS (i.e. can command a higher stab trim angle at a higher rate of stab trim travel) and, running off a single AoA input channel like STS, introduces the potential for a "MCAS runaway" on a single sensor failure scenario that was simply not material with STS. However, disabling MCAS in an AoA disagree condition (logic also out of the KC-767A / KC-46 design), plus the possibility to disable MCAS manually via the existing STAB TRIM CUTOFF AUTO-PILOT switch is deemed sufficient to manage safely any "MCAS runaway" scenario.
  • Level D difference training will be needed to enable pilots to stay out of the relevant flight envelope corners with an inoperative/disabled MCAS. This does not prevent common type rating --> mission accomplished.

Up to this point, everything is cool. If things had been left at that, MAX would have been as safe as NG. Even marginally safer, in the vast majority of the flight envelope and, with an operative MCAS, across 100% of the envelope.

Note: switching only STAB TRIM CUTOFF AUTO-PILOT to CUTOFF while leaving STAB TRIM CUTOFF MAIN ELECT to NORMAL would have disabled all automations while leaving the electrical stab trim servo operative, to help pilots cope with aero loads on the stabilizer via thumb switches. Both Lion Air and Ethiopian crashes could easily have been avoided.

Enter the bad guy (circa 2011): the commercial imperative to avoid simulator training at all costs. Clients are clamoring for minimal transitions costs, and an over-eager sales manager, desperate to secure a mega-order out of Airbus hands, has promised to SW a 1M$/plane discount if any simulator training will be needed. That's a 280M$ penalty at stake. Who's gonna pay that kind of money, you? Then back to the design table and make that training requirement go away. Pronto: flight testing is just a few years away.

  • so no sim training
  • therefore MCAS has to be "always on",
  • therefore the AoA disagree flag needs to go --> compromise: it becomes an extra option, so it does not need to be incorporated into NNCs
  • therefore the STAB TRIM CUTOFF AUTO-PILOT switch needs to go, too. But removing a switch from the cockpit would be a Level D difference, while relabeling is just Level B (no sim training needed). So both STAB TRIM CUTOFF switches get relabeled rather ambiguously PRI and B/U, and both get the same functions - equivalent to the old STAB TRIM CUTOFF MAIN ELECT switch: operating either one will disable the stab trim electric actuator (so only manual trim wheel left). THIS IS A FUNCTIONAL CHANGE (Level C at least) that was hidden as "relabeling": this is the one and only criminal act in this sequence.
  • any "MCAS runaway" scenario needs to be managed through the existing Stab Trim Runaway NNC; luckily, the old checklist never calls for operating the cutout switches separately, so this does not give away the functional change on the cutout switches
  • but this is bound to generate awkward questions, so MCAS gets expunged from the FCOM (but not completely: someone forgets to purge it out of the acronyms reference list).
  • did some engineer raise concerns about this scenario creating a potentially catastrophic scenario on a single sensor failure event, mitigated only by pilot reactions? Probably, but we'll never know if he was silenced by management decision or if he was convinced by a risk analysis. We do know that nobody blew the whistle.

Conjecture, yes, but IMO its chillingly fits.



Damn, great post. This is the kind of well-informed 'speculation' I always hope to find here.
 
User avatar
zeke
Posts: 15746
Joined: Thu Dec 14, 2006 1:42 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 5:17 am

PixelFlight wrote:
Switches are physical devices that only follow the physic's laws, not the letters on a schematic. There is physically no such thing as "primary functions" or "secondary function". Either of the switches cut the power to the electric stab trim motor, end of the story. This is confirmed by the ET302 preliminary report by the way. A quadruple failure mode (2 switches + 1 relay + 1 reason to cutoff) is so remotely improbable that completely out of the design goal.


Tell us all under what conditions the R64 relay closes ? How does the pilot, FCC, and autopilot control the R64 relay if as you suggest that is the only way to remove power from the stab trim ?

Are you suggesting the pilot, FCC, and autopilot cannot trim the stab with flaps down ? How do you trim for takeoff ? How is STS able to operate after takeoff.

I have clearly said the way to control the stab is via A, controlling it means a circuit is completed and power is made available to the motor. It is the light switch in the room that turns the bulb on and off.

PixelFlight wrote:
AoA -> SYMD -> DFDAU -> FDR. A eDFDAU is still a DFDAU .


Prove it. Show us where is says that in the manuals.

An Enhanced DFDAU is not a DFDAU, they are not interchangeable. You cannot put the DFDAU of the NG on the MAX. But like saying a GPWS is the same as EGPWS. Different functions, capabilities, and processing.
“Don't be a show-off. Never be too proud to turn back. There are old pilots and bold pilots, but no old, bold pilots.” E. Hamilton Lee, 1949
 
User avatar
zeke
Posts: 15746
Joined: Thu Dec 14, 2006 1:42 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 5:28 am

flybucky wrote:
I would have thought it would be easier to keep the left/right the same from NG to MAX. Did they really flip the left/right?


No idea, the switches if needed are normally used at the same time.
“Don't be a show-off. Never be too proud to turn back. There are old pilots and bold pilots, but no old, bold pilots.” E. Hamilton Lee, 1949
 
sgrow787
Posts: 450
Joined: Fri May 16, 2014 8:12 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 6:08 am

PacificWest wrote:
jollo wrote:
flybucky wrote:
Peter Lemme said that on pre-MAX 737s, the AUTOPILOT STAB TRIM CUTOFF would disable STS:

The cutout switch function was changed with the 737MAX from all prior 737 models. The legacy switch combination was one switch to cutout electric trim altogether, the other to cutout the autopilot trim commands. MCAS and Speed Trim System are both commands from the “autopilot”. With the legacy switch configuration, the flight crew can disable the autopilot commands and retain electric trim.

https://www.satcom.guru/2019/04/what-ha ... et302.html


Thanks a lot: I'll try to find additional confirmation out of an abundance of caution, but this pretty much clinches the case for me.

IMO, the chain of decision leading to the current MCAS design was something like:
  • aero analysis (or flight testing) highlights non-linear, non-monotonous response to pitch inputs in some flight envelope regions (flaps up, high speed, high AoA, etc.)
  • longitudinal stability augmentation system needed to comply with FAR
  • MCAS out of the KC-767A / KC-46 design is an "off the shelf", proven solution, and is also compatible with 737 "legacy" air data architecture (2 AoA channels, not 3)
  • MCAS is implemented as an "STS extension" run by the FCC; MCAS is disabled when A/P in engaged, so no interference; MCAS operates in rather different flight envelope regions vs. STS, so no interference there either
  • MCAS has stronger authority than STS (i.e. can command a higher stab trim angle at a higher rate of stab trim travel) and, running off a single AoA input channel like STS, introduces the potential for a "MCAS runaway" on a single sensor failure scenario that was simply not material with STS. However, disabling MCAS in an AoA disagree condition (logic also out of the KC-767A / KC-46 design), plus the possibility to disable MCAS manually via the existing STAB TRIM CUTOFF AUTO-PILOT switch is deemed sufficient to manage safely any "MCAS runaway" scenario.
  • Level D difference training will be needed to enable pilots to stay out of the relevant flight envelope corners with an inoperative/disabled MCAS. This does not prevent common type rating --> mission accomplished.

