Zero 52 article part 2

Back Next




The cross section of the fuselage

 The cross section shape of rear fuselage of Tamiya kit is not elliptical but unexpectedly rectangular with rounded corners.







Area between the cowling and firewall

 Tamiya fuselage side is slightly curved at the firewall section in plan view but Hasegawa is straight. I checked references but I couldn't determine which one is correct. For example, POF Zero matches Tamiya, but it is a restored aircraft, and another photo shows straight line. I didn't make any corrections on my model. Hasegawa kit is on the left, Tamiya's on the right. The widths of both firewalls are the same.



The left is Hasegawa, the right is Tamiya. The widths of both firewalls are the same.



Cowling

 The width of cowling is different between Tamiya and Hasegawa. The maximum width is 25.0mm on Tamiya and 24.5mm on Hasegawa. The difference of the cowling may leads to with or without of bent of fuselage. In another approach, the diameter of Sakae engine is 1,150mm and 24.0mm in 1:48 scale, so the gap between engine and cowling is 2cm on Tamiya and 1cm on Hasegawa. Mmmm, which is correct? I heard the rumor that Tamiya measured the actual cowling existing in RAF Cosford Museum. So....

width height length height of opening width of opening
Tamiya 25.0 26.0 22.5 16.3 15.7
Hasegawa 24.5 25.8 22.5 15.7 15.2



Side view

 I tried to compare the side view of Tamiya kit to the famous side view air-to-air photo of captured model 52 (ref.-4 etc). The actual photo was taken from just beside of the nose and the distance could have been about 30m (it is so close) judging from the shape of the main wing or horizontal fin in the photo. So I photographed the kit from the same distance, scaled down to 1/48 of course. Then I compared both photos.

I traced each outline. The red line is actual photo and the blue is the kit. Then I superimposed the images. The two outlines should be identical. Anyhow, we have to consider other conditions such as distortion of the camera lenses and the exact distance from the camera to the aircraft.





The error of perspective

 I considered the distortion caused by the perspective in a photo. In this consideration, I ignored the distortion of lenses-see. Please see the figure below. The point A is the position on the right beside of the camera, the point C is the location of the camera. In many cases, point A is near cockpit, in this case A is the nose.




Next, I calculated the error of the tail fin in a photo. The tail fin is located a little farther than the nose. If segment AB is 10m and segment AC is 30m, then BC becomes 31.6m from the Pythagorean theorem (root of 10x10+30X30). Thus, the tail fin is further than nose by 1.6m. The visible size of object is inversely proportional to the distance. So the tail fin looks 5% (30/31.6=0.95) smaller. I'd call it "Error of the distance" in this case. When we think in modeling terms, if the height of the tail fin is 4cm, the error of distance becomes 2mm. It cannot be ignored.

Furthermore, the tail fin is seen on a skew by angle "theta", then the length appears shorter. Let me calculate this error. The actual length AB looks AE and triangle ABC and ABE is homothetic, so the same calculation leads to 5% error. I'd call it "Error of the angle". In conclusion, the error of tail fin length is the sum of the error of distance and error of angle, 10% total (to be exact 1-(30/31.6)^2=0.90). When we think in modeling terms, if the length of the tail fin is 4cm, the error becomes 4mm. It is too significant!

Now let's calculate the total length of the fuselage. I assume that actual length is AB and apparent length is AD. Then, AD=(AC x sin("theta"/2 x 2) , here AC=30m theta=18.4degrees, so AD=9.6m. The error is 4%. If the length of model is 20cm, the error becomes 8mm. Mmmm.

Calculations can easily be done using Excel-see below. This table assumes the length of the aircraft is 10m (i.e. AB=10m). When the length is different in your case correct it to proper ratio.

