Problems with the 3D model of the wings

[gilbertohasnofb]

The 3D model of the wings need some rework:

1) Glass reflection colliding with vent:

2) The wings are too thin, compare the photograph below with our plane:

3) Our wings are missing the drooped tips ("inverted winglets"), see:

[gilbertohasnofb]

@wlbragg do you think you could also have a look on this since you were fixing the last reflection problems? Look on the image above how the reflection glass is cutting through the middle of the vent cylinder

[wlbragg]

I don't see it? I hate to ask but could you point it out in the picture what your talking about or describe it like your talking to a two year old? :) The vent should have shadows go across it, right? But it shouldn't have glass reflections on it, only behind it, right? I don't see any reflections on the vent?

[gilbertohasnofb]

[the vent] shouldn't have glass reflections on it, only behind it, right?

That's exactly the problem. Look, I uploaded the same image as before, but I marked in red the area where I see the problem. I also made a drawing in the bottom of what I think that is happening: it's a side view of the vent showing where the glass is (cutting through it). Let me know if this makes sense: http://s22.postimg.org/a8fgbgsdd/Untitled.png

[wlbragg]

The layer of glass that has the reflection on it is "glas_interior_sides". The layer of glass that has shadow is "glas". Look at the pictures and see where and how the vent is placed.

glas_interior_sides

glas

So I guess you actually noticed that part of the vent opening would be outside the cockpit glass?

I'm not sure what is better or easier, shorten the vent opening so it doesn't extend beyond the glass or move the glass.

Are we looking at the same thing?

[gilbertohasnofb]

Yes, I believe we are looking at the same thing. In fact, if you look closely on this picture below you will see that there is no glass above this part where the vent is located at all (this part where the wing meets the body): http://www.cruik.org/blog/ventubes/ventube-out.jpg

Better would be to edit the glass so it is not above this part of the wing, another option is to move the glass. I think that moving the vent cylinder so that it is behind the interior glass is a bad idea, because it will be overly short.

[wlbragg]

Oh, wow, that is crystal clear how it is suppose to be. MODELERS, HEEEELP! I'm not sure everyone reads these issues. Do we need to state this problem in the thread to get help? There is a possibility I could remodel the glass, but it would probably take me an enormous amount of trial and error and time to figure it out.

[gilbertohasnofb]

I can post this in the thread, no problems, but who is doing the 3D modelling right now? After @thevirtualfer started having those problems with his internet I thought we were only doing some basic modifications to the 3D model -- and I had the impression that you were the person who knew how to do these things better! :)

[thevirtualfer]

The resolution of my net problem was delayed to June, 21.

And my contract forbids me to cancel the service.

Sorry.

2015-04-20 2:37 GMT-03:00 Gilberto Agostinho notifications@github.com:

I can post this in the thread, no problems, but who is doing the 3D modelling right now? After @thevirtualfer https://github.com/thevirtualfer started having those problems with his internet I thought we were only doing some basic modifications to the 3D model -- and I had the impression that you were the person who knew how to do these things better! :)

— Reply to this email directly or view it on GitHub https://github.com/Juanvvc/c172p-detailed/issues/118#issuecomment-94360737 .

[Juanvvc]

Yes, this issue was in my checklist from the moment we duplicated the glass.

I was waiting for the glass effect to be more or less stable. The lateral angles must be smoother, and the layers of the glass should be probably closer together. I think the glass needs to be modelled again.

[gilbertohasnofb]

@thevirtualfer Oh, I am sorry to hear that! :(

[wlbragg]

OK, I guess we hold tight on this one for now. How come we always seem to make more issues than we ever fix? :(

[gilbertohasnofb]

@wlbragg well, it's true we still have tons of things to fix and improve, but my impression is that we did so much work already and fixed some really big stuff too! Also, we have been without our 3D modeller for long time since thevirtuafer started having those problems, and that really made the progress slow down. Most of the opened issues revolve around the 3D model.

[gilbertohasnofb]

@Juanvvc

I was waiting for the glass effect to be more or less stable. The lateral angles must be smoother, and the layers of the glass should be probably closer together

I was messing around with our model in Blender and I found out that the distance between the two layers of glass is actually given by the thickness of the fuselage, which is really thick (see: https://github.com/Juanvvc/c172p-detailed/issues/6). Fixing one would involve fixing the other, I guess.

[wlbragg]

"How come we always seem to make more issues than we ever fix? :(" FYI, sorry, this was more of a joke. I am elated at the attention to detail and perfection we all appear to posses.

Hopefully thevirtuafer and ludomotico can get to some of this. Maybe someday I might even have the skill set to do it. I have learned so much because of this project.

[gilbertohasnofb]

Yeah, I also tried learning a bit of Blender to see if I could fix some of these 3D issues myself (that thick cockpit "wall" above the panel drives me crazy, I think it looks so ugly and I spend 99% of my time in the cockpit view of course), but the learning curve is just too high for me. At the moment I just can't invest that amount of time on it :( Maybe in a month or so I will be able to really try to learn it properly.

