The Polikarpov I-1, release by ICM a while ago, is the kind of subject that probably no mainstream manufacturer would release. This is an important aviation subject, though, since it was the first monoplane Russian fighter with metal wings. Interestingly, the factory designation was IL-400b (Istrebitel Liberty 400, due to the 400 hp Liberty engine used, and b for bis, because it was an evolution of the first prototype), while the later typical Polikarpov designation (I-1 in this case) was adopted by the time of the third prototype. It first flew in 1923, but the series was developed up to 1927. More about its history here.

Back to the model, this is a simple kit, containing about 30 parts molded in ICM's typical gray styrene and a single clear windscreen, plus the decal sheet. Painting instructions are printed on the back of the flimsy box. This is another project that I started while living in USA during 2009-2010. I started working on it adding several improvements to the cockpit, but shortly after I boxed the project to resume it only months later. More on that in a minute.


Start up: Adding details

As I said, work started by the cockpit. The kits parts are very simple, and the fuselage is internally devoid of any structural detail. So I simulated the tubular structure using plastic rods, as well as a throttle and other levers attached to the fuselage structure. The control panel is from the kit, with the gun breeches improved with pieces of stretched sprue. The instruments came from spare decals and were individually applied, with clear nail enamel simulating the glasses.


A short piece of plastic tubing was glued to one of the fuselage sides, and a small plastic rod glued in the corresponding position behind the control panel. This trick allows the installation of the control panel after finishing the model. I also made a new cockpit floor, with rudder pedals and cables. The pilot's seat is from the kit, but it was modified to accept a cushion and lap belts (I doubt shoulder belts were in order by that time). The kit provided bulkhead was replaced by another item, more squared, made of plastic card.


ICM included a partial "wing spar" sandwiched between the fuselage halves, a piece over which the wings are slotted later. It is a clever system, but you need to be sure that the cockpit floor does not interfere in its alignment. I opted by attaching cockpit floor to the spar with slow setting glue and temporarily closing the fuselage with the assembly in place. This gave me time to tweak the cockpit to its correct position. Since no glue touched the fuselage halves, the cockpit+spar assembly could be disassembled once the glue was set.


The fuselage interior was painted with GSI Gunze Mr.Color Silver (#8), and washed with artists oils. The cockpit floor was painted simulating wood, with the other details were painted in various colors. Small pieces of plastic were installed to act as tabs to make the fuselage alignment easier.



Following (partially) the instructions

Once satisfied with everything in the cockpit area, the fuselage was closed and the seams puttied and sanded. The engine cowling (part A3) was glued in place next. If you take your time (and install some sprue spacers inside the nose), a very decent fit can be obtained. Mr.Surfacer was used to fill deeper seam lines, and the excess cleaned with isopropyl alcohol. It is important to obtain a good fit of the cowling because it would be very difficult to sand the seam without obliterating the fine surface detail and overlapping effects molded on both, cowling and fuselage halves. And besides, we want a subtle panel line separating the cowling (metal) from the fuselage (wood), as in the real thing. The twelve engine exhaust openings were drilled, as well as a couple of vents on the top of the cowling.


The IL-400b had an interesting arrangement for the radiators. The water radiator was embedded to the chin, while the oil radiator was installed like a small wing under the fuselage. Since ICM didn't provide more than an almost imperceptible locating mark, I drilled holes to accept the oil radiator later:


The real aircraft had two prominent beams under the fuselage, ending in handles on both sides for aircraft spotting by ground crew. I sanded the molded on representation and made new ones using plastic strips, following more closely what I saw in photos.

Another important thing to mention here is a slight deviation I made from the instructions. The tail cone (part B14) should be glued after installing the horizontal stabilizers (part B9). However, I wondered it would be very difficult to sand the resulting seam with the stabilizer in place. Instead, I glued part B14 in place without part B9, and sawed the later in the middle, allowing the installation of each half later:


This gave me the necessary room to work the seam resulting from the tail cone installation without destroying the surface details of the stabilizers.


The disadvantage of this approach is obvious, you now have to be careful about the alignment of both stabilizer halves. This was easily solved by using a metal pin connecting the movable surfaces of the stabilizer which, fortunately, exist in most aircraft:



Problems

The next step was the main reason why this model returned to the box after my first attempt to build it: the wings. ICM made a very convincing representation of the corrugated Aluminum on all flying surfaces. Except for a couple of ejection pin marks under the horizontal stabilizers, the effect is really very well done. However, that's where the good news end.

