SBD Dauntless, from scratch

[Wurger]

Smashing job....

 

[Gnomey]

Lovely work so far!

 

[Witold Jaworski]

Wurger, Gnomey - thank you!

In this post I will continue verification of my model by matching it against the photos. This time I will check the wing geometry.

In the first photo from the Pacific Aviation Museum (in my model it is marked as PAM-1) I identified several differences:

First I noticed that the hinge of the upper flap in my model is in the wrong location (I had to shift it forward by 0.7 inch). The upper edge of the aileron bay had slightly different shape on this photo. In this picture the tip of the aileron (the point lying on the wing tip outer edge) is located in the front of the corresponding point in my model. (The difference is less than 1 inch). Surprisingly, the inner (root) rib of the aileron seems to be a little bit higher in my model than on the photo. I can see also similar difference in the root rib of the outer wing panel. Location of the aileron bay upper edge on this photo can also be interpreted as located below the corresponding edge in my model. Does it mean that I made an error in forming this wing? The last visible difference are the outlets of the fixed slats. According the photo they were smaller and set at slightly different angle.

To check the differences in the wing thickness and the details around its trailing edge, I matched my model to another photo:

I opened the wing flaps to better match the projection of my model to this photo. I used the shorter edges of the flaps to precisely determine their deflection. The bottom flaps fits well into this photo (if you take into account that in the depicted airplane the outer flap is bent). To fit the upper flap I had to shift it by 0.8 inch, as shown in the previous photo (as in the first figure in this post). This confirms that there is a difference! What is interesting, the wing on this photo is also slightly thinner than in my model — which confirms that I made a mistake in recreating the shape of its ribs.

As I wrote, I was convinced that I properly recreated the airfoil shape. I used the original coordinates of the NACA-2415 (and NACA-2409) airfoils (as you can see in figure "a", below)! Thus I used another, side photo (PAM-3) to check this finding:

The overall chord of the wing rib in figure above seems to be OK (luckily, on this photo you can see the fragment of the leading edge between the truck forks). The chord of its bottom flaps in my model also fits corresponding chord on the photo. However, the upper edge of the root rib in my model seems to be too high (by about 0.3 inch). I can see clearly that it occurs in the middle of the upper flange of this rib. On the scale plans this difference corresponds to just half of the contour line width! That's why we have to use photos: the drawing conventions alone make the scale plans not as precise as we wish… The PAM-2 photo reveals that this difference (maybe somewhat smaller than at the root) extends over the whole wing span.

Well, so I had to fix it. While lowering the upper part of the center wing was relatively easy (figure "a", below), I had also to modify all the adjacent objects — ribs, spars, and the fuselage

The more difficult was to make similar modification in the outer panels. The difference was smaller at the wing tip. To preserve the straight lines of the spanwise mesh edges I moved the whole selected area down by 0.3 inch, then compensated the difference at the wing tip by small rotation around the wing root chord. (I had to separately rotate each of these "longeron edges"). Of course, then I had to make a lot of minor compensations in the upper flap and the aileron contours.

However, I had to modify these trailing edge details anyway, following the other findings from the photos:

In this modification I had to revert the changes I made three months ago to the aileron bay edge (in my post from 2015-09-12 about details of the outer wing panel). It was the wrong location of the flap hinge, while the aileron bay edge should be in the place depicted on the reference drawings! In fact, drawing these scale plans I assumed that the hinge of the upper flap was directly above the hinge of the bottom flap. (You cannot see the difference on the most common, horizontal photos). Now I know that it is shifted away from the auxiliary rear spar by about 0.8 inch. After this modification I had to shorten the chord of the upper flap and rotate its ribs and spars, adjusting them to the altered directions of the flap skin. It required a few additional hours...

Once I finished with the trailing edge, fixing of the outlets of the fixed slats was easier. I just had to modify the shape of the "cutting tool" auxiliary object, used in their Boolean modifier:

Then I adjusted the slat internal surfaces, fitting their upper edges to these modified openings (as in figure "b", above).