[gilbertohasnofb]

Could we maybe assign the milestone FG 3.8 feature freeze for this one? Since we plan to rework the nose (see #6 and #216), we could as well work on the glass.

[onox]

I started working on the glass in the morning, but I found the leading edge to be quite crappy, so before I could properly fix the glass, I have been trying to smooth the leading edge first. I used a subdivision modifier in Blender, and now it looks like this (look at the leading edge):

Leading edge is a tiny bit shorter now, I can try to move it forward a bit if you guys think it's too much. Still need to check and fix the wing tips (they got rounded too by the modifier).

Also, it's now a separate object, so we can attach that grain effect for splashed bugs.

I also found that in the current model there's some weird rectangle above the pilot:

I also found some weird object, Plane.010 or something. It seems to be unused. I think we should move this issue to 2016.4 so we have more time to properly fix the glass and everything. Then we can also properly shift the whole panel as well.

[gilbertohasnofb]

@onox the leading edge looks much better! :+1: Maybe it could be even shorter a bit, see:

As for the weird rectangles, I can't see what you mean in that picture.

I think we should move this issue to 2016.4 so we have more time to properly fix the glass and everything. Then we can also properly shift the whole panel as well.

Absolutely agree, that should probably be better and having time to test and fix bugs is really fundamental. So since these changes can be quite big, let's move this to 2016.4.

[gilbertohasnofb]

@onox I also think our wings are maybe to thin, take a look at this picture:

Notice how the bottom side of the wing goes almost all the way down to the window (see red line), and in green I marked how I believe our current 3D model looks like. Now take a look from the outside:

Notice how the distance between the top side of the door (marked in green) and the wing is much larger in our model than in the RL plane. So my guess is that we have to do two things here:

• increase the thickness of the fron part of the wing, keeping the back side the same and keeping the wing shape from a top view
• make the front section of the wing (the one visible from the cockpit) rounder, as seen in these photographs above. Ours look too pointy

[gilbertohasnofb]

One more thing about our wings which could be tackled in this same issue, I believe the Cessna 172 has a very small "inverted winglets", seee:

@tigert @Juanvvc could some of you guys confirm this feature is present in the model P as well? It's difficult to tell from most angles, but I believe I can see it in the images of the 172P I have seen.

[wkitty42]

i thought those inverted winglets were addons? that's what i seem to recall from previous research a long while back when i was looking at suggesting winglets be added... i never did make that suggestion because of the addon results... i do know that they affect the efficiency of the craft by reducing vortex drag from the wing tips... not sure how the FDM would be told they exist on the craft so that handling and flight time are affected properly...

[wlbragg]

. i do know that they affect the efficiency of the craft by reducing vortex drag from the wing tips... not sure how the FDM would be told they exist on the craft so that handling and flight time are affected properly...

I'm not an FDM guy but I think it should be doable in the FDM with some adjusting of coefficient's, same as we do for gear changes.

[gilbertohasnofb]

I am not sure those are really addons, I don't remember seeing a picture of a Cessna 172P without them. @wkitty42 could you please try to find this out?

[wkitty42]

On 11/19/2016 01:06 PM, Gilberto Agostinho wrote:

I am not sure those are really addons, I don't remember seeing a picture of a Cessna 172P without them. @wkitty42 https://github.com/wkitty42 could you please try to find this out?

i will try but i don't know that i can duplicate the research i did a year ago when i first got involved with the project...

[time passes]

here's some links i found on a quick uncle google search for "cessna 172p winglets" (without the quotes)...

http://www.diva-portal.se/smash/get/diva2:751994/FULLTEXT01.pdf

here's one that says that the C172's tips are "drooped tips", not winglets... http://www.airliners.net/forum/viewtopic.php?t=741015

that's really about all i have right now... you can follow the $$$ here https://www.google.com/search?q=cessna+172p+winglets

it has been so long that i really don't recall all the various searches i did at that time... the end result, though, was that it was nothing for this craft... the drooped tips are fairly common on the 172s but i don't know about this particular model other than as addons (aka tip replacements)...

[gilbertohasnofb]

Well, as I said I never seen a 172 P without those inverted winglets on its wings and until you wrote about it I just assumed it is a factory design. Also, I find it a bit hard to believe that such an addon would be so widespread that virtually every photograph shows them! Would really nearly all owners buy and install such a thing?

As for your links:

• the first is a theoretical study on the effect of winglets (not inverted) on a Cessna 172. I believe the author selected a 172 for no particular reason, or maybe because he had data to compare his simulation results.
• the second is quite useful, apparently the term is indeed "drooped tips" and so it should be easier to research about it if anyone feels like

As far as I can tell, there are kits with bigger drooped wing tips which one can buy (e.g. for STOL variations), but it seems to me that all models of 172 produced after the early 1970s do have the small ones I showed on the photographs above.

