Flaw In Flight Dynamics?

[jimst57 1]

I saw a video on YouTube that claimed that there is a flaw in the flight dynamics of this DC-6 model.

His reasoning was that this plane has dihedral in the wing and should therefore roll back to level if placed in a bank and the controls are released.

He demonstrated in the video that this does not happened. The plane will remain in whatever bank that was established when the controls are released.

After watching the video, I tried it in hopes that it might have been an old video and this may have been fixed. But not so, the plane remained in the bank.

I know that there's someone of the forum from the flight dynamics team and I was wondering if there are any thoughts of this.

If it's a flaw, I would expect better from a PMDG model. 

Love the plane anyway, just thought I'd mention this.

 

[Chock 17,769]

I'd suggest the person making that claim is vastly over-simplifying things from an aerodynamic standpoint.

Whilst it is true that the dihedral effect will contribute to an aeroplane's roll stability, giving an aeroplane design some dihedral is not some magical choice a designer can make which will instantly mean his or her design will be self righting. All kinds of other things contribute to the dihedral effect, for example, the CoG's position, both vertically and horizontally will affect the amount of dihedral effect the wings produce because it will change the centre of lift position, which in turn alters the fulcrum point of the wings. The dihedral/anhedral angle of the horizontal stabilisers also affects matters considerably, the angle of which is usually chosen as a direct result of how little or much the designer wishes to either enhance, or subtract from the amount of dihedral effect the aeroplane generates from its wings. Then you've got gyroscopic precession from the props, which also contributes to matters, and of course the aerofoil cross section and wash out, which will also be throwing stuff into the mix, not in a big way, but they'll be part of it too. Then you've got wing sweep too, and guess what? Yup, that affects matters as well for how much dihedral effect is created in a sideslip.

In other words, it's not as simple as: 'give it some dihedral and it will self right'. If aeroplane design was that simple, every transport aeroplane you ever saw would have some dihedral, and they quite clearly don't, look at some of those big Russkie ones for example, they actually have anhedral.

Why might that be....?

Too much dihedral effect in a design is not as desirable a feature as it might seem; sure, you want some roll stability for a transport aeroplane so the passengers are comfortable, but too much dihedral will make an aeroplane prone to dutch roll as it continually tries to right itself, which is uncomfortable for passengers and can be dangerous if left to overdevelop. And you never see too much dihedral on a swept wing aeroplane because, as noted, the sweep enhances the effect anyway. But even then that's not the end of the story...

Dihedral is not always part of the design for aerodynamic reasons either, it can be purely to make it easier to land or take off, since it gives the wingtips more clearance. It also allows things to be slung under the outboard wing area, and when we consider that the DC-6 was originally a design aimed at the military, it's not an inconceivable notion to consider this was part of the design decision to cant the wings a bit, simply to make it better in a crosswind landing when banked over.

[randomTOTEN 13]

"However, the effect of the wing-fuselage-tail combination is significant since the vertical placement of the wing on the fuselage can greatly affect the stability of the combination. A wing located at the mid wing position will generally exhibit a dihedral effect no different from that of the wing alone. A low wing location on the fuselage may contribute an effect equivalent to 3 or 4 (degrees) of negative dihedral while a high wing location may contribute a positive dihedral of 2 or 3 (degrees)."

Aerodynamics for Naval Aviators, pg.298

Basically, there are a lot of factors which determine the lateral stability of an aircraft, and designers make choices to determine the roll response of an aircraft, and the stability it will exhibit. It's not nearly as simple (as Chock explains) as "no dihedral=no stability" and "dihedral=stability."

Look at the dihedral on many WW2 fighters, and consider their relatively neutral static stability.

Robert Toten

[Claudius_ 323]

7 hours ago, jimst57 said:

I saw a video on YouTube that claimed...

 

Is it possible for you to share this video with us? 

[jimst57 1]

1 hour ago, rampa said:

Is it possible for you to share this video with us? 

I'll have to look and see if I can find it again. If I find it I'll post a link.

Found it: The part in question starts at 1:00:15

 

[scandinavian13 5,192]

10 hours ago, jimst57 said:

I saw a video on YouTube that claimed that there is a flaw in the flight dynamics of this DC-6 model.