Up to this point, everything is cool. If things had been left at that, MAX would have been as safe as NG. Even marginally safer, in the vast majority of the flight envelope and, with an operative MCAS, across 100% of the envelope.

Note: switching only STAB TRIM CUTOFF AUTO-PILOT to CUTOFF while leaving STAB TRIM CUTOFF MAIN ELECT to NORMAL would have disabled all automations while leaving the electrical stab trim servo operative, to help pilots cope with aero loads on the stabilizer via thumb switches. Both Lion Air and Ethiopian crashes could easily have been avoided.

Enter the bad guy (circa 2011): the commercial imperative to avoid simulator training at all costs. Clients are clamoring for minimal transitions costs, and an over-eager sales manager, desperate to secure a mega-order out of Airbus hands, has promised to SW a 1M$/plane discount if any simulator training will be needed. That's a 280M$ penalty at stake. Who's gonna pay that kind of money, you? Then back to the design table and make that training requirement go away. Pronto: flight testing is just a few years away.

  • so no sim training
  • therefore MCAS has to be "always on",
  • therefore the AoA disagree flag needs to go --> compromise: it becomes an extra option, so it does not need to be incorporated into NNCs
  • therefore the STAB TRIM CUTOFF AUTO-PILOT switch needs to go, too. But removing a switch from the cockpit would be a Level D difference, while relabeling is just Level B (no sim training needed). So both STAB TRIM CUTOFF switches get relabeled rather ambiguously PRI and B/U, and both get the same functions - equivalent to the old STAB TRIM CUTOFF MAIN ELECT switch: operating either one will disable the stab trim electric actuator (so only manual trim wheel left). THIS IS A FUNCTIONAL CHANGE (Level C at least) that was hidden as "relabeling": this is the one and only criminal act in this sequence.
  • any "MCAS runaway" scenario needs to be managed through the existing Stab Trim Runaway NNC; luckily, the old checklist never calls for operating the cutout switches separately, so this does not give away the functional change on the cutout switches
  • but this is bound to generate awkward questions, so MCAS gets expunged from the FCOM (but not completely: someone forgets to purge it out of the acronyms reference list).
  • did some engineer raise concerns about this scenario creating a potentially catastrophic scenario on a single sensor failure event, mitigated only by pilot reactions? Probably, but we'll never know if he was silenced by management decision or if he was convinced by a risk analysis. We do know that nobody blew the whistle.

Conjecture, yes, but IMO its chillingly fits.



Damn, great post. This is the kind of well-informed 'speculation' I always hope to find here.


I'll second that. Even if speculation, a connect-the-dots approach gives context to the discussions.
Just one sensor,
Oh just one se-en-sor,
Just one sensor,
Ooh ooh oo-ooh
Oo-oo-ooh.
 
HaulSudson
Posts: 101
Joined: Tue Feb 12, 2019 6:02 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 7:21 am

sorry if it has been asked before: If the Pilot would have given control to the FO, wouldn't they have survived?

The AoA sensor on the right appeared to be OK, after all. Hence no MCAS intervention.


That would be quite ironic, given the disgusting posts that the late man had to endure here.
 
User avatar
speedbored
Posts: 2230
Joined: Fri Jul 19, 2013 5:14 am

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 7:27 am

zeke wrote:
Tell us all under what conditions the R64 relay closes ? How does the pilot, FCC, and autopilot control the R64 relay if as you suggest that is the only way to remove power from the stab trim?

Are you suggesting the pilot, FCC, and autopilot cannot trim the stab with flaps down ? How do you trim for takeoff ? How is STS able to operate after takeoff.

My reading of the schematic (on page 94 of this thread) is that the R64 relay controls whether 115v power is supplied to the Electric Stab Trim Motor module but it is logic within this module that decides whether or not that power is actually applied to the motor, in which direction, and at what speed the motor should run. So the pilot, FCC and autopilot do not need to control the R64 relay; they control trim via the other connections to the Electric Stab Trim Motor module.

The R64 relay allows power to reach the Electric Stab Trim module only when both of the Stab Trim Cutout switches are set to Norm.
The link from the R64 energizing signal, through the R1192 Main Trim Flaps Relay, appears to be a trim speed signal to vary the trim speed depending on whether flaps are up or down.

zeke wrote:
I have clearly said the way to control the stab is via A, controlling it means a circuit is completed and power is made available to the motor. It is the light switch in the room that turns the bulb on and off.

I'm not convinced about this. The A and B circles on the schematic are simply a way of showing connections without having to route a line on the diagram. So the FCC signal routing through the top cutout switch connects to the A/P Enable Speed And Direction sub-module within the Electric Stab Trim Motor module.

Regardless of the purpose of the connection to the FCC through A, what is clear is that normal trim up/down signals from the FCC are routed via the module labelled F/O Col Cutout SW Mod. There is also a Capt Col Cutout SW Mod. These two Cutout SW Mods control whether or not trim commands from the electric trim switches or FCC are passed on to the trim motor module, based on the Captain and F/O control column movements:
If both pilots are pulling nose up on the yokes, manual electric nose-down trim is disabled
(this functionality can be overridden by the Stab Trim Override Switch)
If both pilots are pushing nose down on the yokes, manual electric nose-up trim is disabled
(this functionality can be overridden by the Stab Trim Override Switch)
If the F/O is pulling nose up on the yokes, FCC nose-down trim is disabled.
(FCC can add nose-down trim even when captain is pulling nose up)
If the F/O is pushing nose down on the yokes, FCC nose-up trim is disabled.
(FCC can add nose-up trim even when captain is pushing nose down)

While there is not enough information in the schematic to determine the exact purpose of the FCC signal via A, the function of the MCAS Engage signal does provide a clue - it disables nose-down commands from the FCC, but not nose-up commands. It also sets the trim motor to apply FCC trim at the same speed as it would when flaps are down. So, in order for MCAS trim commands to get through from the FCC, my suspicion is that MCAS trim commands are routed from the FCC via A. Another clue is the fact that neither of the yokes can override this signal.
 
Spotter1967
Posts: 11
Joined: Fri May 14, 2010 12:13 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 7:32 am

HaulSudson wrote:
sorry if it has been asked before: If the Pilot would have given control to the FO, wouldn't they have survived?

The AoA sensor on the right appeared to be OK, after all. Hence no MCAS intervention.


That would be quite ironic, given the disgusting posts that the late man had to endure here.

MCAS cannot be switched during flight such that it uses the other, functioning, AOA-sensor.
Only after landing and switch off of systems, the other AOA-sensor will be used.
 
HaulSudson
Posts: 101
Joined: Tue Feb 12, 2019 6:02 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 7:36 am

Spotter1967 wrote:
HaulSudson wrote:
sorry if it has been asked before: If the Pilot would have given control to the FO, wouldn't they have survived?

