ICM 1/72nd Tupolev TB-3 by Michael Kendix

Background

An in-box review of this kit and a brief background on the TB-3 appeared in the October 2000 edition of Internet Modeler.

Prologue

My relationship with this model began a long time ago. I bought this kit and the aftermarket photoetch set from Eduard several years ago–each cost about $20. At various intervals I would open the box, gaze at the large stack of green sprues and the instructions, and then replace the box lid. For a fairly long time, I never saw this kit built but gradually I began to see one or two appear at model shows, and the occasional one on the web. One thing that struck me as an avid contest modeler and judge at said events, is that none of these models was ever likely to win me anything at IPMS events in the United States – it appeared the build was just too difficult and I could never build the kit without making too many minor mistakes. However, at some point I saw Aleksandar Šekularac’s amazing rendition of this kit on this web site. Of course, he had gone the whole hog: not only had he used the Eduard aftermarket set but he had added a couple of I-16 type 5’s from kits made by A-Model as parasites under the wing surfaces to create a Zveno. In fact, ICM does provide a Zveno version of the TB-3 but the I-16’s in that kit are Type 24’s, which is not really correct for this TB-3. Of course, this is understandable since ICM makes Type 24’s and not Type 5’s. However, I digress, the more I looked at this model, the more I thought about building it since the alternative was to let it sit in the box forever, plus I had to justify forking out $20 for the Eduard aftermarket set.

The Wing

When I started building this model, I had to make some decisions about certain issues. First, how much extra detail/trouble was I prepared to include here? Second, where am I going to start? I decided that the wing was going to be the trickiest part, and if that didn’t go together well, there would be no point continuing, so back in March 2006, I started assembling the wing. I cannot over-emphasize how much I relied on Aleksandar Šekularac’s pictures and description on the VVS web site because without that, I would have been dead in the water, so to speak.

The wing has about 25 pieces for both the port and starboard sides. I decided I wasn’t going to do the “Trashcan” gun emplacements on the underside of the wing so I blocked those off. The multipart wing contains all the spars and interior components that you would find in the real thing but of course, they don’t show once the thing is assembled. The skin too is in multiple pieces–more than 10 in total, I think. Everything has to be dry-fitted really carefully because as you can obviously see, the entire skin of the aircraft is corrugated, which precludes any kind of regular seam filling and sanding. I managed to fit the various spars together and I glued them to the underside skin. Once that was dry I attempted to add the upper skin of the wing but it didn’t fit! A panicked email to Aleksandar Šekularac and a rapid response revealed that this was the case for him too and that I should expect plenty of fit problems. I pulled out my trusty Dremel tool, flipped the speed to 10,000 RPM and proceeded to sand and melt off a goodly portion of the spar structure, pausing at various intervals to test to see if the top skin would fit. At this point I recall that the plastic was veritably flying off the spars and burning my arms as bits of the melting residue landed on my skin. After an evening of this and with the help of some wonderful adjustable clamps I had bought at the 2005 US Nationals, I managed to get the wing together. Prior to gluing everything down, however, I sprayed the interior of the wing black because the photo etch engine access wing cut-out piece will allow you to see inside the wing just a little. In order to see how things had gone, I sprayed this huge subassembly with Mr. Surfacer. It didn’t look terrible but there were definitely some small gaps and a few glue spots. I couldn’t really think how to get rid of these because of the corrugated surfaces so I put it into a plastic bin container and left it there.

I don’t know what happened exactly but when I pulled it out of the bin about three months later, it didn’t look so bad. During that time I had polished off two other kits and I was ready to start back again on the TB-3.