His reasoning was that this plane has dihedral in the wing and should therefore roll back to level if placed in a bank and the controls are released.

He demonstrated in the video that this does not happened. The plane will remain in whatever bank that was established when the controls are released.

After watching the video, I tried it in hopes that it might have been an old video and this may have been fixed. But not so, the plane remained in the bank.

I know that there's someone of the forum from the flight dynamics team and I was wondering if there are any thoughts of this.

If it's a flaw, I would expect better from a PMDG model. 

Love the plane anyway, just thought I'd mention this.

Welcome to the forum. Please note that full names are required to be placed in your posts here.

 

Issue #1: "I saw a video on YouTube [...]"
Keep in mind that there is a lot of misinformation out on the internet in general, but also on YouTube. Even when something sounds somewhat plausible, it is incredibly important to verify things. The issue with a lot of YouTube videos, particularly in the simming realm, is that a lot of simmers aren't real aviation professionals, or are private pilots with only cursory understandings of the real concepts...but have that burning desire to share and show off their knowledge without having come to fully understand it. Even some of the best pilots only have a simplistic understanding of actual flight physics, as that's more the realm of aerospace engineers for that true, full, understanding of what is going on. That isn't to say pilots don't know what dihedral is or does, but for someone to say that dihedral will make a plane automatically positively stable (roll back to wings level automatically in all cases), means that they misunderstand the concept. If dihedral equaled positive stability at all times, then someone please explain the C-5, C-17, An-124, and An-225 to me. Those should always be negatively stable and impossible to keep wings level, right? Well...no. As Alan mentioned, there are many more factors that go into stability - dihedral is only one of them. In the case of the four aircraft I listed, they all have anhedral to counteract the positive stability inherent in a heavy body, high wing aircraft, providing the end result of neutral stability. Moreover, this.

Issue #2: Blanket assignment of trust in a random unverified source versus a known source
I get that, if something sounds right, then it's human nature to believe it, particularly if it comes with a technical explanation. What somewhat baffles me, though, is that you specifically state "if it's a flaw, I would expect better from a PMDG model." This seems to point to your understanding, and recognition, that we create models that are top notch, given our track record, history, and verified knowledge of topics. All the same, despite all of that, based on a single YouTube video of someone who questioned the model because they misunderstand aerodynamics (and improperly asserted dihedral is essentially roll stability), there's an assertion of a flaw? Instead of taking a moment to consider the alternative (the guy is wrong), you simply jumped in the sim and noted that the plane behaved the same in your sim as it did in his sim. This is also somewhat of a logical flaw, as it does not actually test the theory:
Someone asserted A is false (incorrectly), because of behavior B. You verify that behavior B is present, which you assigned as a verification that A is actually false.
To use a parallel example:
Someone asserts that the DC-6 is modeled incorrectly because they see that carb ice doesn't accumulate when it's below zero outside, even when you're not using carb heat. You verify that the same behavior is present, which you assign as a verification that the assertion is correct.

...the problem is that the formula for ice is "cold + water = ice," and the person was only concentrating on the fact that it was cold. They flew through no visible moisture, in conditions where the moisture would already be frozen, but still asserted (incorrectly) that the air simply being cold will somehow ice a carb. Your verification simply showed that behavior B is the same in your sim as it is on that other person's sim. This does not prove or disprove the assertion of A, however.

Similar assertions are made about our 747 and 777 products in that they claim that the two products "bust" the 250 knot rule, but their claim is based on a huge misunderstanding of that rule (usually because they stopped reading at 91.117a). If one continues to read the full rule, they would easily see the reason the 747 and 777 do this is because of 91.117d (and there is a little bit of further discussion of the concept in the ATC handbook, the 7110.65 in section 5-7-2 Note 1).

 

Sorry for the novel, but stuff like this really gets me. If someone has knowledge that they want to volunteer, then they should be mindful that it is (reasonably) correct before doing so (acknowledging that nobody is always perfect). Aviation is complex. I completely understand that. I think it's awesome that people want to help other people understand how planes work, and how the system works. The problem, though, is that a lot of people in this hobby/community/industry seemingly share bit too carelessly, without fully understanding concepts, so there's a ton of misinformation for people to cut through. Dihedral isn't roll stability. The 250 knot rule isn't a hard rule. Carb/wing/engine ice doesn't just happen because it's cold. Autolands are rare. You don't always magenta line from departure to destination. I'm sure I could probably think of others, but I think I've made my point here. Trust but verify...particularly before making an assertion with them as a source.

