Jump to content

PMDG 737-800 landing with autothrottle - yes or no?


Recommended Posts

Posted
3 minutes ago, Stearmandriver said:

Perhaps this went over your head, but there are things about the FDAS that shouldn't really be discussed in public.  😉

Or nudged at.

  • Replies 53
  • Created
  • Last Reply
Posted
49 minutes ago, Bobsk8 said:

The other issue is what happens if the single pilot becomes incapacitated for some reason? Will they have to hire flight attendants with MSFS experience ?   😉

With the number of sales chances are there's someone on the plane who used MSFS, I mean it can't be harder to land a real A320 than landing the Super Hornet in MSFS, can it? 😉

For transparency: I'm a community mentor at the BATC discord. However, I do not get paid for it in any way.

Posted

I think it will happen, but it's 40-70 years away, my guess about 50 years. So no worries for anyone here.

By that time, it won't matter, the image recognition and scenario processing will be so complete, that not only can it do everything a pilot can do, but it will be able to do it 50x more precisely and no com links will be necessary, it will be flawless in its operation. All emergency procedures will be perfect and executed instantaneously. It will be like an auto-pilot for an auto-pilot for an auto-pilot.

The most honest AI advocates which do a lot of lectures are pointing out that in 30-40 years, a computer will be able to interpret images and scenarios better than a human can. Hence, the vision of a computer will be far more accurate than the vision of a human, because a computer can process every single angle and view simultaneously. The computer will have a 360 degree of "every pixel" around the plane at all times, and be able to see upwards of 50 miles out with perfect accuracy, or in clouds even be able to tell things like spot birds a human could not spot.


 

AMD 5800x | Nvidia 3080 (12gb) | 64gb ram

Posted
On 11/5/2022 at 4:44 PM, Alpine Scenery said:

The most honest AI advocates which do a lot of lectures are pointing out that in 30-40 years, a computer will be able to interpret images and scenarios better than a human can. Hence, the vision of a computer will be far more accurate than the vision of a human, because a computer can process every single angle and view simultaneously. The computer will have a 360 degree of "every pixel" around the plane at all times, and be able to see upwards of 50 miles out with perfect accuracy, or in clouds even be able to tell things like spot birds a human could not spot.

But this is still just pattern recognition.  It's not judgement or decision making or learning on the fly, to work through an out-of-the-box scenario that the computer (or person) never received any training to be prepared for. 

Example: I once experienced triple blocked static lines in an Embraer jet, night IMC.  When we started our descent, all three mach/IAS indications started decelerating.  They were all erroneous, but they were triple independent systems and were ALL IN AGREEMENT.  The chances of this are literally astronomical, and this scenario is to this day not trained in any airline's unreliable airspeed scenarios.  If you ever reach a point in a checklist where you have 3 independent systems in agreement, that airspeed is to be considered reliable, the end. 

AI would have crashed that plane. 

Andrew Crowley

Posted
49 minutes ago, Stearmandriver said:

But this is still just pattern recognition.  It's not judgement or decision making or learning on the fly, to work through an out-of-the-box scenario that the computer (or person) never received any training to be prepared for. 

Example: I once experienced triple blocked static lines in an Embraer jet, night IMC.  When we started our descent, all three mach/IAS indications started decelerating.  They were all erroneous, but they were triple independent systems and were ALL IN AGREEMENT.  The chances of this are literally astronomical, and this scenario is to this day not trained in any airline's unreliable airspeed scenarios.  If you ever reach a point in a checklist where you have 3 independent systems in agreement, that airspeed is to be considered reliable, the end. 

AI would have crashed that plane. 

Whoa that's scary.  How did you ascertain your airspeed?  Did you use DME and the time it took to go a certain distance?  Or did you rely on some throttle setting or the A/T speed setting?

Rhett

7800X3D ♣ 32 GB G.Skill TridentZ  Gigabyte 4090  Crucial P5 Plus 2TB 

Posted
1 hour ago, Stearmandriver said:

AI would have crashed that plane. 

Yikes!

And that's why you deserve the big bucks and our appreciation for getting us safely from A to B, rather than onboard with 'Otto' Pilot (old joke, kids!)

 

56 minutes ago, Mace said:

Whoa that's scary.  How did you ascertain your airspeed?  Did you use DME and the time it took to go a certain distance?  Or did you rely on some throttle setting or the A/T speed setting?

Calculating time elapsed vs. any given distance is measuring groundspeed, not airspeed. Airflow over wings is the important criteria which varies depending on wind speed and direction. Certainly wouldn’t use any of the automatics as the data feeding the computers is incorrect.