ICM inexplicably used a strange parts break up to model the wings... Taking the port wing as an example, we have parts B1 and B2 forming the wing root. Parts A4 and A5 make the outer wing section, and part A8 forms the entire trailing edge of the wing. Therefore, there are four seam lines to tackle in this assembly:

L1: Seam line running along the wing root / main wing section joint.
L2: Seam line formed along the main wing section / trailing edge (A8) joint.
L3: Seam line running along the aileron edge.
L4: Seam line running along the leading edge.


Seam line L1 falls in a natural joint of the corrugated sheets, so it is not a problem. I read somewhere that you have the option to assemble the upper and lower wing parts before joining them as two wing halves. Anyway, the name of the game here is to make the corrugated profiles of upper and lower parts matching as well as possible. Think of a Douglas TBD Devastador wing - we cannot scape from working the L4 line. The seam line L3 is the natural boundary of the ailerons. Finally, line L2 should not exist, but it is almost impossible to eliminate it in a corrugated metal wing, let alone in a natural metal aircraft... I can't see any reason why ICM didn't make part A8 as just a separate aileron.

Anyway, I priorized the leading edge. If seam line L2 could not be eliminated, patience. Modeler Matt Bittner made a nice job in eliminating L2, but his model was incorrectly painted green/blue. I think the Aluminum color would betray the reworked joint. I cut parts A8 and A9 to make the ailerons separate items and glued the remaining of the part to the trailing edge of the wing:


The line L2 was treated with Tamiya Surface Primer.


After an initial inspection, it was clear it would be too risky to try to eliminate that seam line L2. I decided to live with that.


Then I had to work on seam line L4. I started brushing a good coat of liquid glue along the leading edge, using a stiff bristle brush swept perpendicularly to the seam. My solution to produce a nice effect, with the corrugated panels continuously curved along the edge, was to sand each recess of the corrugated panels using a seesaw movement with a sewing line. If the upper and lower wing parts are matching correctly, the line will naturally follow the recess on both sides of the edge, while the abrasive action will 'carve' the continuation of the recess along the seam line :


The result was perfect in one wing but just passable in the other. So I declared it done, and glued wings and stabilizers in place:


Details that had to be installed before painting include the landing gear struts and the rudder. I installed metal pins in the former to be inserted in holes drilled under the fuselage.


The landing gear struts were the next parts to be installed. I had a tough time installing them. In spite of the metal pins inserted in each attachment point, they insisted in an annoying sidesway which compromised the overall alignment of the wings. ICM could have provided positive attachment points here. After two unsuccessful attempts to align the struct, I decided to fix them in place with liquid glue and let it dry overnight. In the next morning the assembly was tweaked to the correct position and secured in place with small drops of super glue.



Painting & weathering

I also drilled holes to accept the rigging later. Next, automotive grey primer was airbrushed over and the model was checked for faults. A polishing disk was used to reduce ICM's grainy surface, particularly over the metal covered areas:


Finally it was time to paint the strange plane. After comparing several metalic colors on a piece of scrap, I selected Gunze Mr. Color Silver for the natural metal areas (wings and cowling), and a 50%-50% mix of Testors Aluminum metalizer and Testors Metalizer sealer for the wood covered areas (basically the fuselage). Strangely, the contrast between the colors found on my test pieces was not so clear on the model:


The decals (there are only two in this model) behaved weirdly. One of them settled beautifully, with the clear film disappearing after a coat of semi-gloss automotive varnish. But the other cracked and lifted the edges. I did my best to correct it with sanding and paint retouches, but it is still ugly as a newbie would do. I tried to find a replacement with no avail, and decided to go on as it was.

After checknig the few photos of the real thing, it was obvious that the aircraft had a lot of dirt accumulated around the engine and on several spots of the wings. I guess this is typical of a corrugated surface... Anyway, artists oils and lighter fluid worked beautifully to show recessed details and provide some weathering:



The undersurface received a heavier coat of wash, specially  the landing gear airfoil and lower fuselage:


And since the airbrush was around, I painted the remaining parts, starting with the propeller. I simulated a dark wood with brown enamels and oils, then masked the leading edge shields and shoot steel color:


Here are the remaining bits (except for the windscreen):


I was pissed with the problem with that decal, but I was not going to order another kit just for the decals, and then repaint the whole fuselage side...