AC : distance from camera (m) 30 40 50 70 100
Theta : angle (degrees) 18 14 11 8 6
Error of distance (%) 5 3 2 1 0.5
Error of angle + distance (%) 10 6 4 2 1


As the result, the error is so large when the distance from camera to aircraft is 30m. Notice that the most of side view photos were taken closer than 30m. At the 50m distance the error decreases and is not a serious problem (1 or 2 mm in 1:48 scale). From 100m the error is ignorable in human eyes. Remember these errors when you look at actual photos.


The error of lenses

 The error of lenses adds as well. I have no knowledge of camera lenses as there are a great variety of them. Some widen near fringe of the frame so the error of distance and angle is reduced when a object is shot using these lenses. As far as my digital camera (Cannon Powershot S3) is concerned the fringe is shot without modification (in other words error of distance and angle are not corrected).



I shot my cutting mat with my camera without zoom. The width of the mat is 20cm and the distance from the camera is about 25cm.
The center and far right of the left photo are combined. The right half shrinks by about 5% in the height and 10% the width.



Fuselage panel lines

 The vertical panel lines which I marked "X" in the photo below are not found on any model 52. In the case of model 21, this "X" panel line exists but also the horizontal line is found as marked by pencil. The vertical pencil line is my error, please ignore it. I didn't check models 32 and 22.

Rearwards of this line a vertical panel line exists near the small square access panel on some models 52 or their later sub-models. If the line is present there are two variants of it - one is located in the center of panel and other is rearwards of the panel as in the picture. I can confirm that this panel line is not found on Nakajima built "standard" model 52 and found on model 52-hei. I cann't establish the exact rule but I suspect the pattern that early production is "without panel line" and late production is "with it".



Model 52 doesn't feature the panel line marked gXh. So you have to delete it.



Fuselage riveting

 Rivet patterns shown in the current drawings are almost accurate. There are some minor errors however. You can refer to my model when riveting yours. I represented rivet line carefully, on the correct side of panel line, single or double rivet line.



Rivets have been engraved but the surface is not sanded yet.
The oval access hatch was engraved using home-made template.





Some rivet lines are still to be applied (will be shown later). The line forward of rudder hinge line is apart from the hinge line at lower end. I mostly used #1 beading tool, #3 on the fringe of the tail fin filet and #0 on the tail cone.


The front half of the fuselage is like this. Basically port and starboard are almost the same but some minor lines are different due to the equipment installation.



There are six lines downward of small square panel on the real aircraft but I was unable to represent them in 1:48 scale. There are four lines on my kit then. The horizontal line just above the wing filet was engraved with #2 beading tool because this panel was not a structural one but only for airstream and attached afterwards during the actual manufacturing. Nakajima Zero was riveted with convex rivets.



Gluing the fuselage and wings

 Overall the fit of the Tamiya kit is excellent but you need to take care connecting the wing and fuselage. Lot of dry fitting is highly recommended. I think the thickness of the wing is the key point.



For the purpose of riveting I reinforced the attachment area of the lower wing and fuselage. This reinforcement can be seen through the cockpit even after completion so I painted it black.
Then I glued the fuselage and the wing. I used super glue on the front and rear, and used thin liquid cement for upper filet side.



Riveting

 The starboard side is here. Please note the border of the filet and wing. As for the aft area of the rear spar, the border of filet and wing is similarly structural joint therefore there is a rib underneath this panel line and consequently rivet lines along it. On the other hand, as for the front area of the rear spar, the border is not a structural one. The filet panel is attached to the wing skin. There is no rib behind this panel line and the rivet line is only on the filet panel.



The rivet line which was filled with light gray putty may be found on port side only. The small hole under the windshield is opened on port side as well. Kit filet is too thin and I made holes during riveting process so be careful.



Previously left off rivet lines were added at this stage.
Underside of the fuselage looks like this.


The integral structure of the fuselage and wing can be understood. In this picture, the horizontal lines are the parts of fuselage frames. Therefore the rivet lines correspond to their locations. The wing rib nr.03 represents the border line of attachment to the fuselage structure. This is the example of Mr. Horikoshi's indigenous design how to balance strength and weight.


