EFB and A/P correct functionality with the LNAV for PMDG in

[Bobsk8 12,879]

6 minutes ago, threegreen said:

I didn't tell you how to fly an approach though.

But even PMDG have acknowledged the issue with the wonkiness so I'm not sure what to make of this unless there's different versions of the addon going around. Any chance you're on a version older than the LNAV update?

I don't see any problem either, so I guess my 45 years as a pilot isn't enough..😉

[mryan75 281]

1 hour ago, threegreen said:

I didn't tell you how to fly an approach though.

But even PMDG have acknowledged the issue with the wonkiness so I'm not sure what to make of this unless there's different versions of the addon going around. Any chance you're on a version older than the LNAV update?

I mean yes they fly a series of straight lines, but that’s exactly how you fly a DME arc in real life. So I don’t have a big issue with it. As far as missing turns and wagging the wings in turns, I saw it in the past and I don’t see it anymore. Improvements can be made and will come but the majority of the issues with flying the PMDG is the person seated in the left seat in my opinion. 

[threexgreen 3,160]

1 hour ago, mryan75 said:

I mean yes they fly a series of straight lines, but that’s exactly how you fly a DME arc in real life. So I don’t have a big issue with it. As far as missing turns and wagging the wings in turns, I saw it in the past and I don’t see it anymore. Improvements can be made and will come but the majority of the issues with flying the PMDG is the person seated in the left seat in my opinion. 

Don't take this as an "attack" on your knowledge as a pilot but since when is a DME arc or RF leg a series of straight lines?

I don't want to be controversial, but how can it be that a known issue like wagging the wings in many turns disappears for some and not for others? It's how PMDG built their simulation of the feature, and that's the same for everyone unless there's different versions flying around. I also don't know what the person in the left seat has to do with how LNAV flies the plane in a turn, with a flight director that's erratic at times or even a lack of RF leg capability.

Edited December 5, 2023 by threegreen

[SAS443 1,731]

4 minutes ago, threegreen said:

Don't take this as an "attack" on your knowledge as a pilot but since when is a DME arc or RF leg a series of straight lines

I think he means when you fly the arc on raw data. You know "twist ten, turn ten" methodology. 
 

[mryan75 281]

6 hours ago, threegreen said:

Don't take this as an "attack" on your knowledge as a pilot but since when is a DME arc or RF leg a series of straight lines?

I don't want to be controversial, but how can it be that a known issue like wagging the wings in many turns disappears for some and not for others? It's how PMDG built their simulation of the feature, and that's the same for everyone unless there's different versions flying around. I also don't know what the person in the left seat has to do with how LNAV flies the plane in a turn, with a flight director that's erratic at times or even a lack of RF leg capability.

I take no offense. Hand-flying a DME arc is actually a series of 10-degree turns in succession. You essentially set a radial in front of yourself, fly a perpendicular heading to that radial, hit it (center the needle), turn the OBS ten degrees, turn your plane ten degrees, fly until you hit the radial again, turn ten, twist ten, so on and so forth. A RTF is indeed different but the difference in how we’re talking about it is really indistinguishable from the real thing. We’re just using a little trick to get the same result. Any time I’ve had an RNP approach go sideways in the sim it was my fault, not the plane’s. 

[threexgreen 3,160]

9 hours ago, mryan75 said:

I take no offense. Hand-flying a DME arc is actually a series of 10-degree turns in succession. You essentially set a radial in front of yourself, fly a perpendicular heading to that radial, hit it (center the needle), turn the OBS ten degrees, turn your plane ten degrees, fly until you hit the radial again, turn ten, twist ten, so on and so forth. A RTF is indeed different but the difference in how we’re talking about it is really indistinguishable from the real thing. We’re just using a little trick to get the same result. Any time I’ve had an RNP approach go sideways in the sim it was my fault, not the plane’s. 

So essentially, what PMDG are doing is akin to the "twist ten, turn ten" technique by creating pseudo waypoints on the arc or radius and fly direct to between them, as in a GA aircraft for example. A 737 however flies it like an actual arc.