If I encountered such surprises on the upper wing surface, what do I find on the bottom of the wing? I started by examining the outer wing panel:

It was surprisingly difficult to find an appropriate projection for this wing — I badly missed the fuselage here! (It would allow me to better determine the proper direction of the camera). The barrel distortion of this photo could also have some influences on this matching. Fortunately, it seems that my model fits better this area of the real wing. The first difference I found was in the fixed slats: minor adjustment of their direction and sizes. I fixed them in the same way as their outlets on the wing upper surface (I will not bother you by describing the details). Another difference is more subtle: it seems that the real wing tip has slightly different shape than in my model!

Of course, I had to check it on another photo, taken from another direction:

This photo confirms my finding: it seems that I made another wrong assumption about the shape of the wing tip. (I assumed that the rear part was a single arc, while it is at least a smaller arc and an unidentified curve — maybe short piece of another arc of larger radius?). Of course I accordingly modified the wing tip (by adjusting location of a few of its vertices — in fact it was not as complicated as it sounds).

For the complete verification of the wing, I used the picture from the SBD manual. I checked the bottom surfaces of the center wing:

To speed up narration in this post, the picture above is showing the updated mesh. I just enlisted the modifications that I made. As you can see I had to adjust the outer edge of the wheel bay (because it was not a simple circle). There were some minor differences in the split lines of the bottom covers (I had to adjust the bottom fuselage! Again!). Ultimately, I discovered that I placed the fixed ribs above the flaps in wrong locations (I really do not know why I not followed the stations diagram— now I corrected this mistake).

In this source *.blend file you can evaluate yourself the model from this post.

This is the last post about this "great verification". Now I am coming back to modeling. In next two posts I will recreate the empennage of this aircraft.

 

[Wurger]

 

[Lucky13]

Fantastic work!

 

[Gnomey]

Looking good so far!

 

[Witold Jaworski]

Wurger, Lucky13, Gnomey - thank you!

In this post I start to work on the tail assembly. The horizontal tailplane has similar structure to the wing — but it is simpler. Thus I started it in the same way as the wing, by forming its root airfoil:

In the most of the aircraft the tailplane has a symmetric airfoil. So it was in the Dauntless. I did not find its signature (family) in any of the reference materials, thus I carefully copied its contour from the photos (its rear part — the elevator — seems to have modified shape, anyway). It has incidence angle of 2⁰, so I rotated the rib object and used a Mirror modifier to generate its bottom part.

During this work I decided that I will use this rib as an auxiliary reference object for shaping the horizontal stabilizer. To precisely match the contour copied from the photos, I rotated part of this curve in the top view. Now it runs along the outer edge of the tailplane fairing:

However, because I am going to copy this rib into the initial edge of the horizontal stabilizer, I already prepared three vertices for the leading edge of the elevator (as in the figure above).

During this work I was struck by the idea that it is stupid thing to model the whole empennage, and then to verify it against the photos. The much better approach is first to "draw" in the 3D space their contours and match them to the photos, then to model their surfaces. In this way I can identify errors in my reference drawings before I start the modeling! The parts formed in this "verified" way and continuously matched to the references will have better quality!

Thus I interrupted forming the horizontal tailplane, and quickly shaped another auxiliary object — the contour of the rudder and fin:

What's more, I decided to recreate in the model the basic reference "trapezes" of the fin and rudder. They are determined by the explicit dimensions in the general arrangement drawing, which I already used some months ago to draw the 2D reference drawings:

While in the model space the 1 unit corresponds to 1 inch, I did not need to multiply every dimension by the scale coefficient. It was a big surprise when the trapeze drawn according these re-applied dimensions occurred shifted left by 0.7"!:

I immediately did the same test for the horizontal tailplane. It also was shifted by 0.7"!:

Well, such a coincidence suggest that I made a kind of systematic error in calculating locations of the elevator and rudder axis for my scale plans. Most probably there was something in their extremely long position, measured from the wing leading edge (see the general arrangement diagram in the fourth figure from this post). For example, it could be a rounding error of the scale coefficient!

If I was wrong in this case, I could made other errors. I decided that it is proper time to re-use the original photos from the web page of Chino Planes of Fame Air Museum. Their resolution is only half of the resolution of the photos from Pacific Aviation Museum Pearl Harbor. However, they were made using a long-lens camera. (You can read that the standard length from the EXIF section of the Chino photos — it was 400 mm. The photos from Pacific Aviation Museum Pearl Harbor were made with the standard lens length: 36 mm).