Unreliable airspeed indication means using defined pitch and power settings for the aircraft (a memory item, I believe), until troubleshooting underway by reference to the aircraft's QRH.

AMD Ryzen 5800X3D; MSI RTX 3080 Ti ; 32GB Corsair 3200 MHz; ASUS VG35VQ 35" (3440 x 1440)
Fulcrum One yoke; Thrustmaster TCA Captain Pack Airbus edition; MFG Crosswind rudder pedals; miniCockpit FCU; CPFlight MCP 737; Logitech FIP x3; TrackIR

MSFS; Fenix A320; A2A PA-24; HPG H145; PMDG 737-600; AIG; RealTraffic; PSXTraffic; FSiPanel; REX AccuSeason Adv; FSDT GSX Pro; FS2Crew RAAS Pro; FS-ATC Chatter

Posted
55 minutes ago, Mace said:

How did you ascertain your airspeed?  Did you use DME and the time it took to go a certain distance?  Or did you rely on some throttle setting or the A/T speed setting?

Purely pitch and power, or realistically, pitch since we were in an idle descent.  Our assumption was that this was a frozen contaminant in the line issue and since we were descending into a warm midwest summer night, at least one system would unblock eventually - which did happen down around 10,000ft.  

The problem became recognizable by a wrong "feeling," and was then confirmed by experiencing mach buffet and a mild tuck when I disconnected the automation.  My brief on the way down was "now we know what too fast feels like.  A shaker is too slow. We stay between those things and we're flying."  Thankfully, all stall protection on that plane was purely AoA based and so there was no risk of experiencing unreliable stall protection activation. That's something I don't love about the 737.

Realistically, after many years and thousands of hours flying and teaching on that plane, I'll bet that once we got it corralled and just held correct attitude for an idle descent, we stayed within 20kts of desired speed. But a computer could never have handled the concept of disregarding 3 independent, concurring, but equally incorrect systems.  No pilot training covers that either as it's not supposed to be possible... but this is what a human brain can do that a computer cannot. We can learn and adapt - really, really fast when we need to. 

14 minutes ago, F737MAX said:

Calculating time elapsed vs. any given distance is measuring groundspeed, not airspeed. Airflow over wings is the important criteria which varies depending on wind speed and direction. Certainly wouldn’t use any of the automatics as the data feeding the computers is incorrect.

If any of you are timing and calculating in a scenario like this, you have a lot more available mental bandwidth than I do.   😉  Realistically, yes we had groundspeed indications as that comes directly from the GPS signals to the FMCs, but it doesn't do you a ton of good in the flight levels.  Groundspeed gets more useful down low and we did discuss flying a pitch/power/groundspeed approach if necessary but it thankfully was not. 

If I remeber right, we didn't have unreliable airspeed tables in our QRH at the time - certainly there were no memory items. The accident that drove these things was the Air France heavy airbus out of South America, and our event was probably within a year of that. I know I referenced it in my ASAP, but industry response hadn't caught up yet.  Today, absolutely all airlines have these tables. 

Andrew Crowley

Posted
On 11/4/2022 at 3:38 PM, Alpine Scenery said:

I just wonder how much longer there will even be a pilot at the helm, I guess it might take 50 years or so because people won't feel comfortable without a pilot.
 

Alexa, start engine #2.....

Only a matter of time......

Posted
On 11/3/2022 at 3:33 PM, regis9 said:

I’ve always disconnected the auto throttle at the same time as the AP on the 737.  I understand this is standard at most (but perhaps not all) 737 operators.  

That has always been my understanding too.

Howard
MSI Mag B650 Tomahawk MB, Ryzen7-7800X3D CPU@5ghz, Arctic AIO II 360 cooler, Nvidia RTX3090 GPU, 32gb DDR5@6000Mhz, SSD/2Tb+SSD/500Gb+OS, Corsair 1000W PSU, LG Ultragear 48"4K, MFG Crosswinds, TQ6 Throttle, Fulcrum One Yoke
My FlightSim YouTube Channel: https://www.youtube.com/@skyhigh776

Archived

This topic is now archived and is closed to further replies.

  • Tom Allensworth,
    Founder of AVSIM Online


  • Flight Simulation's Premier Resource!

    AVSIM is a free service to the flight simulation community. AVSIM is staffed completely by volunteers and all funds donated to AVSIM go directly back to supporting the community. Your donation here helps to pay our bandwidth costs, emergency funding, and other general costs that crop up from time to time. Thank you for your support!

    Click here for more information and to see all donations year to date.
×
×
  • Create New...