As for an RNP, the plane used to be bad at following its own magenta line, so it wasn't able to comply with RNP at times on an approach because even though ANP in numbers showed as smaller than RNP you could see on the map it was sometimes greater than RNP because it had trouble following the line closely. It does this much better now. However, I think it was with this update that the wing wagging started. That was definitely not there before. I've found it to be more noticeble in cruise than on an approach.

Edited December 6, 2023 by threegreen

[LRBS 953]

The main issue here is that some people are entirely unfamiliar with how this (other) airplane actually flies or any of these modes (A/P, LNAV, VNAV, etc.)

It's understandable and expected for a desktop pilot not to understand systems operations fully; for us (the real pilots), when we do transitions or a new type rating, it takes at least 1 ½ month 

of intensive training with all high-tech ground schools and simulators, and after we finish the process and start IOE, we still discover/learn stuff. What is puzzling is this persistent denial of the fact that this PMDG airplane behavior needs

a serious fix. PMDG as an entity and all the supporters need to understand that in real life, "we use LNAV/VNAV for about 95% of the total flight duration." and this product is very messy. Actually, this type of

problem will get entered into the maintenance book, and A/P or FMC might get deferred under MEL/DDG and the airplane can be grounded.

While this is a "game," we are not talking about any safety; or how much we paid for; it's just that it is not correct, and if we compare it to the P3D version, we didn't have these issues of this magnitude.

Then there are lame excuses/pretexts from the PMDG of this nature "“For 99% of users the change will be practically invisible.”, well times have changed, people see this, and is not a good business practice to ignore reality, and keep on developing products after products with the same bugs.    

[Stearmandriver 2,333]

22 hours ago, mryan75 said:

I take no offense. Hand-flying a DME arc is actually a series of 10-degree turns in succession. You essentially set a radial in front of yourself, fly a perpendicular heading to that radial, hit it (center the needle), turn the OBS ten degrees, turn your plane ten degrees, fly until you hit the radial again, turn ten, twist ten, so on and so forth. A RTF is indeed different but the difference in how we’re talking about it is really indistinguishable from the real thing. We’re just using a little trick to get the same result. Any time I’ve had an RNP approach go sideways in the sim it was my fault, not the plane’s. 

Hand flying a DME arc in a real 737 is nothing at all like what you're describing, is the point.  In reality, an arc is drawn as an smooth, single-radius RF segment.  The flight guidance tracks this smooth continuous curve perfectly.  It commands one smooth roll into a turn, maintains the bank angle necessary, and commands one smooth roll out.  When handflying, this is what you follow.  You absolutely do not make a series of 10 degree segmented turns like you would in a bug smasher.  

PMDG's LNAV flight guidance is definitely not correct, and this is definitely not open to interpretation or opinion - it's demonstrable.  Yes, it will get you there in the end, but it's very sloppy compared to the real airplane.  Anyone on this thread that thinks it works correctly, feel free to post a video of you flying a complex LNAV procedure in the PMDG that you think went well, and I guarantee you that myself and others here can tell you all the things wrong with it.  

[Stearmandriver 2,333]

On 12/2/2023 at 7:23 AM, LRBS said:

On the EFB, one of the standard worldwide requirements is to have an accelerate-stop distance calculation information. 

I will point out that this is not necessarily true.  Accelerate go and accelerate stop (and many other factors) are all accounted for correctly in a proper takeoff data solution that is specific to runway, conditions, and weight.  At work, I never see accelerate-stop or accelerate-go.  I just use the takeoff data given: thrust setting, flaps, bleeds on/off, and V speeds.  Abort for anything below 80kts, high-speed abort for fire, failure, fear, or shear prior to V1, go afterwards.  That's all you need to know is V1, assuming you have well-calculated data.  In fact there is a school of thought that says it's better not to provide specific margins to pilots because it may bias them to abort or go, and we really just want them to fly standard and use the V1 call as decision point.

