The RC3-2 in 1:72: The fastest thing in the world, 1925 edition


The 1925 Schneider Trophy-winning Curtiss R3C-2, as flown by Jimmy Doolittle

The Coupe d’Aviation Maritime Jacques Schneider, better known as the Schneider Trophy, helped propel aviation forward in the early years of flight. Sponsored by the scion of the Schneider family, French makers of steel and armaments, the race reflected the younger Schneider’s view that seaplanes were the wave of the future, since they didnot require land bases and expensive runways.

The race called for a timed competition over a triangular course of at least 150 miles. The winner of each race would receive a £1,000 award and ownership of the Schneider Trophy until the next race. Should one nation win the award in three consecutive races within three years, it would win permanent possession of the trophy.

The race became the focal point for national pride as French, British and Italian entries traded victories. In 1924, the U.S Navy jumped into the race with a converted Curtiss CR-2 landplane and won the race at a speed of 173.347 mph.

Because of the win, in 1925 the race came to America. The 1925 race was scheduled for Baltimore, MD, between September 19 and 21, but neither Britain nor Italy had any aircraft ready at that time. The Americans delayed the race until October 23–26, giving time for Italy to prepare two Macchi M.33 flying boats and the British to send their new Supermarine S.4 and a Gloster III. , During a trial flight before the race, the S.4 developed aileron flutter and crashed into the Chesapeake. Though its pilot survived, the S.4 was a write-off.

Up against this competition was a trio of Curtiss R3C-2s, two from the Army and one from the Navy. The R3C-1 landplane had won the Pulitzer Trophy race just two weeks earlier, when Cyrus Bettis flew the aircraft to an average speed of 249.8 mph. Powered by a 610-hp Curtiss V-1400 engine, the floatplane version was called the R3C-2. Three R3C-2s were entered, flown by Lts. George Cuddihy  and Ralph Oftsie of the Navy and Jimmy Doolittle of the Army.

Doolittle was already an emerging aviation superstar, having earned the first-ever doctorate in aeronautics from MIT earlier in 1925. He found the R3C-2 a sweet-flying plane, by getting it off the water was a little tricky, with a long takeoff run before suddenly coming un-stuck from the water. “I would taxi out at 90 degrees to the take-off area, pull the stick into my stomach, advance the throttle until full, duck my head down into the cockpit to avoid the water spray from the prop,” Doolittle wrote in 1988. “The plane would run to the left 90 degrees before reaching take-off speed, and all of a sudden it would break loose, pitch up and I would push the stick forward to level off.”

Doolittle had thought race tactics through before lifting the R3C-2 into the air. As he orbited the triangular course, each time he approached a pylon he would climb under full power, then make a steep, banked diving turn, which he believed gave him the critical speed advantage. He also looked after the engine – which is why he finished the race while Cuddihy and Oftsie did not.

Doolittle won the race with an average speed of 232.57 mph, besting Hubert Broad in the Gloster III and his 199.16 mph speed – a margin of more than 33 mph. The next day, with the same plane on a straight course, Doolittle reached 245.7 mph, a new world record for seaplanes. It also cemented the Curtiss biplane as the fastest thing on earth at the time – the land speed record at the time was 150.87 mph – and earned Jimmy Doolittle the reputation as the fastest racer on earth. Doolittle would be the last man to win the Schneider Trophy in a biplane – and the last American to claim the title.

Doolittle would win the Bendix Trophy race from Burbank, CA to Cleveland in the Laird Super Solution in 1931 and the Thompson Trophy in 1932 flying the Gee Bee R-1. After completing this triple crown of air racing, he retired from racing. “I have yet to hear of anyone engaged in this work dying of old age,” he quipped.

The R3C-2s would receive new engines for the 1926 Schneider Trophy race and finish in second, but by then military enthusiasm for racing had waned in the U.S. It wasn’t until 1931 and Reginald Mitchell’s S.6b that the Schneider Trophy found a permanent home. Doolittle’s R3C-2 also found a permanent home in the Smithsonian Institution.

Since Doolittle was a hero of mine since childhood – and since I live in Alameda, the town where he was born – building Doolittle’s planes holds a special attraction. Building exclusively in 1:72 scale meant my options for an R3C-2 were limited, with my best bet being the CMR resin kit. This is not a kit for the faint of heart – some of the parts don’t fit particularly well, there’s no cockpit to speak of, and the instructions are so rudimentary they suggest what you would have found in the caves of Lasceaux had models existed in the Stone Age. The packaging is dubious as well – the cover sheet in the bag has a nice profile of one of the Navy’s R3C-2s, with its pale blue fuselage and chrome yellow floats. The decals provided allow you to do the Army racer, with a dark blue-black fuselage and floats. Markings provided give you Doolittle’s racer and the Pulitzer Trophy-winning R3C-1 flown by Lt. Cyrus Bettis. But, for the most part, the castings were bubble-free and straight – and that made it a fairly easy model to build, once you figured out where things went.