The AoA sensor on the right appeared to be OK, after all. Hence no MCAS intervention.


That would be quite ironic, given the disgusting posts that the late man had to endure here.

MCAS cannot be switched during flight such that it uses the other, functioning, AOA-sensor.
Only after landing and switch off of systems, the other AOA-sensor will be used.


OK, thanks. I mistakenly thought I had read somewhere that it would use data by the sensor(s) on the side of the pilot flying.
 
Spotter1967
Posts: 11
Joined: Fri May 14, 2010 12:13 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 7:42 am

zeke wrote:
flybucky wrote:
I would have thought it would be easier to keep the left/right the same from NG to MAX. Did they really flip the left/right?


No idea, the switches if needed are normally used at the same time.

Lots of discussion here on the function of the ELEC STAB TRIM cutout switches on the NG and MAX.

FACT: Boeing changed the function as well as the nomenclature. It can be seen in the 1 hour iPad training required for pilots to train on the MAX. Screenshot see link below.

On the NG:
The left switch was used to disable the stab trim motor.
The right switch labeled AUTO PILOT disables AUTO PILOT initiated command to the horizontal stab (like STS)
So by switching the AUTO PILOT switch to cutout, manual trim using the yoke switch was still possible, while STS was disabled.

On the MAX
Both PRI an BU switches share the same function: disable the stab trim motor
Pilots do not have the possibily to disable AUTO PILOT commands for the stabilizer (like STS but Boeing could have decided to have MCAS use this switch as well to disable MCAS)
So no way to disable MCAS completely besides using the STAB TRIM cutout switches (as described in runaway stabilzer procedure)

More info here: https://feitoffake.wordpress.com/2019/0 ... g-737-max/
 
User avatar
zeke
Posts: 15746
Joined: Thu Dec 14, 2006 1:42 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 10:08 am

Spotter1967 wrote:


A number of key statements made in that article are incorrect and conflict with the excerpts of Boeing material contained within.
“Don't be a show-off. Never be too proud to turn back. There are old pilots and bold pilots, but no old, bold pilots.” E. Hamilton Lee, 1949
 
jollo
Posts: 398
Joined: Mon Aug 08, 2011 7:24 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 2:23 pm

zeke wrote:
Spotter1967 wrote:


A number of key statements made in that article are incorrect and conflict with the excerpts of Boeing material contained within.


Zeke, I'm sure you're very knowledgeable, but it would help the quality of the discussion if, in addition to posting negatives, you could also contribute positive info, e.g. identify which specific bits of info are incorrect and what would be correct instead. I bet most members here would be genuinely eager to know.

BTW, rooting out inconsistencies/omissions/ambiguities in Boeing's documents and statements is kind of the name of the game, nowadays.

zeke wrote:
The primary has very similar function to the autopilot switch on the NG, it will remove power from the FCCs.


Assuming the PRImary stab trim cutoff switch is the top one in this functional diagram, it does appear that setting PRI to CUTOFF in a MAX disables both trim inputs from the FCCs and trim inputs from the yoke thumb switches. This is also confirmed by Boeing's iPad training course (screenshot), saying:

The Stab Trim switch nomenclature has changed as follows:
  • NG "MAIN ELEC" --> MAX "PRI"
  • NG "AUTO PILOT" --> MAX "B/U"
The MAX PRI and B/U switches will both deactivate the main electrical and auto pilot stab trim operation.


How is this "very similar" to the AUTO PILOT STAB TRIM switch on the NG? According to the FCOM (and to basically every other source linked to in this thread), setting AUTO PILOT to CUTOFF in a NG disables only FCC's trim inputs (including STS), while leaving thumb switches operative. Do you have any evidence proving otherwise?

(BTW, if this is confirmed, we have yet another inconsistency/intentional obfuscation in Boeing's material: writing NG "AUTO PILOT" --> MAX "B/U" and "The ... B/U switch will ... deactivate the main electrical ... stab trim operation" in the same slide conveys, beyond the main claim "This is just a relabeling!", the implicit message that, on the NG, setting AUTO PILOT to CUTOFF will disable the "main electrical stab trim operation". Could this be verified hands-on by a helpful a.netter who is also a NG driver?)

I am aware of the fact that procedures (including NNCs) on the NG never call for operating the stab trim cutout switches separately, but the point to be confirmed here is that NG pilots blessed by a full understanding of the system can, if needed, deactivate automated trim inputs while retaining electric stab trim servo. In the MAX pilots do not have this option any more (which could have been quite useful, and potentially life-saving, in both the Lion Air and Ethiopian accidents).
 
Spotter1967
Posts: 11
Joined: Fri May 14, 2010 12:13 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 2:39 pm

Fully agree with Jollo here.
If someone states "A number of key statements made in that article are incorrect and conflict with the excerpts of Boeing material contained within." make sure to explain

Additionally it is very clear Boeing not only relabled the ELEC TRIM switches but it also changed the function of these switches.
These two crashes could have been prevented with proper documentation by Boeing,
 
morrisond
Posts: 3189
Joined: Thu Jan 07, 2010 12:22 am

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 6:18 pm

Spotter1967 wrote:
zeke wrote:
flybucky wrote:
I would have thought it would be easier to keep the left/right the same from NG to MAX. Did they really flip the left/right?


No idea, the switches if needed are normally used at the same time.

Lots of discussion here on the function of the ELEC STAB TRIM cutout switches on the NG and MAX.

FACT: Boeing changed the function as well as the nomenclature. It can be seen in the 1 hour iPad training required for pilots to train on the MAX. Screenshot see link below.

On the NG:
The left switch was used to disable the stab trim motor.
The right switch labeled AUTO PILOT disables AUTO PILOT initiated command to the horizontal stab (like STS)
So by switching the AUTO PILOT switch to cutout, manual trim using the yoke switch was still possible, while STS was disabled.

On the MAX
Both PRI an BU switches share the same function: disable the stab trim motor
Pilots do not have the possibily to disable AUTO PILOT commands for the stabilizer (like STS but Boeing could have decided to have MCAS use this switch as well to disable MCAS)
So no way to disable MCAS completely besides using the STAB TRIM cutout switches (as described in runaway stabilzer procedure)

More info here: https://feitoffake.wordpress.com/2019/0 ... g-737-max/


I wonder if the ET302 Pilots who were trained on both the NG and MAX misunderstood what the switches on the MAX did.

Being familiar with the NG they would have known that hitting the right switch turned off HAL but still gave them the Electric Trim.

They might have understood the MCAS memo to be Hit Both switches to turn off MCAS and A/P but that Electric Trim would still be available.

That would be the sensible thing to assume as who would design a plane that at the first hint of the STS or A/P or MCAS failure - Electric Trim would not be available as it was on the NG?

That fits the timeline a lot better when the FO came back to the pilot only 8 seconds later saying the Manual didn't work - that could have been the Pilot asking to try the Electric Trim with his trim switch. You would have assumed the FO would have asked the Pilot to try and help him with the crank if it was hard to move.

They then spent the next few minutes trying to get the electric trim to work again without HAL before giving up and flipping the switches back on.