Using this focus length, it is easier to fit the model and the photo:

As you can see, this is a flying airplane — and there is no visible dynamic deformation of the wingtip! This means that the whole "theory" about the wing deformation that I described in my post from 5th December was wrong! The wing is much stiffer than I thought. The deformation of the historical photo can have other reason. It could be significant barrel distortion of its lens, or the deformation of the negative. I do not know.

I verified contours of the horizontal tailplane by matching the model to another photo:

Note, that this is another photo that I used to draw my scale plans. However, this time I left it unaltered, to avoid eventual errors that I could made by setting it horizontally and scaling.

In general, the model fits this photo pretty well. However, there are small differences at tailplane and wing tips. I started to suspect that such a photo can still have a small barrel distortion.

Finally, I used the third Chino photo to further verify the side view:

The dimensioned contours of the empennage helped me to match better the other photos. For example, I slightly updated the projection parameters of the SBD-5 pictures from Pacific Aviation Museum Pearl Harbor:

The contours of the tail and fuselage fits the photo pretty well. There is just a visible difference in the wing tip spans — I think that this is the effect of the barrel distortion.

In this source *.blend file you can evaluate yourself the model from this post.

Well, I started to build the tailplane in this post, but this process ended in another verification. However, it spared me from similar check that I would have to perform on the finished empennage. Now I can quickly build this assembly — in the next post I will finish the horizontal tailplane.

 

[Wurger]

 

[Lucky13]

 

[Gnomey]

Nicely done so far!

 

[Witold Jaworski]

Wurger, Lucky13, Gnomey - thank you!

After some verification of the reference contours that I described in the previous week, I am coming back to modeling of the horizontal tailplane.

In the previous post I created the reference airfoil of its root rib. Now I copied it into a new object, straighten along the fuselage centerline, and finally extruded spanwise:

I checked the resulting shape, ensuring that the thickness of the tip ribs matches their counterparts on the photos:

When this base shape was verified, I started to form the curved contour of the tip. Basically, it was an arc, thus I shaped it by extruding and rotating subsequent mesh segments:

Preparing the horizontal tailplane for such a mesh topology, I used the same number of rib vertices to form the leading and trailing edges of its root airfoil.

In the next step I created an additional gap in this mesh, at the point where it will be split between the stabilizer and the elevator (as in figure "a" below):

Such a gap deforms the original circular contour of the tip. To restore its shape, I had to move a little two nearest vertices on each side of the gap. Facilitating this task, I used an auxiliary circle as the reference shape.

To fill the empty space inside the tip, I extruded the internal edges of the last rib (as in figure "b", above).

I slid the last vertex of the edge that runs along the elevator leading edge, forming in this way the angle visible on the reference drawings (as in figure "a", below). In fact, its location was re-checked on the reference photos, thus it lies in a slightly different place than you can see on the underlying scale plans.

Of course, I also scaled the thickness of this newly formed "rib" (as in figure "b", above), aligning it to the slope of the previous, trapezoidal segment of this tailplane.

In the next step I started to build different topologies in each part of this tip mesh. In the "stabilizer" part I joined the "tab" of the internal faces and the leading edge (as in figure "a", below). In the "elevator" part, I removed the first and the last face of this tab, and shifted the vertices of the middle face, forming a thinner trapeze. Then I extruded the outer edge of this face several times, rotating them around the "corner" of the elevator leading edge. Note that each of these faces corresponds to a single mesh segment on the tip external contour:

I also joined the "gap" in the tip contour into a single, "sharp" (Crease = 1) edge. (In fact, I should create it as a single sharp edge in the beginning). Such an arrangement allows me to quickly create an array of new faces that closed the tip of the elevator (as in figure "b", above). Note that I filled the gap in the "corner" using two quad faces.

To match the topology of the elevator tip, I had to add additional "rib" to the stabilizer mesh. I did it in three steps. First, I created edges that joined the corresponding vertices of the tip external contour and the internal faces (as in figure "a", below). Then I split them by half (using the Subdivide command). Finally I used all these vertices to create new faces (as in figure "b", below):

Note that I had to create a single triangular face near the leading edge (see figure "b", above). Fortunately, the mesh curvature in this place is low enough that it does not disturb the resulting, smooth shape of the tip.

