Below Follows a report extracted from the Modell Flugsport 3/2007 magazine. Quite long, but well worth the read!
Reproduced here with kind permission from Reto Fiolka reto_fiolka@yahoo.com the original author: 
What Is F3B?
F3B is an FAI competition class for remote controlled (RC) sailplanes that covers the full bandwidth of performance radio control gliding. The three tasks, thermal duration, distance flight and speed flight pose contradictory requirements on the pilots and models, because one has to perform all these tasks with the same model. Effectively there is even a fourth task, namely the launch at the winch. The launch is the beginning of every task and a perfect winch launch is absolutely mandatory in F3B: “Launching is everything” (quote by Tobias Knoblauch!)
These four tasks require the optimization of completely different flying states: flying with maximum lift during launching, with minimal sink in thermal duration, with the best gliding ratio in the distance task and with minimal drag in the speed task.
Beside the flying skills of the pilot and the performance of the plane, the weather also plays an important role in F3B as thermal activity can sometimes change quite quickly. To avoid “weather luck” in an F3B competition, the thermal duration and distance tasks are flown in groups (typically up to 12 pilots in thermal duration, up to 6 in the distance task). The best pilot of each group scores 1000 points, the others proportionally less. Therefore the influence of the weather can be compensated; the winner is the pilot who can most successfully handle the current conditions.
In contrast, the pilots have to fly one after the other in the speed task and the pilot with the fastest time scores 1000 points. Of course, the weather conditions can strongly influence the speed performance; therefore each pilot has a working time of 4 minutes in which the speed flight has to be accomplished. Thus an experienced competition pilot can choose - within some limits - the best moment to start. Only if the weather changes drastically (e.g. when it suddenly starts to rain), can the organizer split the speed task into smaller groups.
What Is the Challenge?
The reader might ask himself what is so special about these rather simple tasks in an F3B competition.
Nearly every modeler has experienced successful thermal flights, all the F3B ships feature superb gliding ratios and flying fast with them is a piece of cake anyway, so what is so difficult about F3B?
Well, when there is good thermal activity, the duration task is not that difficult, but what about difficult conditions, extremely poor air, turbulent wind or just really bad weather? It would be very naive to hope that in such conditions the other pilots will also not be able to fly the 10 minutes, because at least one pilot in each group will somehow manage to fly the full time.
Furthermore, let's assume that the models in the distance task have exactly the same gliding ratio (although flying a plane at the best gliding ratio requires a skilled pilot and good settings) and the weather conditions are constant (e.g. calm air in the evening). Despite the equal conditions, a beginner in F3B would normally not score the 1000 points. This can be very disappointing, because one cannot directly see what went wrong, either the launch height was not good enough, the turns were not optimal or something else was not perfect.
But the distance task is even more challenging, because the weather conditions are mostly not at all constant, thermal bubbles rise in some locations, but a little later there is only poor air left.
Therefore the distance task is a four-dimensional optimization problem, because it matters where and when one starts the flight. Furthermore, it is allowed to land and launch the plane again if a pilot realizes that he is flying in the wrong location.
Because there are so many possibilities, a pilot alone could not fly an optimal distance task by himself - he would for example not notice that other pilots have found good thermal activity. Therefore team work is required in this task and the pilots advised by up to three helpers. Their job is to observe the other pilots, indicate the flying speed and to announce the turns at either end of the course. All in all, the distance flight is a very tactical task that, beside optimal control of the plane, requires a lot of experience from the pilot and the helpers.
But the speed task must be easy, F3B planes are fast anyway and we fly crisp speed turns each weekend at the slope, right?
Well flying an optimal speed turn is a different story, because the plane should conserve most of the kinetic energy, but flying this turn exactly at the end of the course, that is the real problem! Each meter the plane flies too far is doubled, because one also has to fly this extra distance in the opposite direction. On the other hand, if a pilot turns too early, no signal will sound and he has to fly back again, which costs a lot of speed and time.