Has anyone seen the iPad course? Does it specifically say that when when you turn off either or both of these switches Electric Trim is not available?

Edit - Sorry I didn't read Jollo's great post above. It appears it states it.
 
fadecfault
Posts: 164
Joined: Mon Nov 05, 2007 8:44 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 6:24 pm

Those of you arguing over the switch wiring need to know that diagram does not fully represent the actual system. The actual WDM needs to be seen to make any kind of factual statement.
The views and opinions written here are my own and do not reflect those of my employer.
 
Spotter1967
Posts: 11
Joined: Fri May 14, 2010 12:13 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 6:35 pm

[url][/url]
fadecfault wrote:
Those of you arguing over the switch wiring need to know that diagram does not fully represent the actual system. The actual WDM needs to be seen to make any kind of factual statement.

The iPad course for pilots is clear. Both ELEC TRIM switches (labeled PRI and B/U) share the same function. Disable the horizontal stabilizer trim motor. Why would the switches have different functions? B/U of backup means it does the same as PRImary.
 
RedBob
Posts: 12
Joined: Wed Apr 10, 2019 5:47 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 7:59 pm

jollo wrote:
MCAS out of the KC-767A / KC-46 design is an "off the shelf", proven solution, and is also compatible with 737 "legacy" air data architecture (2 AoA channels, not 3)


This is something that I'm actually not sure about.

If MCAS does what it says it does, then it essentially should compensate for the augmented plane characteristics with the position of the stabilizer based on several values. Problem is, when you want to compensate something with the stabilizer, the position of the stabilizer is essentially becomes one of your inputs.

The way I understand 767 works is that you set trim digitally, and the adjustment would amount to the difference between the physical position of the stabilizer and the location it is set to be in, and this difference would be calculated based on the number of parameters. In this case, you have a clear trim input and output.

However, in 737 there's no target trim in 767 sense: the trim position as indicated by trim wheel, and the physical position of the stabilizer are one and the same. Therefore, your "adjusted" trim output in one cycle can be your actual trim input in another cycle. With this in mind, the cyclical nature of MCAS and "reset" would make some sense, as it essentially tries to see if the trim is accepted by the pilots, and after a while essentially assumes that it's what they wanted. However, this obviously can produce some very odd feedback loops.

I think more details are needed on how the adjustment works exactly, because personally I find it odd that the system that is said to improve handling characteristics or compensate for lift would work in 10-second cycles, have reset positions, and the like - all while being remarkably ignorant of the stick inputs.
 
freakyrat
Posts: 2195
Joined: Fri Aug 22, 2008 1:04 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 8:19 pm

mjoelnir wrote:
ikramerica wrote:
planecane wrote:

You are very insistent that the "obsolete trim system" of the 737 must be completely redesigned. Perhaps the components of the trim system are just extremely reliable and the instances where it needs to be cut out (except when the terrible MCAS implementation was added) is exceedingly rare. Perhaps if the electric trim stops working somewhere near where the trim should be, the aircraft can still be controlled and landed using the elevators or, in that scenario, the manual wheel works.

Obviously runaway trim can't be very common at all on the pre-MCAS/pre-MAX 737's or the issue of being able to recover from it ending up impossible would have come up many years ago.

Using the history of the 737 series in service as the data, there is nothing wrong with flying around with the obsolete trim system. The problems with the trim system are with the algorithm that is commanding the trim motor. That's the most "state of the art" part of the whole thing. They could have put the 787 trim system on the MAX and it would have resulted in the same two crashes if MCAS was there (which it would have been).

It isn't your vision of modern technology for a cable to be running from the cockpit to the tail and physically turning a jackscrew but it works fine.

While I don't agree with a lot of what he is saying, there is still a case to be made that if airplanes were designed today, the controls would be very different. Same goes for cars.

The more we find out about the 737 system, the more we see how aircraft systems in general are designed so they "feel" like obsolete technology and use control mechanisms that mimic designs that should be outmoded. I'm not talking grandfathering exclusively. But the main paradigm of the feedback the aircraft gives is based on how much, much, much older aircraft with manual controls reacted and how they were controlled. Many of the paradigms go back to the piston era.

This happens because, even with a new type, the PILOTS are grandfathered in. To get experienced pilots, you need to have them already know how to fly the plane based on their experience. If you changed it up completely, you would have no one with experience to fly any of the planes you sell. Yes, you can train people, but getting enough people up to speed with enough experience would be astronomically costly and probably impossible.

So we have various aircraft of various vintages all trying to approximate how an older aircraft would behave just because. And as technology advances further and further, there will be more conflicts between how it was always done/needs to be done and how it really should be done.

The really ironic part is that as aircraft are more automated, the pilots are doing less actual flying anyway and too often forget what they are supposed to know in a panic situation because they barely ever use it, so completely changing the paradigm shouldn't matter as much.


Airbus broke with that tradition. The joystick has no feedback. Even pilots trained on the 737 seem to be able to switch over to the A320 and fly those frames.


I just a small aircraft GA pilot here who has some experience in an FDS A320 Flight Simulator. It took me about 20-30 minutes to get use to flying with a sidestick. It was quite easy after I got the hang of it.
 
User avatar
PixelFlight
Posts: 1030
Joined: Thu Nov 08, 2018 11:09 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 8:27 pm

zeke wrote:
Tell us all under what conditions the R64 relay closes ?

"28V DC BUS 2 = nominal" & "STAB TRIM CONTROL fuse = good" & "PRI STAB TRIM CUTOUT SW = NORM" & "B/U STAB TRIM CUTOUT SW = NORM"

zeke wrote:
How does the pilot, FCC, and autopilot control the R64 relay if as you suggest that is the only way to remove power from the stab trim ?

* The pilot can control the R64 by moving "PRI STAB TRIM CUTOUT SW = CUTOUT" and/or "B/U STAB TRIM CUTOUT SW = CUTOUT"
* The FCC can't control the R64.
* The autopilot can't control the R64.

zeke wrote:
Are you suggesting the pilot, FCC, and autopilot cannot trim the stab with flaps down ?

No.

zeke wrote:
How do you trim for takeoff ?

"F/O STAB TRIM SW" and/or "CAPT STAB TRIM SW"
The limit of "STAB TRIM" "NOSE DOWN" action depend on "FLAPS DOWN"
I don't know if there is a special other condition before takeoff, as least this is not show on that schematic.

zeke wrote:
How is STS able to operate after takeoff.

Good question. There is a pilot that say that the STS actions are not cut by the "F/O COl CUTOUT SW MOD", so the STS signals can't pass through the "F/O COl CUTOUT SW MOD" and can't be the other "FCC" signals just at his left. If this is the "A" signal it need to carry at least 3 states [IDLE, ANP, AND]. This is technically possible with some very simple modulation, but I have yet to read a document that describes that scenario.

zeke wrote:
I have clearly said the way to control the stab is via A, controlling it means a circuit is completed and power is made available to the motor. It is the light switch in the room that turns the bulb on and off.