When the overall shape of the tailplane was ready, I split it into the stabilizer and elevator objects. I did it by copying the original object and then removing the "elevator" part of its mesh faces (figure "a", below):

Similarly, I removed the "stabilizer" faces from the elevator object (see figure "b", above). Ultimately I also simplified this mesh by removing one of its "longeron" edges. (It seems that the tip contour in the front view requires just a three-point curve).

The elevator of the SBD Dauntless had an oval leading edge (it was the aerodynamic compensation, an area shifted in the front of the hinge line). I started to form this element by inserting on the symmetry plane a circle (consisting 12 vertices):

Then I extruded it spanwise, adapting its radius to the local airfoil thickness (as in figure "a", below):

In the next step I removed the rear faces from the leading edge cone, and joined it with the rest of the elevator mesh (see figure "b", above).

The presence of a single middle edge in the elevator tip allowed me to remove similar edge from the stabilizer tip (as in figure "a", below):

Of course, it would be even easier to not create this edge at all — but this is typical situation, when I modify the initial concept of the mesh topology during the progress of the work. Figure "b" (above) displays the resulting tailplane assembly.

In this source *.blend file you can evaluate yourself the model from this post.

In the next post I will describe my work on tailplane fairing.

 

[Wurger]

 

[Gnomey]

Lovely work so far!

 

[Witold Jaworski]

Wurger, Gnomey - thank you!

In the previous post I formed horizontal tailplane of the SBD Dauntless. In this part I will describe how I created the fairing between this tailplane and the fuselage. It is an easier part than the wing root fairing, because it is smaller and most of its cross sections are not circular.

At the beginning I cut out from the stabilizer its middle segment, along the root rib:

Then I "draw" the outer contour of this fairing in the side view. I also checked it in the reference photo (as you can see in the figure above).

Then I projected this "sketched" outer contour onto the fuselage. I did it by extruding its polyline into a face strip that crosses fuselage surface (figure "a", below), then finding the intersection edge of this mesh with the fuselage (figure "b", below):

The intersection edge was calculated by one of my Blender add-ons (named Intersection — you can download it from here). In general, it would be easier to extrude this edge horizontally (because I sketched this contour in the side view). However, I was afraid that the add-on will lost the track of the upper rear part of this mesh (the part that crosses just the upper tip of the fuselage surface). That's why I initially shifted this contour close to the fuselage, and extruded it in a more-or-less perpendicular direction to the fuselage surface.

All in all, after this operation I have the three edges, which is enough to create the first version of a smooth fairing:

Figure below shows the initial smooth, subdivided mesh based on these three edges:

It starts to resemble the original element. I created here the new row of faces, from the middle edge to the outer contour. Then, before creating this screenshot, I switched the display mode to the resulting subdivision surface.

To have better control over the shape of this fairing, I inserted two additional edge loops into this mesh

These additional vertices were extremely useful in shaping the bottom edge of this fairing, which had a semi-circular cross section:

In the most of the aircraft designs the fairing is just a piece of sheet metal bolted over the fuselage skin. In the SBD Dauntless it was an integral part of the fuselage skin (except the area around the stabilizer leading edge). Thus I had to extend the bottom part of this mesh, copying the fragment of the fuselage surface (as you can see in the figure above).

Figure below shows the finished fairing. As you can see, it smoothly fits the fuselage:

It was not necessary, but I also created the rear spar of this tailplane — just because I do not like to see a large empty space in a finished element (the fuselage is not finished, yet!).

In this source *.blend file you can evaluate yourself the model from this post.

In the next post I will create the fin and the rudder.

 

[Wurger]

 

[Gnomey]

Good work so far!

 

[Witold Jaworski]

Wurger, Gnomey - thank you for following this thread!

This week I have worked on the SBD vertical tailplane. I started by forming its root airfoil (see the figure below). I had no description nor a direct photo of the airfoil used here. However, the reference photos reveal that it could have similar shape to the airfoil of the horizontal tailplane. Thus I copied that curve into this mesh.