Additionally, the speed task is often the most important task to separate the top pilots. So imagine the following situation: The last speed task is run in an inverted sequence (the best pilots start last) and the pilot that flies before you achieves an excellent speed flight.
Now you know that you have to perform equally or even better to keep your position in the final ranking. And the pressure increases the nearer you are to the top pilots - at the 2005 World Championships the very last speed flight decided who became world champion!
Beside that, the speed task is very spectacular because the planes fly at very high velocities (up to 160 km/h), so don't make mistakes in this task; otherwise you have to pick up your plane in very small pieces!
And Why Should One Do That?
So far we have learned that F3B requires a lot of training, is extremely challenging and is also not very cheap. So why are there people doing F3B?!
F3B is the challenge and the fascination of flying against pilots from all over the world and of raising the limits higher and higher. Subjective impressions about personal flying skills are replaced by solid facts, because individual performances in the F3B tasks can be quantified exactly.
When one flies alone, one can never tell if the flying performance was good or bad, because one has no reference. In an F3B competition, one gets a direct feedback - either you have the 1000 points or the other pilots were better than you!
Of course this can be very disappointing in the beginning, but when you score the 1000 points for the first time, it is a rush! And you can share your joy with your team mates, because F3B is a team sport and you won't make it to the top alone.
Colleagues and friendship are also an integral part of the F3B scene, because it sometimes requires extreme situations to find really good friends! A lot of good friendships, collaborations and networks have been developed in the history of F3B and that should not be underestimated.
Furthermore F3B helps you to increase your flying skills because you learn from the best. Being able to find the right settings for a model, to read the air and to fly precisely will payoff quickly because all of a sudden you will fly longer and get more performance out of your sailplane.
Thermal DurationThe competitor has to fly for exactly ten minutes, and has to land exactly in the centre of a circle with 15 meters radius. This task is similar to F3J, but there are some differences: In F3B the pilots have a working time of 12 minutes, so there is no need for risky simultaneous starts and landings as in the F3J class. Furthermore, F3B planes achieve such excellent launch heights that they can effectively just soar the 10 minutes in calm air. Nevertheless, this task can be extremely exciting in bad weather, because 10 minutes can then become very long...
Distance-Task 
The pilot has 4 minutes in which to fly as many legs as possible of a 150 meter course. The working time for this task is seven minutes and there is the possibility to restart the plane if needed. The world record in this task is 34 legs, nevertheless in bad weather conditions sometimes 11 legs can be enough for 1000 points.
Speed-TaskIn a working time of 4 minutes the pilots have to effectuate a speed flight of 4 legs, each leg having a length of 150 m. There is the possibility to restart the plane within the working time.
The world record is 13, 8 seconds which corresponds to an average velocity of 156 km/h.
The Optical Reference 
To check that the models fly the correct distance in the distance and speed tasks, an optical reference system is placed at each end of the course: Two wires are arranged so that they span the corresponding turnaround plane. By focusing over the two wires, helpers can determine optically when a plane arrives at the end of the course and at that moment they press a buzzer.
The Winch launch 
In F3B the planes are launched using an electric winch with a turnaround pulley. The power of the winch is limited and the overall length of the launching line (typically a nylon line with a diameter of 0, 8 - 1, 4 mm) has to be 400 m. The diameter of the nylon line is not limited and can be optimized for weather conditions. The tremendous launch heights that can be achieved with modern F3B planes originate from the enormous lift the planes can generate during the first phase of the launch. This places severe strain on the nylon line and it is stretched, so that at the summit of the launch, the plane can be accelerated by the winch and the nylon line itself.
After catapulting the plane off the line, the plane can rise nearly vertically by an additional 100 meters. Therefore launch heights of 350 m are possible in ideal launching conditions.
The F3B Plane
Technical RequirementsAn F3B plane has to fulfill contradictory and extreme aerodynamic and static requirements: When the plane is being launched it is first in a high lift regime, at the summit it is instantaneously accelerated to approx 180 km/h, then it is pulled violently off the line to rise by an additional 100 meters. This causes the airframe to be subjected to enormous g-forces, and it should furthermore feature exceptional stiffness in order to convert the kinetic energy efficiently into altitude (and not into deformations!).