I jumped ahead to the end for my first step. Looking at the rigging diagram, I planned where the wires would enter the fuselage and wings and carefully drilled tiny holes with a No. 80 bit in my pin vise. I did not have a good reference, so I translated as best I could from the instructions. Later, I would find that I’d missed a few, but the holes gave me a place to start. Precision was the watchword here – any variance from left to right would mean the rigging would look out of whack at the end.

The cockpit came next. The real aircraft’s interior was all wood, with instruments fastened where they needed to be. I made a seat, an instrument panel (such as it was) and throttle assembly, and made a small seat to replace the backless lump given in the kit. WWI-style belts from my spares box went on the seat, and I painted the interior a wood color. There was no need to go crazy with wood grain or anything else, because the cockpit opening is very small. I decided not to open the cockpit side panels, because that would have meant major surgery on resin parts, and cutting into resin often leads to unpleasant discoveries in the form of air bubbles and other issues.

This is in there. I swear.


The fuselage halves were joined and I was faced with fit issues at the tail and the nose – a bit of warp had afflicted one side. A dip in a little hot water and some thumb pressure straightened the bend, but there was still significant seam work to be done, especially around the tail cone and the bottom of the rear fuselage, which I used a shim of styrene strip and CA glue to remedy.

The vertical fin and horizontal stabilizers were provided as separate pieces. The vertical fin went first, with copious amounts of CA being used to blend it in to the contours of the fuselage. The horizontal stabilizers were butt-joined to the fuselage, and I had to take care to make sure they were aligned to each other exactly.

I put the vertical fin on first – and checked that it was perfectly aligned to the fuselage – so I could use it as a guide to get the wings on straight. Having built a Curtiss F9C-2, I had a process for building shoulder-mounted biplanes like this one: start with the wing that fits worst. That way, you’ll be able to attack the fit issues from more angles without that second pesky wing getting in the way. In this case, the top wing fit the worst, and getting it to fit would affect some fuselage details. I CA-glued it in place, then went after it with sanding sticks of various coarseness until it was blended to the fuselage, taking care to ensure it was aligned at a right angle to the fuselage and vertical tail and aligned with the horizontal tails. A trick here is to look at the model head-on, then tilt it (nose down in this case). As you sight along the fuselage, the tips of the horizontals should disappear at the same time.

The R3C-2 had a small teardrop-shaped fairing atop the nose to accommodate the engine. CMR molded this fairing in three parts – one half on the wing, and a quarter on each half of the fuselage parts. While putting the wing on, I simply eradicated the fairing, then made a styrene strip replacement.

The floats came in halves, and spilt side-to-side. I put them together, addressed the seams, and made sure they were symmetric, since they’d throw the geometry of the finished model off otherwise.

The fuselage detail – minimal as it was on this streamlined racer – was carefully restored with a scribing tool. I had to pay attention to where the upper wing interfaced with the fuselage, especially around the cockpit opening, and I carved away some rough moldings to get it to look right.

The bottom wing went on next. The fit here was also kind of rough, but most of the sanding work was confined to the bottom of the aircraft, not the area between the wings. Again, alignment was a priority – I repeatedly checked the lower wing against the top wing to make sure it didn’t twist and that the distance between them was the same from right to left. Once the lower wing was successfully blended to the fuselage, I painted the radiators with ModelMaster Metalizer brass – and then put the model away for six years!

Brassy! ModelMaster metalizer brass proved to be fairly tough and stood up well to masking.

Eventually, during our shelter-in-place interlude caused by the Covid-19 breakout, I decided to tackle the model. It had been kept in a wooden cigar box, safe from accidents or breakages, and it was easy to resume the work. I did a little clean up on wing-to-fuselage seams, then decided it was time to paint the outer wings. These were “chrome yellow” on the real plane, which at the time was a bit more saturated than the modern version of the color. ModelMaster deep yellow was a good match. I masked the radiators and sprayed the outer wings yellow. I also sprayed the elevators at the same time.

Before you paint yellow (or red), lay down a coat of white…

…And your color will be much more vibrant than if you spray the paint on gray plastic (or beige resin).



While that dried, I assembled the floats. I was missing one of the resin cross-braces, so I swapped in some styrene strip in its place. It fit neatly into the recesses in the pontoons and I was assured that both cross-braces were the same length! Once these were in place, I added the resin vertical braces; these were a bit harder since they butt-joined the floats and there were no locators on the body of the model. The rear braces met at the base of the fuselage; I built these first, then aligned the forward braces off of them.

Mess up the geometry here, and your plane will never fly…

With the wings now dry, I masked them and the elevators and stuck a bit of wet toilet tissue in the cockpit opening before painting the fuselage and floats a very dark blue. This was made by adding a bit of cobalt blue to flat black – there’s a hint of blue in there if the lighting is right. The floats were somewhat challenging – you had to spray from an assortment of angles to get full coverage of every surface. Once the fuselage had dried, I carefully painted the exhausts silver using metallizer aluminum and a tiny pointed brush.  I had also cleaned up the interplane struts and painted them blue-black; these fit very precisely, and I was able to wedge them into place before running a little super-thin CA along their bases. I also drilled a small hole into the right wing for the pitot tube, which would be added at the very end of the build.

How much blacker could it be? A little, to be honest.

The propeller and spinner required a lot of attention. The casting was very imperfect and a strange, irregular blemish stretched along the spinner between the blades on both sides. Once this was eradicated with sanding sticks, I knocked off any flash on the blades and polished them in preparation for a metal finish. Ultimately, I airbrushed aluminum metalizer over the entire assembly, then carefully brush-painted the spinner before giving it a coat of Pledge Floor Magic (or “Future,” as we modelers know it).

The entire model was then given a coat of Future, the floats separate from the rest of the plane. The real aircraft was rather glossy, and the Future was a great base for the decals. I was apprehensive about the kit decals, but I had no reason to be – they performed spectacularly. I was really concerned about the tri-color tail flash with the tiny “Curtiss” logo in the white field; I was worried the white would turn into a muddy gray against the black background. There was no reason for concern!

Kit decals on…


…And looking pretty good!

Now I had a mostly finished plane and a set of floats – they needed to come together somehow. The rear attachment point was approximately even with the rear of the cockpit; it was the sturdiest of the mounting points, so I added the floats starting there. Once I thought I had it in place, I used CA to attach the forward struts to the fuselage, a difficult task since the real plane’s attachments were very small. Scale that down to 1:72 and you don’t have much surface area for glue. When I thought I had it right, I checked it – the plane was leaning one or two degrees to the left, making the lower wing clearly out of alignment with the float cross-braces. Gentle coercion and a little more CA made everything geometrically even again.

Look at all the perpendicular and parallel angles – an IPMS judge’s dream come true!

Before attacking the rigging, I scratch-built the windscreen. This was basically a glass tunnel with three flat sides; the kit provided a template for you to make your own. I used the plastic from the blister packaging for two AA batteries as my source material. I traced the diagram with a No. 11 blade, then cut out the windscreen and bent it along the scored edges with a pair of flat-nosed pliers. The first attempt was a little short, so I adjusted for the diagram’s shortcomings on the second try and got a piece that fit neatly. The frames were created by running the edges of the plastic windscreen across the tip of a Sharpie pen.

Now it was time for the point in the build I was dreading: the rigging. The aircraft had just two wires between each wing, but the floats were attached with five wires on each side. The forward-most wires ran to the upper wings; others cross-braced the struts, and the middle set ran to the lower wings. I had run across some .5mm nickel-silver metal wire from Albion Alloys that I’d bought for some purpose, and I thought it might be useful as a rigging alternative. This would be a chance to test it out.

Luckily, at around this time, I found a great reference from the Smithsonian Air & Space Museum – a set of images of the R3C-2 on display, showing the aircraft from all angles. They can be found here . The location of all the rigging was no longer a mystery!

I carefully measured and cut lengths of wire to match the distances between the rigging points. Precision was the key: too short, and the wire wouldn’t connect and just dangle; too long, and the wire would deform. Testing and trimming until I had the right length, I started with the inner float braces and worked from the inside out, securing the wires with small amounts of Woodland Scenics scenic glue. This dries hard, clear and flat, but is easy to remove during assembly if you make a mistake and clean it up quickly. Where I had a hole to work with, I’d glue the wire to the edge of the hole and extend it up to its terminating point, checking constantly for alignment. The last to go on were the float-to-upper wing wires. I found the .5mm wire would sag over long runs – but nothing on this model is more than three inches! In that short space, as long as you don’t impart a bend into it, the wire will remain stiff and taut.

This view show most of the rigging lines in place. Build from the inside out!

The last things to do were to add the pitot tube (a .4-inch length of Albion Alloys tubing, painted silver) and the windscreen, which was lowered into place with tweezers and then fixed in place with a brush-applied bit of Future.

The windscreen in place, showing off its Sharpie-ink framing.

This was my third all-resin kit, and my second Curtiss shoulder-wing biplane. The tricks I’d learned came in handy, but really this model with primarily about anticipating problems with geometry and heading them off before they happened. It may be a small model, but it’s one of my favorites!

This view show most of the rigging lines in place. Build from the inside out!