Note that I used here a thin strip of the faces instead of a single curve (which I used in the case of the horizontal tailplane or the wing). The reason is simple: on the single subdivision curve I cannot mark a "sharp corner" at a control point (original mesh vertex). On the face "strip" I can mark the corresponding edge as sharp (increasing its Crease coefficient to 1). I marked in this way the edge at the split between the rudder and fin. Simultaneously I can form such a face strip in the top view as easily as a single curve. (I just have to remember to select its vertices using the group select commands (Border-select or Circle-select), instead of the simple mouse click). Why didn't I use this method in the previous cases? Well, good ideas require some time to emerge…

Once the root airfoil was ready, I extruded it into the basic trapeze:

Then I split this mesh into the rudder and fin (i.e. into separate objects, as in figure below):

Note that I added additional "rib" edges to the mesh of the fin. They will be useful in forming the forward fragment of this part.

Initially, I extruded the first approximation of the dorsal fin from the bottom edges of the lower ribs:

However, I decided that the resulting topology of this mesh differs too much from the original layout of the panel seams (and the original ribs and spars). To make a better approximation, I used the fin shaped in the previous step as the reference object (in the figure below it is in red):

I cut out the forward part of the original fin, forming in this mesh the first vertical edge. Then I extruded it into next segment:

The next segments were extruded in similar way:

In each new segment the vertical cross section is significantly smaller than in the previous one. I had to compensate it by cutting out its bottom fragment (using the Knife tool — as in figure above) and reducing the number of remaining faces.

Figure below shows the resulting dorsal fin:

The tip of this fin was formed in an unusual way — it was stamped in the cover of a fuselage hatch (as you can see on the photo). I will form this cover later.

Figure below shows the objects created in this posts:

You can examine them in this source *.blend file.

In the next post I will describe my work on the fairing of this fin (it seems quite simple, but occurred more difficult than I expected!).

 

[Wurger]

 

[Witold Jaworski]

Wurger, thank you!

In the SBD Dauntless the fillet along the fin and the fuselage was formed from the bent bottom edges of the fin panels. I am showing it in the figure below:

(To make some of these panel seams more visible on thee photos, I sketched along them thin lines). You can observe that each fin panel overlaps the next one, starting from the tip stamped as the part of one of the fuselage doors (see the second-last figure in the previous post). The outer contours of these panes are not perfectly aligned: you can see small overlaps on the photos (see the figure above). Surprisingly, such a detail makes the modeling more difficult. However, the most difficult part will be the seam between the fin and the horizontal tailplane fairings (as in the figure above). It runs along the fuselage longeron, across the fillet between the stabilizers and fuselage.

Well, I started this fillet by extruding some faces from the bottom edges of the fin mesh:

As you can see, I already recreated the sharp panel corners in these extruded faces. Then I lowered their outer edges and aligned them to their contours on the top view:

This is the first, rough approximation that fits these panels to the fuselage surface. In this process I discovered that I had to make some modifications to the upper part of the tailplane fairing. However, I was not entirely satisfied with the result: comparing to the photos, something was wrong at station 242 (see the figure above). The outer seam of the fin fillet should be a little bit wider here!

After some additional deliberations I decided that the fuselage under the fin was somewhat higher (by about 0.5") than on my reference drawings, and the upper arc of these bulkheads had larger radius. Thus I had to modify this part of the fuselage:

I rotated a little this mesh fragment, then scaled up the upper part of each of its three bulkheads.

I had no photo of the SBD fuselage without the fin, taken from the side. In fact, the shape of this fragment on my scale plans is in 80% my assumption! Such small anomalies as this one helps me to discover the real shape of this airplane.​

Basing on the corrected fuselage, I was able to fit better the outer edges of this fin to the fuselage and the tailplane fairing:

However, I was not satisfied with the forward part of the seam between these two fairings: despite all my efforts, it looked a little bit sharp!

Ultimately, I had to reshape a little bit the tailplane fairing and cut out the excess of its surface along this seam using a Boolean modifier:

This method produced results that resemble the smooth shape that I can see on the photos:

However, I do not like such a "Boolean – based" solution. It seems too complicated. In the next post I will try to eliminate such a "hard" seam between these two panels. I think that much better solution here will be a continuous, smooth surface. The seam can be recreated later, using textures.

Anyway, in this source *.blend file you can evaluate yourself the model from this post.

 

[Wurger]