The following flying tasks require optimal performance in a wide velocity range (9 m/s in thermal duration, up to 60 m/s in the speed task), so a wide range of Reynolds numbers have to be considered in the airfoil and plane design. The contradictory requirements need a lot of compromises, a plane that is only optimized for one task would have no chance in a competition.
History of F3B Plane Development
In the last 30 years F3B planes have changed a lot, new construction technologies (molding, glass fiber and carbon fiber reinforced plastic, GRP and CFRP) and better aerodynamic knowledge (computer based airfoil design) have revolutionized the scene.
The first planes were only controlled by elevator, rudder and spoilers and the wings were constructed with ribs or sometimes in sandwich technology. But at the end of the seventies the first all-molded GRP planes appeared with this technology, one could reproducibly manufacture the planes with high precision. Carbon fibers soon enabled significantly stiffer airframes and a further reduction of the all-up weight. In the mid eighties Rolf Girsberger developed the RG 15 airfoil which was an aerodynamic breakthrough. Finally an airfoil was available that combined good speed performance with sufficient lift for the launch and thermal duration. Several World Championships were won using the RG 15 airfoil and the first commercial F3B planes also employed this airfoil (e.g. with the Ellipse 1 or the V-Ultra plus).
With the arrival of the commercial version of the Ellipse 1 in the early nineties, a competitive plane was available to everyone for the first time. Up to now, the market for F3B planes has continuously grown, but nevertheless there are still a lot of teams that very successfully develop their own planes, like the LOGO team or the Herrig brothers.
In the mid-nineties high-lift airfoils like the MH 32 were introduced into the F3B scene. These airfoils enabled stunning launch heights by optimization of the maximum lift. The slightly reduced speed performance was overcompensated by high wing loading and exceptional launch heights. Due to this development, the duration task was increased to 10 minutes, because the new planes could fly out the seven minutes too easily.
Today, the models are again optimized for the speed task and very thin airfoils have been developed. High aspect ratios are used to guarantee good launch performance and new ultra-high-modulus carbon fibers enable even stiffer airframes.
So Is It Only a Question of Money...?
One could draw the conclusion that today one could win a competition by investing the most money to have the best F3B plane. Fortunately this is not true, because the most important factor remains the pilot, “The best plane of the world is no use at all, if you fly in the wrong place” (quoted from Paul Bohlen!)
Of course, the planes are continuously developed further and there is always competition among the various engineers to design the best model. Nevertheless progress only happens in small steps and the “miracle airfoil” that beats all others by orders of magnitude does not exist.
So a normal modeler might be very disappointed if he buys the latest F3B ship and finds out later that it does not fly significantly better than, for example, an Ellipse 2. If one does not fly F3B competitions, the small gains in performance might be not that obvious.
The story is the same with the tailplanes: In the nineties the V-tail was successfully introduced in the F3B scene and it became the de facto standard. But at the end of the nineties more and more planes with a cross-tail were used in F3B and they were praised just like the V-tail was some years before. Today the trend is more towards the V-tail again. Apparently, it is more a personal question which tail type a pilot favors; therefore many manufacturers sell their models in a cross-tail and a V-tail version.
So the key to success in F3B is finding the right settings for the competition plane and extensive training. State-of-the-art F3B technology is available to everyone and it is not possible to just invest more money to have an advantage over the other competitors. This is important; one should just consider the development in other classes like F5B, where some pilots use new batteries for each flight to have a superior performance.
Off course the F3B equipment is not cheap, but you can use the planes for several years in competition. When the plane is no longer competitive (most times it is more the pilot and not the plane), one can use the plane for normal flying outside the competition. Because F3B planes are very durable and have an excellent performance, they are great fun on the slope.
State-of the-Art F3B Planes
Although the myth of the superior model was disproved in the previous chapter, some of today's successful F3B planes should be mentioned: Estrella, Diavolo, Europhia, Supersonic EX/R, Crossfire, Tool, Tresher, Ceres and Radical.