Engines

One of the less interesting aspects about large aircraft models is that there is multiples of everything and in this case, there are four engines, so the process of construction is somewhat repetitious. Please don’t misunderstand me, the kit’s engines are excellent, especially considering that almost nothing of them can be seen once the engine cover is glued on. To cut through this potential boredom, I resorted to the Eduard photoetch set, which contains some nice detail for the rear section of each engine and one engine cover that can be raised to show the engine detail for one of the engines. While the engine detail is more than sufficient when the cover is closed, this is not the case when the cover is in the open position. I rectified this by scratch building an engine to go into one of the engine compartments, relying heavily on Mr. Šekularac’s build and some references on the web that he showed me. I used the kit’s engine block as a basis and added the various cylinder parts and other features using pieces of plastic rod and card, and I made the tank behind the engine using the same. One problem with the photo etch engine cover is that the configuration of its holes is different to those of the kit engine covers. In order to match them all up I followed Mr. Šekularac’s advice and “moved” the kit’s holes. Part of this little open-engine vignette entailed installing the photoetch for the access step next to said engine. In order to do this, I had to cut out the appropriate section of the wing’s leading edge and insert the photoetch pieces. However, at this point I only installed the wing piece, not the hanging-down step itself, which would have broken off too easily during the building process. The engines themselves are nice enough but assembling their housings was not straightforward. Each engine housing has six outer pieces, all of which are butt joined and must be carefully prodded in place. Once glued, various gaps were apparent and other panel lines were lost. It took several sequences of filling, spraying with primer and re-scribing to bring these components up to an acceptable standard. Once this was completed, I glued the engines into their positions in the wing. Once again, there were significant fit problems but one has to recognize that such problems are an inevitable price for the way the kit is constructed: it’s almost impossible to align the pieces during construction so that such gaps will not occur. As mentioned above, you cannot simply fill the gaps and sand the area flush because the entire wing and fuselage surfaces are corrugated. In order to deal with this I followed the advice of Mr. Šekularac and used thin strips of plastic, which I glued into the gaps. Once again, there were several round of filling, prodding and primer spraying.

Fuselage

The construction of the fuselage was the most interesting and satisfying part of this build. The detail is incredible even without the aftermarket photoetch but with it, you can really dress up the main cockpit area, and this is really the only interior area that can be seen once the model is finished. The photoetch seatbelts, instrument panel and other main cockpit components will all be in plain view so it’s great if you can add these various extra pieces. A great deal of detail is in the forward section underneath the nose gunner’s position, however, even though there are a lot of clear parts surrounding this area, little of the inside can be seen once the fuselage is assembled. The fuselage components, including the outer skin and all nine of the bulkheads went together quite well. Naturally, I had to use extra care dry fitting these pieces but compared to the wing, this part of the build went quite smoothly. The main problem I faced was gluing the straight fuselage sides to the curved fuselage base. In order to do this I used those clamps again and glued the parts together in stages. Starting with the forward area, I glued that section, clamped it and waited for that to set-up. I repeated this sequence working my way gradually to the rear. At this point you also have to assemble and glue in the front section of the tail and once again, care is required in assembling this part because it is not completely obvious exactly how the pieces are to fit with each other. One problem with the forward part of the horizontal tail was the lack of any decent attachment point, which meant this thing kept on coming unglued.

Before adding the fuselage top, I glued in the clear parts. Unfortunately, every single one was too big, which gave me a choice: either widen the window frames or reduce the size of the clear parts. I chose the latter, frequently dry-fitting the piece each time I sanded it a little. I glued these in with Elmer’s white glue so as not to make a mess of them and I was fortunate that they all held up to the stress of masking them and painting.

Once the fuselage was completed, I glued it onto the wing, which was far from straightforward. The over-engineered construction of the wing and fuselage components makes it easy to allow a slight asymmetry to creep in, which lead to significant gaps at the wing-fuselage joins. Again, I glued in slivers of plastic to fill these gaps, and the process of filling, sanding a little then spraying with primer to ensure that all the gaps were filled properly, took some time. Finally, I added the large clear part that goes on the front of the fuselage. This clear part fit well but I was uncertain whether the underside of this contained windows or not and I decided it did not, however, further investigation has lead me to believe I made a mistake here. Although it would have meant that people could actually walk out over and onto the glass windows, I think that’s how it was.