Sorry but the "A" signal on the schematic that goes from the FCC to the "A/P ENABLE SPEED & DIRECTION" inside the "ELECTRIC STAB TRIM MOTOR" can't replace the "115V AC XFR BUS2" to power the "ELECTRIC STAB TRIM MOTOR". No way. His exact function is largely unknown yet on this forum at least.
Last edited by PixelFlight on Wed Apr 10, 2019 8:45 pm, edited 2 times in total.
:stirthepot: 737-8 MAX: "For all speeds higher than 220 Kts and trim set at a value of 2.5 units, the difficulity level of turning the manual trim wheel was level A (trim wheel not movable)." :stirthepot:
 
freakyrat
Posts: 2195
Joined: Fri Aug 22, 2008 1:04 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 8:34 pm

Looks like we have more information from the Ethiopian FDR readout. The aircraft suffered a bird strike of some sort after takeoff that ripped the AOA sensor vane off of the aircraft. FDR data also indicated that the flight crew didn't totally follow the correct procedures for a runaway stabilizer.

My question for Boeing. Why didn't you build redundancy into the MCAS system? Surely the enginners should have thought of the possibility of a bird strike taking out the single sensor. The software should have also had a command in it to disable the MCAS if it lost sensor data. The FAA although delegating certification to Boeing should have been asking the same questions before thay even approved the aircraft.
 
User avatar
PixelFlight
Posts: 1030
Joined: Thu Nov 08, 2018 11:09 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 8:42 pm

fadecfault wrote:
Those of you arguing over the switch wiring need to know that diagram does not fully represent the actual system. The actual WDM needs to be seen to make any kind of factual statement.

I challenge anyone here to show a factual document that contradict the finding that can be made on this schematic.
Can't lost at this game: either the schematic is right, either we will learn something new.
:stirthepot: 737-8 MAX: "For all speeds higher than 220 Kts and trim set at a value of 2.5 units, the difficulity level of turning the manual trim wheel was level A (trim wheel not movable)." :stirthepot:
 
jollo
Posts: 398
Joined: Mon Aug 08, 2011 7:24 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 10:43 pm

freakyrat wrote:
mjoelnir wrote:
ikramerica wrote:
While I don't agree with a lot of what he is saying, there is still a case to be made that if airplanes were designed today, the controls would be very different. Same goes for cars.

The more we find out about the 737 system, the more we see how aircraft systems in general are designed so they "feel" like obsolete technology and use control mechanisms that mimic designs that should be outmoded. I'm not talking grandfathering exclusively. But the main paradigm of the feedback the aircraft gives is based on how much, much, much older aircraft with manual controls reacted and how they were controlled. Many of the paradigms go back to the piston era.

This happens because, even with a new type, the PILOTS are grandfathered in. To get experienced pilots, you need to have them already know how to fly the plane based on their experience. If you changed it up completely, you would have no one with experience to fly any of the planes you sell. Yes, you can train people, but getting enough people up to speed with enough experience would be astronomically costly and probably impossible.

So we have various aircraft of various vintages all trying to approximate how an older aircraft would behave just because. And as technology advances further and further, there will be more conflicts between how it was always done/needs to be done and how it really should be done.

The really ironic part is that as aircraft are more automated, the pilots are doing less actual flying anyway and too often forget what they are supposed to know in a panic situation because they barely ever use it, so completely changing the paradigm shouldn't matter as much.


Airbus broke with that tradition. The joystick has no feedback. Even pilots trained on the 737 seem to be able to switch over to the A320 and fly those frames.


I just a small aircraft GA pilot here who has some experience in an FDS A320 Flight Simulator. It took me about 20-30 minutes to get use to flying with a sidestick. It was quite easy after I got the hang of it.


[disclaimer]Just nitpicking here, and quite OT[/disclaimer] It's technically incorrect to say that A's joysticks have no feedback: they have, it's the return force by the stick spring. I know that your meaning is "the joystick gives no artificial feel proportional to aerodynamic loads", but it turns out that a linear feedback depending only on stick deflection is an adequate substitute (as thousands of Airbus pilots prove daily). It also automatically complies with 14CFR §25.203(a): as long as the control law requires an increasing pitch input approaching to stall, the control force gradient is linear (duh, it's a spring). Besides, hand flying with actual zero force feedback (i.e. without the centering spring) would be very, very difficult as you would have no tactile clue about the stick's neutral position (1 g load in pitch, 0 roll rate demand in normal law).
 
jollo
Posts: 398
Joined: Mon Aug 08, 2011 7:24 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 10:46 pm

freakyrat wrote:
My question for Boeing. Why didn't you build redundancy into the MCAS system? Surely the enginners should have thought of the possibility of a bird strike taking out the single sensor. The software should have also had a command in it to disable the MCAS if it lost sensor data. The FAA although delegating certification to Boeing should have been asking the same questions before thay even approved the aircraft.


Ah, the proverbial 280 million $ question...
 
mjoelnir
Posts: 9422
Joined: Sun Feb 03, 2013 11:06 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Wed Apr 10, 2019 11:17 pm

jollo wrote:
freakyrat wrote:
mjoelnir wrote:

Airbus broke with that tradition. The joystick has no feedback. Even pilots trained on the 737 seem to be able to switch over to the A320 and fly those frames.


I just a small aircraft GA pilot here who has some experience in an FDS A320 Flight Simulator. It took me about 20-30 minutes to get use to flying with a sidestick. It was quite easy after I got the hang of it.


[disclaimer]Just nitpicking here, and quite OT[/disclaimer] It's technically incorrect to say that A's joysticks have no feedback: they have, it's the return force by the stick spring. I know that your meaning is "the joystick gives no artificial feel proportional to aerodynamic loads", but it turns out that a linear feedback depending only on stick deflection is an adequate substitute (as thousands of Airbus pilots prove daily). It also automatically complies with 14CFR §25.203(a): as long as the control law requires an increasing pitch input approaching to stall, the control force gradient is linear (duh, it's a spring). Besides, hand flying with actual zero force feedback (i.e. without the centering spring) would be very, very difficult as you would have no tactile clue about the stick's neutral position (1 g load in pitch, 0 roll rate demand in normal law).


So we got to hair splitting what feedback means.
IMO feedback is feeding back to you an reaction to your input into a system.
So the spring load in a joystick is in no way a feedback. It is a spring load not connected to the input in the system and the reaction of the system to your input.
 
planecane
Posts: 1644
Joined: Thu Feb 09, 2017 4:58 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Thu Apr 11, 2019 12:11 am

I ask this question with ignorance on proper procedure. Why did they raise the flaps with the Captain's stick shaker going off?
 
User avatar
zeke
Posts: 15746
Joined: Thu Dec 14, 2006 1:42 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Thu Apr 11, 2019 1:08 am

Spotter1967 wrote:
Fully agree with Jollo here.
If someone states "A number of key statements made in that article are incorrect and conflict with the excerpts of Boeing material contained within." make sure to explain


The webpage that was referenced has just lifted material from pprune (eg https://www.pprune.org/rumours-news/614 ... ta-68.html ), and then we have people passing off their version of it on here as their own work. I’m not going to go around correcting it. The onus is on the people posting it to be factual in the first place.