Cowling

 I carefully re-scribed kit panel lines usin a needle and sanded off the parting lines.



This is a custom-made template from 0.2mm thick plastic sheet. It is fixed with double sided tape and regular scotch tape.



Cowl flaps

 The kit parts represent closed position featuring slight opening to facilitate release from the mold. So I cut this part using PE saw and bent it.



Original kit is on the right (1 degree open). Modified kit is on the left (-2 degrees close). No too much change visible?



Cowling rivet lines

 Every drawings I've seen don't depict them correctly. These lines are easily visible in the photos of the real aircraft. But there are very few materials for figuring out rivet lines on the upper and lower portions. So some lines have to be guessed from inner structure.

There are some variations found on the existing aircraft. I don't know the origin of these variations. They may have been caused by a model sub-type, period of manufacture, manufacturer or restoration work. Anyway I tried to depict the pattern of Nakajima-built standard model 52. Mainly I used the photos of existing cowling fromf RAF Museum in Cosford (this cowling and engine is said to be from the same aircraft as an IWM Zero fuselage).


One of the inferences of a model 53-hei

 Model 53-hei was equipped with Sakae 31 engine and the cowling upper line was bulged. This is visible in some WWII original photos and can be distinguished by the gap (step) between cowl upper line and upper fuselage forward of windshield. And now, how is the shape of this cowling in 3D?

The change of the shape of cowling was required due to the airflow of the engine. So if I was an engineer in charge, I would think that the changes on the production line should to be minimized, then the cowl flaps and gun tubes not to be changed and use the maximum of the existing 52 cowling. Thus only the minimal portion of the panel between the two gun tubes would be cut and the new bulged panel just installed..... Well, this is my inference. What do you think?


Rivetting



Riveting was almost finished. I used #1 beading tool but it's a little over size.
The forward edges of the cowl flaps are not straight but slightly curved.

The upper side is like this.
Comments are as follows.


Nakajima built Zero has eight convex rivets at the red line. This is confirmed by POF, cowling from Cosford and the original WWII photo. I don't know about Mitsubishi-built Zero but some restored Mitsubishi aircraft don't have these convex rivets. The blue lines are for fixing of the air duct. This air duct can be confirmed from Cosford cowling etc. Accuracy of the yellow lines is deducted from the inner structure which fills the gap between the engine and cowling.



There are double crank shaped lines on the lower half of the cowl. These lines are confirmed by POF aircraft. This shape seems to be created by the inner parts made of canvas to absorb the vibration (reference-27) and each corner corresponds to cylinder heads.
This canvas is also put on upper cowling but there is gap and the canvas is put on the inner structure so the crank rivet line does not appear on the cowling surface.



Exhaust pipes



This is the starboard exhaust. The welded line is depicted with a stretched sprue.
Pipe ends were drilled out.



Heatproof plates

 The rectangle heat proof plates aft of the exhausts differ between Nakajima and Mitsubishi built Zero. Tamiya kit represents Nakajima. These plates were about 50% shorter on Mitsubishi planes. But the accurate sizes and shapes are not clear since there is no clear photo of them. As for port side, photos of ref.-4 page 60 and ref.-20 p.83 are insufficient but precious evidences regardless.

As for starboard side, my friend send me photos of Mitsubishi 52 kou which was discovered on Guam islands after the war and the starboard heat proof plates are half size of Nakajima. But these photos only show two upper plates so the lower plates' size remains unknown (but I guess they were the same as the upper ones). It is also said that the early production of Mitsubishi 52 didn't carry those plates at all. Model Art Magazine article stated that starboard plates were larger than Nakajima, but I have my doubts.


The horizontal tail surfaces

 This is the upper port side. The under surface or starboard are the same. Some drawings depict some double rivets lines, but I couldn't confirm it.









Back Next

HOME