So it's not really unrealistic for PMDG to not provide these distances.

[LRBS 953]

1 hour ago, Stearmandriver said:

I will point out that this is not necessarily true.  Accelerate go and accelerate stop (and many other factors) are all accounted for correctly in a proper takeoff data solution that is specific to runway, conditions, and weight.  At work, I never see accelerate-stop or accelerate-go.  I just use the takeoff data given: thrust setting, flaps, bleeds on/off, and V speeds.  Abort for anything below 80kts, high-speed abort for fire, failure, fear, or shear prior to V1, go afterwards.  That's all you need to know is V1, assuming you have well-calculated data.  In fact there is a school of thought that says it's better not to provide specific margins to pilots because it may bias them to abort or go, and we really just want them to fly standard and use the V1 call as decision point.

So it's not really unrealistic for PMDG to not provide these distances.

Probably so. When I used to fly the 737, we had a laptop similar to EFB. The FAA had an issue with the unit because it didn't show the accelerate stop distance. Today, on other airplanes, other SOPs, and on EFB, we also have that data available. Besides the information provided by the company, we can also utilize EFB on our aircraft to calculate data.  When at performance limit or max t.o. weights, we are specifically informed of the stop distance in purpose for the pilots to know. If we find that the acc stop dist is less than 500FT and we are not comfortable (even legal) our SOP allows us to use a few other options, another assumed temperature, another rating, other flaps setting, packs off, etc. 

[Stearmandriver 2,333]

Sure, you can always manually define if you want data for a specific flap or bleed setting or thrust rating.  The idea is just that once you have this data, margins don't matter - just fly a standard profile.

[LRBS 953]

26 minutes ago, Stearmandriver said:

Sure, you can always manually define if you want data for a specific flap or bleed setting or thrust rating.  The idea is just that once you have this data, margins don't matter - just fly a standard profile.

Hi, I'm sorry for the confusion. I'm referring to just before V1, not after. After V1, I completely agree with you.

[Stearmandriver 2,333]

38 minutes ago, LRBS said:

Hi, I'm sorry for the confusion. I'm referring to just before V1, not after. After V1, I completely agree with you.

I'm with you.  I'm just describing the other school of thought, which I do see merit in after years as a check airman: we'd really rather not have pilots evaluating an abort decision on a case by case basis.  That's a recipe for delayed reaction times, fatigue-induced indecision or confusion on a multi leg day etc.  There's value in having the decision be basically automatic: you're rejecting for these items within this speed regime, and you're continuing for anything else, and at V1 you're going for everything, period.  Every takeoff, without exception.  Adhering to those standards, margin is irrelevant.

Now the flip side is, I do see the value in using margin as a determining factor as to whether you'd choose to request more conservative data.  The flip side to THAT is, if your airline has default required margins set to a comfortable value, you won't need to... Can be argued either way but I think I fall in the camp of not wanting to bias a captain's abort decision, but just keeping it standard.

Of course in MSFS, it's irrelevant.  I was just saying that it's not necessarily unrealistic that the EFB not provide these margins, as many real world operations intentionally do not.

Also though,I thought margin was graphically displayed on the runway graphic at the bottom of the tablet screen?  I've only tried it a few times so not sure I'm remembering right though.

[mryan75 281]

4 hours ago, Stearmandriver said:

Hand flying a DME arc in a real 737 is nothing at all like what you're describing, is the point. 

I wasn’t talking about hand-flying a 737, I was describing doing it in a Cherokee. 

[Stearmandriver 2,333]

2 hours ago, mryan75 said:

I wasn’t talking about hand-flying a 737, I was describing doing it in a Cherokee. 

Yeah I get that; it's just that you only said "handflying a DME arc" without specifying aircraft type as if it was done that way in all airplanes.  Just pointing out that it's done differently in the plane the thread is about.  ;)

Handflying any LNAV procedure in the PMDG is less than comfortable these days because of the way the lateral guidance fluctuates.  It's quite a separation from reality.