Look at the manufactures QRH checklist on how the cutout switches are to be used on all 737 models, and then compare it with what was said on that webpage and on this thread.

PixelFlight wrote:
zeke wrote:
Tell us all under what conditions the R64 relay closes ?

"28V DC BUS 2 = nominal" & "STAB TRIM CONTROL fuse = good" & "PRI STAB TRIM CUTOUT SW = NORM" & "B/U STAB TRIM CUTOUT SW = NORM"

zeke wrote:
How does the pilot, FCC, and autopilot control the R64 relay if as you suggest that is the only way to remove power from the stab trim ?

* The pilot can control the R64 by moving "PRI STAB TRIM CUTOUT SW = CUTOUT" and/or "B/U STAB TRIM CUTOUT SW = CUTOUT"
* The FCC can't control the R64.
* The autopilot can't control the R64.

zeke wrote:
Are you suggesting the pilot, FCC, and autopilot cannot trim the stab with flaps down ?

No.

zeke wrote:
How do you trim for takeoff ?

"F/O STAB TRIM SW" and/or "CAPT STAB TRIM SW"
The limit of "STAB TRIM" "NOSE DOWN" action depend on "FLAPS DOWN"
I don't know if there is a special other condition before takeoff, as least this is not show on that schematic.

zeke wrote:
How is STS able to operate after takeoff.

Good question. There is a pilot that say that the STS actions are not cut by the "F/O COl CUTOUT SW MOD", so the STS signals can't pass through the "F/O COl CUTOUT SW MOD" and can't be the other "FCC" signals just at his left. If this is the "A" signal it need to carry at least 3 states [IDLE, ANP, AND]. This is technically possible with some very simple modulation, but I have yet to read a document that describes that scenario.

zeke wrote:
I have clearly said the way to control the stab is via A, controlling it means a circuit is completed and power is made available to the motor. It is the light switch in the room that turns the bulb on and off.

Sorry but the "A" signal on the schematic that goes from the FCC to the "A/P ENABLE SPEED & DIRECTION" inside the "ELECTRIC STAB TRIM MOTOR" can't replace the "115V AC XFR BUS2" to power the "ELECTRIC STAB TRIM MOTOR". No way. His exact function is largely unknown yet on this forum at least.


That post just demonstrates how much you don’t understand about how it works.

PixelFlight wrote:
I challenge anyone here to show a factual document that contradict the finding that can be made on this schematic.
Can't lost at this game: either the schematic is right, either we will learn something new.


That diagram is not a schematic, and fadecfault has told you which document contradicts what you are saying.
The onus is on you to post factually correct information in the first place, not on others going around correct your misinformation.

fadecfault wrote:
Those of you arguing over the switch wiring need to know that diagram does not fully represent the actual system. The actual WDM needs to be seen to make any kind of factual statement.


Exactly, but I guessing most of the people on this thread would not know what a WDM is let alone how to read it. They would need someone to post it elsewhere to be able to repost it on here and pass it off as their own work.
“Don't be a show-off. Never be too proud to turn back. There are old pilots and bold pilots, but no old, bold pilots.” E. Hamilton Lee, 1949
 
User avatar
zeke
Posts: 15746
Joined: Thu Dec 14, 2006 1:42 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Thu Apr 11, 2019 1:10 am

planecane wrote:
I ask this question with ignorance on proper procedure. Why did they raise the flaps with the Captain's stick shaker going off?


Because they probably looked at their pitch attitude, thrust, and airspeed to know that it was in error.
“Don't be a show-off. Never be too proud to turn back. There are old pilots and bold pilots, but no old, bold pilots.” E. Hamilton Lee, 1949
 
planecane
Posts: 1644
Joined: Thu Feb 09, 2017 4:58 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Thu Apr 11, 2019 1:51 am

zeke wrote:
planecane wrote:
I ask this question with ignorance on proper procedure. Why did they raise the flaps with the Captain's stick shaker going off?


Because they probably looked at their pitch attitude, thrust, and airspeed to know that it was in error.


That makes sense. My next question to you (since you are a pilot and I'm not) is that wouldn't they also know that the captain's stick shaker going off but not the FO's that it indicated an AoA disagree even without the AoA disagree warning light?

If they would have known that there was an AoA disagree and following the AD and publicity about MCAS following Lion Air that retracting the flaps would lead to MCAS activating?

I'm not asking these questions in defense of Boeing as there is no excuse for the way MCAS responds to a faulty AoA sensor. I'm asking to try and "get in the heads" of the pilots with the situation that they faced prior to it becoming an MCAS induced emergency.
 
User avatar
zeke
Posts: 15746
Joined: Thu Dec 14, 2006 1:42 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Thu Apr 11, 2019 2:50 am

planecane wrote:
That makes sense. My next question to you (since you are a pilot and I'm not) is that wouldn't they also know that the captain's stick shaker going off but not the FO's that it indicated an AoA disagree even without the AoA disagree warning light?


You raise a good point, stick shaker was not the only difference. The traces show airspeed, altitude and flight director pitch bar values from the left side were deviating from the right. This was similar on the Indonesian accident where the crew were saying they had no airspeed or altitude data ATC.

Personally if faced with such unresolved data so close to the ground I would not be changing configuration, I would use then maximum thrust available with a known good pitch attaitude to climb up above the platform altitude and then troubleshoot and clean up. Stick shaker just after liftoff would not be that unexpected. A fairly common scenario pilots would have thought about would be a bird strike during the takeoff roll impacting of sensors.

The other issue is how is the stick shaker command being generated. Is it coming from AOA probe directly, or the SYMD ? Why is there no SYMD in the MAX MEL like the NG, has the SYMD been removed on the MAX and now just another software application run within the FCC ? If it is within the FCC, where is the A to D conversion of the AOA being made, in the ADIRU or FCC ?
“Don't be a show-off. Never be too proud to turn back. There are old pilots and bold pilots, but no old, bold pilots.” E. Hamilton Lee, 1949
 
User avatar
seahawk
Posts: 10080
Joined: Fri May 27, 2005 1:29 am

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Thu Apr 11, 2019 5:24 am

And maximum thrust was their fault, as it let to an overspeed situation and that made the plane unable to recover.
 
User avatar
speedbored
Posts: 2230
Joined: Fri Jul 19, 2013 5:14 am

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Thu Apr 11, 2019 5:26 am

planecane wrote:
That makes sense. My next question to you (since you are a pilot and I'm not) is that wouldn't they also know that the captain's stick shaker going off but not the FO's that it indicated an AoA disagree even without the AoA disagree warning light?

One would hope that the fact that the captain's AoA display had gone all blue while the FO's was still ~40% brown would also be a clue.

zeke wrote:
Personally if faced with such unresolved data so close to the ground I would not be changing configuration, I would use then maximum thrust available with a known good pitch attaitude to climb up above the platform altitude and then troubleshoot and clean up.