[jimst57 1]

Thanks Scandinavian13 for the long answer. I didn't mean to make you type that much!

I understand perfectly what you said about taking things you find on the Internet with a grain of salt.

I guess what I was really asking, and may have worded it wrong, was "Is he right?" and was looking for someone to comment who had a better understanding.

Anyway thanks again.

Jim Stravato 

[Chock 17,769]

On 9/20/2017 at 0:47 PM, jimst57 said:

I guess what I was really asking, and may have worded it wrong, was "Is he right?" and was looking for someone to comment who had a better understanding.

Well, he is kind of right. A bit. The part about there being a different angle of attack of the wings when in a bank is certainly true, but if you look at the lift vector lines in the sim on his video, they don't show any influence on the lift vector from the horizontal stabiliser, so it's (quite literally) only half the story.

As you probably know, the main function of the horizontal stabiliser on an aeroplane (apart from being a place to put the elevators), is as its name suggests, to aid horizontal stability. Specifically, it does that by essentially being an 'upside down wing' which functions by forcing the tail downward to counteract the tendency for a wing to flip up from its trailing edge as it generates lift. But it has a secondary function, and that is, if you put a dihedral angle on it, it aids in stability. Does the DC-6 have any dihedral on its horizontal stabiliser? Nope, sure doesn't, and that means it will counteract the dihedral effect of the wings in a sideslip, and this is particularly true when one looks at how far back the DC-6's horizontal stabiliser is owing to its long rear fuselage, which further aids its effect on the wing as that distance means it produces a lot of leverage.

Compare this to for example, the Vickers Viscount, which is an airliner of a similar vintage to the DC-6. Google a picture of the Viscount and you will see it has a massive dihedral angle on the horizontal stabiliser, and that its horizontal stabiliser is approximately half the wingspan of the aircraft, and it has a big fillet on the vertical stabiliser too, this is largely because the Viscount is fairly short in length, so it needs that big horizontal stabiliser, and the big vertical stabiliser too, to be providing plenty of leverage and stability. Now look at the DC-6 again and observe that it too has a pretty large horizontal stabiliser, not quite the wingspan ratio of the Viscounts', but then again it is further back, so it doesn't need to be quite as big, also note that it is lower down in comparison to the location of the Viscount's stabiliser, which also affects its influence. From this you can determine that the horizontal stabiliser design and placement can have a big influence on the behaviour of an aeroplane; designers don't simply give an aeroplane's horizontal stabiliser anhedral or dihedral because they think it looks cool, it is to influence the wing's behaviour, which indeed it does.

Check out some pictures of the early prototypes of some aeroplanes and then look at later revisions of them. Most of the time you will see that it is the tailplane and horizontal stabilisers which undergo the biggest alterations, and this is usually as a result of flight testing for stability effects in level flight and in turns. The best example of this is probably the evolution of the Boeing B17 Flying Fortress, compare an early B17 model A, B or C's tail to that of a later Boeing B17 D, E, F or G variant and look at the size difference of the rudder and tailfin with its big long fillet along the top of the rear fuselage. This was done to make the aeroplane more stable at high altitude in thinner air so that it would be a better bombing platform when up at 20,000 feet or more. If that doesn't tell you that the tail of a plane affects its stability, then nothing will.

 

[skelsey 765]

The short answer is -- as Alan says (much more eloquently), there is no such thing as a free lunch when it comes to designing aeroplanes and many factors combine to produce the final result. Thus, it is basically impossible to know how the aeroplane will fly by looking at the wing design alone.

2 hours ago, scandinavian13 said:

...the problem is that the formula for ice is "cold + water = ice," and the person was only concentrating on the fact that it was cold. They flew through no visible moisture, in conditions where the moisture would already be frozen, but still asserted (incorrectly) that the air simply being cold will somehow ice a carb.

Off-topic, but as we're being precise...