Final assembly

The bits above were almost all glued in place, and the little Polikarpov started to look like an airplane:


The radiators were painted using Humbrol Metal Cotes and installed. Since these paints are sensible to mineral spirits, I applied a sludge wash using watercolors. The wheels also received a bit of dirt from pigments and washes, while the tailskid was painted to simulate wood with a protective metal shield. Next came a thing not provided by ICM: the wingtip skids, which were basically a semi-circular metal bar installed on the wingtips to prevent them from touching the ground. ICM provided the holes under the wings, though, and suggests to made your own using a 0.8 mm rod. I made mine using 0.5 mm steel wire bent to shape, painted flat Aluminum and glued in place:


Almost there. This was the "to do" list to finish the model:



More detail

From what I could understand from the period photos, the headrest was merely a piece of wood installed atop the pilot's seat. I made mine cutting a piece of plastic card to conform the shape of the fuselage behind the seat and simulated wood using standard methods.

The biggest problem by then was that cushing installed around the cockpit opening, so characteristic in early planes with open cockpits. I antecipated the problem early in this project when I made a male copy of the cockpit opening. First, the cockpit opening was covered with Aluminum foil which was rubbed to conform well to the cockpit borders. Next, a pool of Tamiya light curing putty was applied over the foiled area:


After mere 5 minutes under light, the foil was removed and a male mold of the cockpit opening was easily obtained. The mold was mounted on a handle and reserved until the end. I was not sure if it would be really needed, but I'm glad I made it:


Milliput was the weapon of choice to make the cushion. The mold received a good coat of talcum powder and a thin sausage of Milliput was carefully pressed along the perimeter of the cockpit. After drying overnight, the cushion was detatched from the mold and I had a decent part without damaging the cockpit. I fattened the part with a coat of Gunze's Mr. Dissolved Putty and declared my cushion done. A tad overengineered, but it worked:


I would do it differently today... Anyway, after painting the cushion it was coated with Vallejo semi-gloss varnish and glued in place. The windscreen was also installed, although it had the wrong shape. The correct should be a section of a cone (straight when viewed from the side), while the kit's part is curved:


The rigging wires which reinforce the wheel struts were made from stretched sprue and installed in the corresponding holes:



Display base

I then started to work on the display base. Two things were considered: (a) I wanted to simulated melted snow scattered on the grass, like in a sunny afternoon after a snowy night, and (b) I needed to place one or two figures next to the port fuselage to conceal the problem with the decal.

The base was made pretty much like my other bases. I cut a piece of cardboard to conform a wood base, and applied a liberal coat of wood filler, intentionally leaving a few flat spots. These areas would be pools of water/mud accumulated from the melted snow. The base was coated with Krylon gray primer and covered with Woodland static grass, except for the pools. I don't like the shiny appearance of static grass straight from the package, so different shades of green and brown were airbrushed over it, using the brown around the lower areas of the terrain and around the pools. I also glued small lengths of Campbell's railroad timber stained with Weather All to serve as platform for the wheels:


The melting snow was made by mixing Noch Powdery Snow with artist's gloss medium and Tamiya X-22. Yes, I know this is not the way Noch snow is supposed to be used, but it worked... sort of...


As it turned out after drying, the snow looked too flat and too much irregular, and I wanted more round snow tops. So I went back and mixed gloss medium (undiluted) with marble powder and top coated the previous snow application, smoothing everything:


It is not perfect, but I liked it. As for the figures, I modified a Prieser figure to become a technician talking to the test pilot. The pilot is a Plum Blossom resin figure, dressed appropriately for a cold flight. I thought a few bottles of Vallejo paint and thirty minutes would do. No mess... I couldn't be more wrong:


After watching a couple of videos on painting figures I spent a whole night painting the small guys. I did my best, but definitely I'm not a figure modeler:


After chosing a good position for the figures, they were glued on the base using holes previously drilled in the base:



Chalk up one more for me!

The last thing to do was to paint the gun muzzles, using a fine pointed brush. The model was finished...


Finally I could place the model on its base... a good feeling:



In case you are wondering, this is the photo of it two days before: