The "Little Donkey"

Polikarpov I-16 Ishak

in 1:33 scale from Fly Model

by William H. Geoghegan

Model photographs by the author

Cover Art

Click on a thumbnail for a larger image


Model: Polikarpov I-16 (Variously known as Ishak, Rata or Mosca), probably a Type 24 (1939-40)
Kit: Fly Model #53 (the I-16 is part of a 3 plane package that includes the I-16, IL-2, and Ju-87)
Scale: 1:33
Price: US $12.00 Plus S&H for the 3-plane package from Paper Models International
Difficulty: Medium (3 out of 5)
Number of parts:  Approximately 125
Instructions: Very brief and in Polish. The more important points can be teased out with the aid of material on the translation page at Saul Jacob's card modeling Web site.
Diagrams: Excellent. They occupy the equivalent of two A4 sheets.
Fit: Good to excellent with one significant exception (see below).
Printing: Fair. There were some color registration problems and blemishes. Printing was much too light on the black and white sheet containing gun and bulkhead patterns, engine detail, etc. These had to be scanned and reprinted to get a good black. Panel lines were nicely scribed, and the cardstock used was very good.
Coloring and Artwork: Good. Only three colors are used (other than black and grey): olive (for Russian topside green), sky blue, and red. This leads to some odd shortcuts -- like having the starboard wingtip light printed sky blue rather than green. Except for the wheel wells, surface detail is fairly sparse.
Resources: A considerable amount of material on the I-16 is available on the Web, including drawings, photos, and detailed writeups that cover almost every variant produced. Good reference sites include the Russian Aviation Museum and the New Zealand Fighter Pilots Museum. The latter was involved in the recovery and reconstruction of six I-16s from Russia, and they host some fantastic photos and written accounts of the plane. Special thanks go to David Okamura, Fred Bultman and Wayne Cutrell for sources and commentary. Peter Ansoff's review of the Mwe I-16 on Saul Jacob's Web site contains a wealth of information about the plane itself, not to mention a great illustration of how to write a review.


Nikolai Polikarpov's I-16 heralded a revolutionary new approach to fighter aircraft design when it was introduced in late 1933. It was the first operational cantilever monoplane fighter with fully retractable landing gear; and its performance was near top of the class in comparison to other fighters entering service at that time. It made a worthy showing in the Spanish Civil War, where it was used by Republican forces, and during the early stages of the Sino-Japanese conflict. The I-16 appeared in many variants, and performed in a wide variety of roles. It served as a fighter for several air forces; it was used by the Soviet Union in a ground attack role; it performed as a trainer in both single and two-seat versions; it was carried by Russian heavy bombers as a parasite escort and dive bomber; and it was used late in the war as a high speed reconnaissance aircraft. A few were serving as trainers in Spain as late as the early 1950's. Actual production figures are hard to come by; but estimates range from a low of just over 8,600 to almost 20,000 on the high side.

The I-16 never received an "official" name from the Russian authorities, but it was popularly known as Ishak or "little donkey" -- appropriate given the multitude of roles it was called upon to play. The Spanish Nationalists referred to it by the pejorative Rata ("rat"), which for some unfortunate reason has stuck with it over the years. The Spanish Republicans called it Mosca ("fly"), which seems a little more appropriate. This kit models a Russian plane, however, so I'll stick with Ishak. "Little donkey" it is.

An interesting sidebar to the history of the I-16 concerns the efforts of Tim Wallis, a New Zealand businessman, to find and restore rare Russian aircraft of World War II. Working with the Russian government in the early 1990's, Wallis was able to recover six wrecked I-16's and have them restored to flying condition at one of the original manufacturing facilities in Siberia. (Three Polikarpov I-153's were also restored in a related project sponsored by Wallis.) After being test flown in Russia, all six planes were taken to New Zealand. One will stay in the NZ Fighter Pilots Museum, and the rest will be sold to recover the restoration costs. (I believe one has already been sold to a buyer in California, so it may soon be appearing in Stateside airshows.) This remarkable story of recovery and restoration can be found in a 1998 article available at the NZFPM site.

The Fly Model kit:

I developed a real love-hate relationship with this kit. I liked the subject, the ease with which so much of the kit went together, the fit of most (but not all) components, and diagrams so precise as to make the sparse Polish instructions almost irrelevant. What frustrated me most were the seemingly "minor" design errors that mushroomed into serious assembly problems with some of the most critical components. In other respects the engineering is excellent, and even approaches the precision of computer-assisted designs. But contrasted with this is the marginal quality of some of the printing, and the rework that was required on the non-card sheet of parts. (One of the other kits in this three-pack, the IL-2, had a large area on the upper surface of one wing where ink was incompletely applied, yielding a blotched appearance.) All that aside, the end result is still very pleasing, maybe even "cute," if that can be said of a warplane.

The type of I-16 represented by this kit is somewhat open to question. The gun fairings on the fuselage make it a Type 10 or later. Armament and oil cooler (?) intakes suggest that it is probably a Type 24.

And, finally, as if the designers wanted to make sure that the builder didn't run out of problems to solve, "left" ("L" or lewa) and right ("P" or prawa) are reversed! The "right" wing, for example, is the wing seen on the right when the plane is viewed from the front -- not as seen from the cockpit. (The wing on the seated pilot's right is the "left" wing in this kit.) This caused no end of confusion as I tried to figure out how this kit was supposed to fit together. The other two kits in the three-pack use the standard notation. Go figure....


The fuselage uses standard segmented construction; adjoining segments are glued bulkhead to bulkhead without connecting strips. The fit is very good, but you can help both accuracy and appearance by test-fitting adjoining bulkheads and making any needed adjustments to shape and taper before attaching the outer skin. Once the skin is fitted and glued, you can lightly sand the bulkhead end on a sheet of fine sandpaper laid flat on your work surface. This helps insure a tight fit with little visible gap between segments.

The cockpit floor and sides make into a separate assembly that is slid into place and glued after being formed. It would be fairly easy to do some extra detailing of controls, cabling, instrument panel, the landing gear winch (the gear was raised and lowered by hand!), etc., before the cabin is glued in place. The detailed cockpit photos on the NZFPM site would be an excellent guide. I settled for a scratchbuilt joystick modeled after the NZ restorations.

Construction of the two forward fuselage segments is a complicated matter. Both of these segments embrace portions of the wheel well assemblies. The forward bulkhead in the second segment also serves double duty as the wing spar and base for mounting the two wheel well assemblies, as shown below:

Wheel well assembly

The wheel well assemblies themselves need to be completed and mounted on the spar before the fuselage skin is applied. Unfortunately, the fit of the wing assembly cannot be tested until this part of the model has been glued together. I found a slight mismatch between the wheel well sides and the fuselage skin in segments 1 and 2 that could probably have been avoided through dryfitting the parts before final gluing. The best approach is probably to "tack" components in place, test the fit, and then make adjustments as needed before final gluing. I ended up forcing pieces into place, hoping the glue would dry before they slipped back out of alignment.

The engine exhaust ports (which I installed as one of the last steps) were supplied in a roll-up format that looked as though it would present significant problems in shaping, sanding, and coloring. I decided to use J. J. van der Sande's CONE program (version 1.3 is required) to lay out an exhaust port with the correct angle, diameter, and length. I imported the DXF output into Paint Shop Pro 7.0, colored it black, and duplicated it multiple times in a single sheet image (with PSP 7). I printed this out on 47 pound coated paper, and then cut and formed the individual ports on a bamboo skewer (perfect size for a 3mm diameter tube). The black printing was rolled on the inside, and the edges were butt glued. I painted the outside with Floquil gunmetal black. The result matched the shape, thickness and colors of the NZ restorations as well as you could want. These were glued in place with cyanoacrylate so as to protrude approximately .5 mm above the fuselage skin, a best approximation to what was shown in the photos I had available.

The dorsal "hump" (part 7) and the fairing that mates it to the rudder should be installed after the stabilizer and rudder have been glued in place.


The wing construction is unusual in that there are no ribs to aid in shaping the airfoil. Tabs on the wing covering, and slots in the fuselage help with shaping, along with gluing tabs on the underside of the wheel well housing. I added a folded gluing strip at the trailing edge to help preserve the airfoil at that point.

The most significant problem was in fitting the underside of the wing covering to the fuselage and wheel well housings. Unfortunately, I cut the opening in one wing on the lines indicated before test-fitting the overall covering; and only then discovered that the wing opening extended beyond the housing by about 2mm, or more than 1/16". I filled these sections in with scrap card and painted over the section to hide the patch. On the other wing, I cut well inside the outline, dry-fitted and trimmed it to shape, and then painted over the remaining white space and construction lines. The problem areas are marked by red ovals in the photo below. A second problem was that one part of the housing extended too far toward the fuselage, into the area exposed by the wing covering. I ended up cutting that section of the housing loose, removing a short section, reshaping and regluing it, before finally attaching the wing covering. The problem area is flagged by the green oval in the picture below. All of this changed the shape of the wheel well, of course, which had a slight (but not too serious) effect on placement of the landing gear assembly. This is what I mean by a "mushrooming" problem.

Apart from this, the wing covering went into place without a hitch, and held its shape surprisingly well, despite the lack of wing ribs. The wing fairings shaped easily and fit perfectly.


The stabilizer and rudder go together very easily, and a well designed and reinforced spar assembly built into the fuselage bulkhead helps to maintain a proper airfoil. It's probably a good idea to create a folded gluing strip for the trailing edge of each to insure the proper shape. The fairings that mate them to the fuselage fit perfectly.

Nose and propeller assembly:

The nose assembly can be tricky, and it requires some advance preparation if you want a rotating propeller. The assembly is formed from two shallow conical segments (about 1/8" or 3mm deep) backed with a bulkhead and fronted with a louver assembly for the air cooled radial engine. A connecting strip is used to mate the two sections. Inside this assembly is the engine mockup (representation of 9 cylinders backed with hardcard and slightly rounded) and the cooling louver. (See picture below.)

If you want a rotating propeller, then retrieve part 1g (earlier cut from the center of bulkhead 1h), and drill a center hole in it and the other components of the nose assembly. Build the assembly, but do not mount it to the fuselage at this time. Color the bulkhead at the rear of the nose assembly so it doesn't show white through the louvers. The most difficult step is that of fitting the fixed outside louver to the front of the nose assembly. A lot of fitting, adjustment and touch-up will be required. Make sure the top louver is correctly aligned with the centerline of the plane (mine was not). The propeller and hub assembly is straightforward. Once completed, a pin can be inserted through the nose assembly and glued to the propeller hub.

Landing gear:

Diagrams for the tripod landing gear are quite detailed and very helpful. I used .020" (about .5mm) piano wire for the internal supports, and rolled the struts tight enough to fit snugly on the wire. These were trimmed according to the diagrams, and glued in place without any significant problems. Some adjustment was needed to accommodate changes to the wheel well outline, and 2mm disks were added to axle sections of the wheel struts to lock the wheels in place and permit them to rotate. There is only a single template from which to cut the segments used to build up the wheels, and a draftsman's compass fitted with a knife blade really helps. I had to cut 20 circles out of posterboard to form the two wheels (each about 1/8" thick). The wheels were cut, drilled, glued, sanded, colored and polished before mounting. The wheel well coverings, mounted on the outside of the struts, went into place without a problem. Make sure that the reinforcing wire in the struts extends beyond the end of the strut only enough to insure a secure mounting. If it is too long, it may puncture the upper surface of the wing. Stiffen the bottom of the tailwheel assembly with cyanoacrylate glue before mounting. This will prevent the wheel support from bending sideways.


The windscreen is made up of two frames, each backed with thin acetate (not provided). It is a delicate arrangement, and very difficult to assemble and mount while maintaining the correct curvature. I think a single-piece assembly would have worked better. This is the one part of my finished model with which I am most unhappy. I scanned the windscreen frames and templates, and will probably try to tweak the image and rebuild the component in a single piece.

Final touches:

Guns were rolled, colored and installed without a problem. The pitot tube was made from .025" piano wire and painted gunmetal (don't know if this was the right color).

The model required a fair amount of touch-up coloring, mainly with the nose assembly, the undersurface of the wing (where surgery had to be performed) and the usual exposed cut edges. I found it difficult to match the colors used in the kit. The problem, which I hadn't run into with other models, concerned the behavior of paints in different lighting situations. In this instance, the mixture I used for touch-up in the olive drab / topside green areas, worked well in the mix of incandescent and fluorescent light in my work area, but became a much greener color under camera flash and sunlight. (This is apparent in the photos.) I also tried using acrylic paints for the first time on this model, and found that they were too glossy in comparison with the printed card stock. I may decide to go back and redo some of the touch-up work.

I did not install an antenna because I could not tell from the information available how it was configured in the original. Once I find out, I'll put it in. Since I'll probably replace the windscreen, it seems best to leave the antenna off until the rework is complete.


For all the frustration, I'm pretty satisfied with the result. But it's the kit that frustrates -- the finished model and its original that satisfies. In retrospect, I'm probably getting spoiled by the precision of design, detail, coloring and printing exhibited by the newer computer assisted kit designs from Digital Navy, Halinski, GPM, and others. Fly Model's I-16 is a little more "hand made," and it suffers from a few inaccuracies. When I think about how much of a nightmare this kit must have been to design (wheel wells that transitioned from wing to fuselage, for example), the result is very, very good.

But above all, it's the plane itself that appeals to me. Short, fat, stubby, lots of wing -- it almost looks like a GeeBee with guns. And what other fighter could be known as a rat, a fly or, strangest of all, a little donkey?