It is quite important to recognise that "cold + visible moisture" is a good guide for identifying conditions conducive to airframe icing. Carb icing, however, can and does very much occur without visible moisture: the fundamental problem is that as the intake air accelerates through the carb venturi, its temperature drops (because its velocity increases and thus its pressure reduces) and therefore moisture contained within that air in the form of water vapour may condense out and freeze.

Thus it is entirely possible to experience severe carb icing on a completely clear day and at ambient temperatures well above zero -- even up to 25-30C. There's a graphic somewhere produced by the Australian CAA (IIRC) which essentially points out that the average, humid, northern European summer's day is absolute prime carb icing territory, even without a cloud in the sky! Ironically, the risk is probably lower on a cold, crisp winter's day when the relative humidity is lower. Likewise, I would expect carb icing to be less of a problem at high altitude (where I presume the DC-6 tends to spend a lot of time) where the air is much drier.

[scandinavian13 5,192]

1 hour ago, jimst57 said:

I didn't mean to make you type that much!

You didn't force me to. It's just a topic that I'm keenly aware of. I've been the person that has thrown information out there in the past because I thought I knew it all. I get it. It just frustrates me to see because I understand the motivations, but I also see how it can really negatively impact everyone's experience.

1 hour ago, jimst57 said:

I guess what I was really asking, and may have worded it wrong, was "Is he right?" and was looking for someone to comment who had a better understanding.

Not quite right, no, but that's not to say his understanding is entirely wrong. It's just somewhat misapplied (assertion that dihedral = roll stability is incorrect, along with some of the other things). Alan's post, above, explains it pretty well, though. Keep in mind that the DC-6 still flies, and we had a number of current, recent, and former 6 pilots (and mechanics, even) validate the performance. Someone with a YouTube channel may be knowledgeable, but when it comes to actual flight dynamic and feel, I'm going to go with the real pilots of the plane. Theories are all well and good, but reality is reality. If you've ever flown a plane before, and have looked at the charts, you've struggled with that difference many times (look at charted fuel burn versus actual sometime).

[downscc 1,327]

I have two cents to add to this. We had several DC6 pilots on board during technical and beta testing and things such as dynamic stability and impulse response were tested.  We also had a great developer on the team dedicated to the flight dynamics and in my opinion he gave us a simulation that is both very realistic and fun to fly.  Such mundane topics as adverse yaw and roll dynamics were often on the table.As I recall, the enthusiast behind those YouTube videos appeared early after the DC6 release and his opinions were dealt with and we haven't heard from him since. I can only conclude that his goals were to increase subscribership to his platform.

[jimst57 1]

Thanks for all the great answers guys.

 

Jim Stravato

[Claudius_ 323]

My only reference on Youtube is to see the real plane, but I didn't find one on this subject. Anyway the guy on youtube is talking about the DC6 xplane version, and in my opinion xplane in general has a very different "ambient" where the planes can exercise their dynamics. That DC6 in xplane seems flying on rails. Anyway I tried today to do the same with P3d4 and ASP4 in the background, the DC6 behaves in the same way, but not on "rails". Also, in general a plane doesn't roll back when banking because the centrifugal force and the other forces developped while the plane turns, this is a very interesting reading on the argument: http://design-plane.blogspot.it/2009/11/aircraft-forces-in-turns.html

I don't know how the real DC6 behaves when turning, but a plane is not a car, if you keep the yoke turning the plane will raise the turning angle drammatically, but this is depending by the meteo conditions also.

[metzgergva 28]

Well just my background as the developer on flight dynamics for the FSX/P3D versions. As Dan pointed out we had multiple pilots of old propliners on the testing team and I have been flying DC-3 and L-1049. The banking back to level is indeed dependent on many circumstances, just as mention on design and also CoG relative to lifting surfaces. For normal bank angles we had consensus that I design it to keep the bank in the "unnatural" situation of no weather at standard load. That's what I did and depending on speed and angle of attack you may see a very small tendency t level back.

Wing dihedral is very important for low wing construction to keep the plane stable in straight and level flight and avoid that any little difference on lift from the wings would make her instable and would require pilot corrections. Just as said in Aerodynamics for Naval Aviators, pg.298

So I can assure you she flies like the real thing within the limitations of a simulator.