Isn't this pretty much what the crew did? Things only really started going pear-shaped when they had to raise the flaps because they had reached their speed limit.
 
User avatar
speedbored
Posts: 2230
Joined: Fri Jul 19, 2013 5:14 am

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Thu Apr 11, 2019 5:35 am

seahawk wrote:
And maximum thrust was their fault, as it let to an overspeed situation and that made the plane unable to recover.

The very minor amount of overspeed (until the final dive) really shouldn't have been enough to render the aircraft uncontrollable.

In any case, there are multiple reports out there now of sim tests, with scenarios similar to what these pilots faced, that have shown that the control problems (primarily an inability to manually correct the adverse trim) would still have been present even if the aircraft had been below Vmo.
 
User avatar
zeke
Posts: 15746
Joined: Thu Dec 14, 2006 1:42 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Thu Apr 11, 2019 5:42 am

seahawk wrote:
And maximum thrust was their fault, as it let to an overspeed situation and that made the plane unable to recover.


I would maintain the configuration, thrust, and attitude until above the platform height (terrain is the bigger threat), after that you would set an attitude and thrust appropriate for level flight and clean up. The speed should not change that much if that sequence is followed.
“Don't be a show-off. Never be too proud to turn back. There are old pilots and bold pilots, but no old, bold pilots.” E. Hamilton Lee, 1949
 
Spotter1967
Posts: 11
Joined: Fri May 14, 2010 12:13 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Thu Apr 11, 2019 6:48 am

Zeke!
zeke wrote:
Spotter1967 wrote:
Fully agree with Jollo here.
If someone states "A number of key statements made in that article are incorrect and conflict with the excerpts of Boeing material contained within." make sure to explain


The webpage that was referenced has just lifted material from pprune (eg https://www.pprune.org/rumours-news/614 ... ta-68.html ), and then we have people passing off their version of it on here as their own work. I’m not going to go around correcting it. The onus is on the people posting it to be factual in the first place.



To summarize: you state key statements are incorrect. When asked to elaborate on that, your response is: "information was taken from other sites and I am not going to provide arguments why statements are incorrect. "
I leave it to the readers how they value your response. Indeed many sources, not just pprune but also Peter Lemme, Mentourpilot, airliners.net and Seattle Times were used to compile an overview of the many failures by Boeing, Nobody is claiming it is his own work. (which has nothing to do with factual incorrect btw)

https://feitoffake.wordpress.com/2019/0 ... g-737-max/

Peter Lemme, a former Boeing employee, was subpoenaed demanding he provide testimony and documents related to the 737 Max.
He states
The cutout switch function was changed with the 737MAX from all prior 737 models

Make sure to read his blogpost on ET302
https://www.satcom.guru/2019/04/what-ha ... et302.html
 
jollo
Posts: 398
Joined: Mon Aug 08, 2011 7:24 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Thu Apr 11, 2019 7:13 am

mjoelnir wrote:
jollo wrote:
freakyrat wrote:

I just a small aircraft GA pilot here who has some experience in an FDS A320 Flight Simulator. It took me about 20-30 minutes to get use to flying with a sidestick. It was quite easy after I got the hang of it.


[disclaimer]Just nitpicking here, and quite OT[/disclaimer] It's technically incorrect to say that A's joysticks have no feedback: they have, it's the return force by the stick spring. I know that your meaning is "the joystick gives no artificial feel proportional to aerodynamic loads", but it turns out that a linear feedback depending only on stick deflection is an adequate substitute (as thousands of Airbus pilots prove daily). It also automatically complies with 14CFR §25.203(a): as long as the control law requires an increasing pitch input approaching to stall, the control force gradient is linear (duh, it's a spring). Besides, hand flying with actual zero force feedback (i.e. without the centering spring) would be very, very difficult as you would have no tactile clue about the stick's neutral position (1 g load in pitch, 0 roll rate demand in normal law).


So we got to hair splitting what feedback means.
IMO feedback is feeding back to you an reaction to your input into a system.
So the spring load in a joystick is in no way a feedback. It is a spring load not connected to the input in the system and the reaction of the system to your input.


My disclaimer made it clear this is OT nitpicking ... but since we're splitting hairs:
  • "IMO feedback is feeding back to you an reaction to your input into a system.": by this definition the stick's spring load is a feedback.
  • "...it is a spring load not connected to the input in the system....": this bit is technically incorrect, as the return force of the spring is linearly proportional to inputs (stick deflection)
  • "...it is a spring load not connected to ... the reaction of the system to your input.": this bit is spot on.


The definition you're looking for, using your own words, is: "feedback is feeding back to you ... the reaction of the system to your input." By this definition I agree that A-style sidesticks give no feedback. Ok, enough hair-splitting, I apologize for the diversion, back to the topic...
 
jollo
Posts: 398
Joined: Mon Aug 08, 2011 7:24 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Thu Apr 11, 2019 8:27 am

zeke wrote:
Look at the manufactures QRH checklist on how the cutout switches are to be used on all 737 models, and then compare it with what was said on that webpage and on this thread.


Zeke, I am well aware of the fact that the 737's QRH only instructs to operate both stab trim cutout switches together - in the Runaway Stabilizer and the Stabilizer Trim Inoperative NNCs (see last sentence of post #4817 ).

The points here is not how the manufacturer supplied procedures tell pilots to use cutout switches, but how they actually work. And, if their function changed between the NG and the MAX, why Boeing felt the need to change it. I'm sure the basic engineering principle of "leave well enough alone" is well known to Boeing designers, and if they did change how the switches are wired there's a reason for it.

The fact that procedures never highlight the difference in function between the switches is what allowed Boeing to get away with presenting a functional change as a simple relabeling. I bet the reason there are two switches is historical to begin with, rooted in the distant past of the Jurassic and transmitted to us through a long series of "grandfatherings".

zeke wrote:
The other issue is how is the stick shaker command being generated. Is it coming from AOA probe directly, or the SYMD ? Why is there no SYMD in the MAX MEL like the NG, has the SYMD been removed on the MAX and now just another software application run within the FCC ? If it is within the FCC, where is the A to D conversion of the AOA being made, in the ADIRU or FCC ?


That's an excellent question, would love to get a genuine MAX wiring manual to try to work it out. Alas, I don't have an online reference: do you?

WARNING to members: beware of downloads! Many purported "free" ebooks out there are infected.
 
mjoelnir
Posts: 9422
Joined: Sun Feb 03, 2013 11:06 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Thu Apr 11, 2019 9:58 am

jollo wrote:
mjoelnir wrote:
jollo wrote:

[disclaimer]Just nitpicking here, and quite OT[/disclaimer] It's technically incorrect to say that A's joysticks have no feedback: they have, it's the return force by the stick spring. I know that your meaning is "the joystick gives no artificial feel proportional to aerodynamic loads", but it turns out that a linear feedback depending only on stick deflection is an adequate substitute (as thousands of Airbus pilots prove daily). It also automatically complies with 14CFR §25.203(a): as long as the control law requires an increasing pitch input approaching to stall, the control force gradient is linear (duh, it's a spring). Besides, hand flying with actual zero force feedback (i.e. without the centering spring) would be very, very difficult as you would have no tactile clue about the stick's neutral position (1 g load in pitch, 0 roll rate demand in normal law).