[gilbertohasnofb]

I have searched for images of the Cessna 172P in a databse of aviation photographs (www.airliners.net) and absolutely all Model P have these drooped wings, e.g.:

I will update the issue title and description to better reflect everything that should be done here (glass colliding with vent, wing thickness and drooped tip)

[gilbertohasnofb]

Just adding one more pic from a different angle showing the issue with the cockpit glass shape:

[wlbragg]

A simple proof of concept here, I thickened the wing a tad bit. Notice I didn't move or adjust any of the peripherals, just a quick and dirty. With a little more effort I probably could make this look really nice by adding some edges to make it more rounded. I'm not sure I want to tackle it although it isn't really that much work. Except, we would have to reposition the tie-downs a tad bit and even this little bit of thickening is impeding the doors from opening. Also the taxi and landing light would be slightly affected. The vents would probably require repositioning. And while we were at it I think we would have to do the wingtips.

[gilbertohasnofb]

@wlbragg this is already looking so much better, many thanks for taking yet one more issue into this huge PR! If you feel like making the leading edge more round (i.e. more polygons) then we could possibly close this together with all other issues, though I understand that you might not want to tackle all this at once.

And while we were at it I think we would have to do the wingtips.

Yeah, I was just going to add that our wingtips aren't that great looking.

It's up to you, but if you feel like doing it then I would be happy to help with whatever I can 😄

[wlbragg]

OK, guess were doing the wings!

vents were moved... x = 0.01828 z = -0.01825 from their original positions.

TiedownHotSpotRight and TieDownRight were moved... y = -0.06198 z = -0.03995

TiedownHotSpotLeft and TieDownLeft were moved... y = 0.06112 z = -0.03807

[wlbragg]

Nothing else requires repositioning so far. I'm going to check the taxi and landing light and see if we can leave it as is. In my opinion it is OCD to move it, but we can if it really looks wrong.

[wlbragg]

I went ahead and repositioned the pitot and cover, oat and vents, tie-downs. The tie-downs are repositioned and rotated in such a way it is more realistic as the strap now sits more centered and where you might expect it to be with tension. Plus the added benefit that we didn't have to reposition the straps.

I fixed the taxi and landing bar light so is now visible. I had to comment out a section that was actually eliminating it being seen except when the lights were on. Now it behaves as before except you can also see it when the lights are off. Not sure the effect on Rembrandt.

The trim wheel needed to be removed from the interior effect or when you rotate it would start to go black (shadow) and back to light colored. It looked like crap. Now it has no shadow or interior lighting effects but that is the price we have to pay I guess. If anyone wants to take a look at it, please do so.

The wings are finished except the tips which I may still do. That is going to be a bit of a job so well see. Just in case, are they supposed to look like this...

[wlbragg]

For sure our landing/taxi light is not in the wingtip like this, but what about the shape and pointy end?

[wlbragg]

OK, I pulled up a bunch of reference pictures, i think I know what we need to do. I decided to do this. It's going to take a bit though because these wingtips are really dorked up, like the tail was. I'm actually surprised they look as good as they do in the sim because they are so screwed up. There are vertices overlapping and all kinds of issues. I have to delete the end face and start over after reshaping the new end of the wings. Then I can taper it to match the reference images I have. FYI: I am aware of the "dent" or anomaly in the top of the port wing starting from the landing light recess and extending to the wing tip. It is caused be vertices and faces that are refracting light differently because they are cut and positioned different that the rest of the wing and also different from the starboard wing.

[gilbertohasnofb]

I decided to do this.

Great, many thanks! I am really looking forward to this one!

It's going to take a bit though because these wingtips are really dorked up, like the tail was.

No rush, please take your time and let me know if I can help with anything other than testing.

[onox]

@wlbragg Do you plan to rework the leading edge as well? Make it a separate object for #737 ?

[wlbragg]

Sure, let me take a look a little later tonight and see what it will take to separate it into its own object for the effect.

[wlbragg]

@onox the leading edge of the wing is now a separate object so when someone is ready to do the dirt effect it is ready to go. I think I named it wing-leading-edge.

[gilbertohasnofb]

@wlbragg about the smoothness of the air vent tubes, compare my first image from 2015 to your latest one:

It really looks to me we lost the smoothness somewhere along the way.

[wlbragg]

Yep, your right once again. I just spent about 2 hours testing the smooth function and the auto smooth function in Blender until I finally figured out how they work. Please check the entire aircraft out carefully and see if there is anything that looks out of place. Hopefully you'll notice some improvements (I did get the vents smoothed like they were).

Things to pay special attention to...

Switches Knobs Radio stack vertical stabilizer rudder fuselage wing flaps ailerons

For example... There may be some objects that need to have sharper edges, like the switches. The top face of the wings where the fuel caps are had to have the auto smooth angle turned down to less than 90 deg or the recess for the caps got all screwed up. The fuselage had to have the auto smooth angle turned down to 60 or the step in front of the wing spar was messed up.

The picture below compares c172 in fgdata (left) and c172 one in github (right). If you look closely you can see the difference between the un-smoothed switches in fgdata (left) and the smoothed switches in github (right). Please let me know what changes you'd like to see if any?

[wlbragg]

I guess reopening this isn't really appropriate as this was for the wings. But I want to get this smoothing business straightened out.

You can check the smoothing of all the model mesh I did in the sim with branch issue-118 #971

[gilbertohasnofb]

@wlbragg I will be commenting on the PR, as I think that makes it easier for us to track progress