Painting

After much consideration, I decided to paint the model as “Blue 9”: a scheme that comes with the kit–250 APDD, summer, 1942–a somewhat later version than some of the other kit schemes. The upper surfaces are AII black and AII green and the lower surface is AII blue. There is some uncertainty as to whether the red stars were outlined in black and whether they appeared on the wing’s upper surface. I chose to do plain red stars and omit the upper wing surface stars – more on the decals later.

After stuffing the gun emplacements and cockpit areas with tissue paper, I painted the underside using Testors flat blue, which I lightened with a little white. At this phase, it looked a little bright but I eventually toned it done with an oil wash later in the building process. Once this had been done, I masked off the blue and sprayed the black for which I used straight black lightened slightly with a little white. At this point, the model looked like a sort of night fighter. Once the black was dry, I masked and sprayed the green for which I used Model Masters US Interior Green. Despite the corrugations, masking with 3M blue masking tape went well, and I only needed to do a few touch-ups.

Decals

The decals for this kit are not that great, which contributed to my decision to not bother with the upper side wing decals. I put one of the star decals in a bowl of water to test its structural integrity and it just about disintegrated. I took some Microscale Liquid Decal Film and brushed a couple of coats carefully over the decals I planned to use. I trimmed the edges of the decals as much as I could and brushed on a few coats of Future to prepare the surface for the decals. Those on the underside went on fairly well but the fuselage stars suffered from some discoloration, which I was able to disguise in part when I gave the model an oil wash. The “9” decals on the tail went on relatively easily though some crumbling of their white surround meant I had to touch them up with a little paint.

Peripherals

The bits and pieces that one tacks onto this model are excellent. The wheels and bogies are excellent–each side is a little kit in itself. My research showed that this version of the TB-3 had wheels with spokes. Using the method suggested by Aleksandar Šekularac, I glued together the wheel halves, drilled out the centers, filled the tires with Squadron White Putty and glued on the wheel spokes from the Eduard photo etch set. The landing gear struts presented some problems. I had to replace some of the ends of the struts in order for the parts to sit right. I also had to replace the piece that goes through the wheel subassembly with brass tubing so it didn’t bend and make the wheels splay outward.

I used the photo etch set for the open engine cover and the engine maintenance access step. I glued on the bomb racks but it was clear that the bombs would have interfered with the engine access step so I left the bombs off. The photo etch set also provides a number of replacement pieces for the guns and gun mounts, however, I simply could not get these together so I used only the gun sight. I created a wood grain appearance for the propellers by taking burnt umber and dark brown gouache water paints and swiping them onto the surface with a Q-tip. I then sprayed them with Tamiya clear yellow and finished the surface with a coat of Future floor polish. The propeller blades have metal edges that I painted with Testors metallic brass paint. I hollowed out the engines’ exhausts with a number 11 blade on my Exacto knife and a drill, and added the rigging on the tail and the antennae using 0.005-inch steel straight wire from Smallparts. Inc. I gave the entire model an oil wash with burnt umber oil paint thinned with Turpenoid and then some light weathering with pastel chalks for the stains behind the engines’ exhausts.

Conclusion

I have about 75 built models on my shelf including several vacuform models and an Ilya Mourametz but this was the most difficult kit I have built. The key to building it was perseverance and patience. It simply cannot be rushed without it being ruined and even then ultra-care is required. I also had a lot of helpful advice and as mentioned above, I could not have done it without being able to follow the web build of Aleksandar Šekularac. My finished model is no contest winner because it contains far too many blobs and scrapes but it is an extremely impressive-looking model. I recommend it conditionally for those who really must have one of these in their collection.

Acknowledgements

I would like to thank the following for their invaluable advice and help. In alphabetical order, I would like to thank Matt Bittner for his advice and general encouragement, Michael Lampros for references, Erik Pilawskii for advice and references, Aleksandar Šekularac for advice and his phenomenal build on the VVS web site, and various members of the VVS web site forum for their encouragement.