So we got to hair splitting what feedback means.
IMO feedback is feeding back to you an reaction to your input into a system.
So the spring load in a joystick is in no way a feedback. It is a spring load not connected to the input in the system and the reaction of the system to your input.


My disclaimer made it clear this is OT nitpicking ... but since we're splitting hairs:
  • "IMO feedback is feeding back to you an reaction to your input into a system.": by this definition the stick's spring load is a feedback.
  • "...it is a spring load not connected to the input in the system....": this bit is technically incorrect, as the return force of the spring is linearly proportional to inputs (stick deflection)
  • "...it is a spring load not connected to ... the reaction of the system to your input.": this bit is spot on.


The definition you're looking for, using your own words, is: "feedback is feeding back to you ... the reaction of the system to your input." By this definition I agree that A-style sidesticks give no feedback. Ok, enough hair-splitting, I apologize for the diversion, back to the topic...


A spring is not feedback. Airbus has been chastised for exactly having no feedback in the stick. It has been highlighted the A220/c-series system as more modern, partly because there the joystick provides a feed back.
Now you com and declare the that the Airbus standard joy stick gives a feedback.

You can go to a computer store, you can buy a joystick without feedback with sprigs returning it to zero or you can buy a joystick much more expensive with feedback, little motors or something to give you the feedback of the system.

You are out of whack with the industries definition of feedback.
 
User avatar
ExperimentalFTE
Posts: 52
Joined: Sun Mar 10, 2019 7:59 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Thu Apr 11, 2019 10:20 am

mjoelnir wrote:
jollo wrote:
mjoelnir wrote:

So we got to hair splitting what feedback means.
IMO feedback is feeding back to you an reaction to your input into a system.
So the spring load in a joystick is in no way a feedback. It is a spring load not connected to the input in the system and the reaction of the system to your input.


My disclaimer made it clear this is OT nitpicking ... but since we're splitting hairs:
  • "IMO feedback is feeding back to you an reaction to your input into a system.": by this definition the stick's spring load is a feedback.
  • "...it is a spring load not connected to the input in the system....": this bit is technically incorrect, as the return force of the spring is linearly proportional to inputs (stick deflection)
  • "...it is a spring load not connected to ... the reaction of the system to your input.": this bit is spot on.


The definition you're looking for, using your own words, is: "feedback is feeding back to you ... the reaction of the system to your input." By this definition I agree that A-style sidesticks give no feedback. Ok, enough hair-splitting, I apologize for the diversion, back to the topic...


A spring is not feedback. Airbus has been chastised for exactly having no feedback in the stick. It has been highlighted the A220/c-series system as more modern, partly because there the joystick provides a feed back.
Now you com and declare the that the Airbus standard joy stick gives a feedback.

You can go to a computer store, you can buy a joystick without feedback with sprigs returning it to zero or you can buy a joystick much more expensive with feedback, little motors or something to give you the feedback of the system.

You are out of whack with the industries definition of feedback.


C-Series / A220 does NOT have force feedback system, it has spring feel system. It works marvellously btw...
 
Spotter1967
Posts: 11
Joined: Fri May 14, 2010 12:13 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Thu Apr 11, 2019 10:22 am

I am not sure if this was explained in previous posts. Hope someone can explain:
At 05:40:35, the First-Officer called out “stab trim cut-out” two times. Captain agreed and First-Officer confirmed stab trim cut-out

From 05:40:42 to 05:43:11 (about two and a half minutes), the stabilizer position gradually moved in the AND direction from 2.3 units to 2.1 units

So the motor of the stabilzer was not operational anymore as result of cutout switches.
How come the stabilizer position moved AND? I believe this is only possible because one the pilots turned on the stabilzer trim wheel. Why nose down? The crew wanted to check if the stabilzer was stuck? Turning stabilizer AND requires less muscle force than ANU in that situation?
 
User avatar
speedbored
Posts: 2230
Joined: Fri Jul 19, 2013 5:14 am

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Thu Apr 11, 2019 10:53 am

Spotter1967 wrote:
I am not sure if this was explained in previous posts. Hope someone can explain:
At 05:40:35, the First-Officer called out “stab trim cut-out” two times. Captain agreed and First-Officer confirmed stab trim cut-out

From 05:40:42 to 05:43:11 (about two and a half minutes), the stabilizer position gradually moved in the AND direction from 2.3 units to 2.1 units

So the motor of the stabilzer was not operational anymore as result of cutout switches.
How come the stabilizer position moved AND? I believe this is only possible because one the pilots turned on the stabilzer trim wheel. Why nose down? The crew wanted to check if the stabilzer was stuck? Turning stabilizer AND requires less muscle force than ANU in that situation?

Given that the trim moved such a small amount, I suspect that this was due to the pilots trying very hard to use the manual wheel to move the trim ANU. It could be that every time they tried pushing the wheel in the ANU direction, and then released the pressure when nothing happened, the tension in the system caused a slight bounce back in the opposite direction, partly due to the very slight springiness inherent in even the best cables, and partly due to the forces on the elevator, which are opposing.ANU adjustments, but will assist movement in the AND direction.
 
michi
Posts: 266
Joined: Thu Jul 15, 2004 9:18 pm

Re: Ethiopian Airlines 737MAX crashes enroute to Nairobi

Thu Apr 11, 2019 10:59 am

speedbored wrote:
One would hope that the fact that the captain's AoA display had gone all blue while the FO's was still ~40% brown would also be a clue.


The display with the blue and brown is called attitude indicator (aka artificial horizon). It displays pitch and bank. However, this is not an AoA indicator!

https://en.wikipedia.org/wiki/Attitude_indicator

Angle of attack is the angle between where the nose points to (pitch attitude) and the flight path.

https://en.wikipedia.org/wiki/Angle_of_attack

For ET302 both CPTs and FOs attitude indicator should have displayed the same blue and brown colours.

Popular Searches On Airliners.net

Top Photos of Last:   24 Hours  •  48 Hours  •  7 Days  •  30 Days  •  180 Days  •  365 Days  •  All Time

Military Aircraft Every type from fighters to helicopters from air forces around the globe

Classic Airliners Props and jets from the good old days

Flight Decks Views from inside the cockpit

Aircraft Cabins Passenger cabin shots showing seat arrangements as well as cargo aircraft interior

Cargo Aircraft Pictures of great freighter aircraft

Government Aircraft Aircraft flying government officials

Helicopters Our large helicopter section. Both military and civil versions

Blimps / Airships Everything from the Goodyear blimp to the Zeppelin

Night Photos Beautiful shots taken while the sun is below the horizon

Accidents Accident, incident and crash related photos

Air to Air Photos taken by airborne photographers of airborne aircraft

Special Paint Schemes Aircraft painted in beautiful and original liveries

Airport Overviews Airport overviews from the air or ground

Tails and Winglets Tail and Winglet closeups with beautiful airline logos