The Diavolo is the successor of the famous and widely used Estrella. It is a new design and features a relatively short tail boom, single dihedral and a thin airfoil.
A world champion model, the Crossfire is used by many pilots because of its exceptional launching and speed capabilities. A V-tail version is now also available to decrease the all-up weight of the plane. The designers of the Crossfire have already engineered a new plane, the Tool. This plane also features a very thin airfoil, is rather compact and employs a V-tail as well.
The Logo team's Tresher (successor of the famous Woebegone) and the Radical developed by the Herrig brothers and Martin Weberschock are fully competitive designs, the latter is even considered as one of the best F3B planes at the moment. Both planes are optimized for the speed and the distance task.
History of F3B
The first FAI sporting code for the F3B class appeared 1973. Essentially it described various kinds of RC sailplane competitions, namely for slope soaring, powered gliders and winch-launched gliders. For the latter, a duration task of six minutes, a distance task limited to twelve legs and a speed task consisting of two legs were described and this kind of competition soon became very popular.
F3B World Championships - Hall of Fame
Year - Country and Place - 1st - 2nd - 3rd
1977 RSA- Pretoria - Skip Miller (USA) Fricki Roos (RSA) Sean Bannister (GRB)
1979 BEL- Amay - Anton Wackarie (AUT) Ralf Decker (GER) Roy Spavins (RSA)
1981 USA- Sacramento - Dwight Holley (USA) Sean Bannister (GBR) Schaefer (GER)
1983 GBR- York - Ralf Decker (GER) Helmut Quabeck (GER) David Worall (GBR)
1985 AUS- Waikerie - Ralf Decker (GER) David Worall (GBR) Karl Wasner (AUT)
1987 GER- Osnabruck - Reinhard Liese (GER) Peter Hoffmann (AUT) Samuele Vilani (ITA)
1989 FRA- Melun - Nic Wright (GBR) Peter Hoffmann (AUT) Joris ten Holt (NED)
1991 NED- Terlet - Joe Wurts (USA) Daryl Perkins (USA) Stephen Haley (GBR)
1993 ISR- Kefar Sava - Denis Duchesne (BEL) Joe Wurts (USA) Klaus Kowalski (GER)
1995 ROM- Brasov - Daryl Perkins (USA) Denis Duchesne (BEL) Joe Wurts (USA)
1997 TUR- Ankara - Daryl Perkins (USA) Pasi Vaisanen (SWE) Joakim Stahl (SWE)
1999 RSA- Rustenberg - Daryl Perkins (USA) Dieter Perlick (GER) Roland Hofmann (SUI)
2001 CZE- Chrudim - Daryl Perkins (USA) Joe Wurts (USA) Stefan Knechtie (SUI)
2003 GER- Kirchheim/Teck - Andreas Bohlen (SUI) Pasi Vaisanen (SWE) Reinard Liese (GER)
2005 FIN- Lappeenranta - Andreas Bohlen (SUI) Martin Herrig (GER) Reinard Liese (GER)
2007 SWE- Emmen - Martin Herrig (GER) Andreas Herrig (GER) Fidel Frick (LIE)
Thirty years ago, the first F3B World Championships were held in Pretoria, South Africa. Most of the planes at that time were only controlled by rudder and elevator and had rib or sandwich wings. The Swiss team, for example, used a Brillant V and a Hobie Hawk. Today it is unbelievable how one could fly the speed task with such planes!
Two years later at the World Championships in Amay, Belgium, the models had evolved remarkably. The first all-molded F3B planes were introduced by the Austrian AME team, the Germans Ralph Decker, Dieter Pfefferkorn and the Italian Eugenio Pagliano. The AME team dominated the World Championships with their Dassel owing to its superior speed performance, rigorous training and a solid team performance. If one compares the models of the world champions from 1977 (Skip Miller) and from 1979 (Anton Wackerle) one can obviously see the tremendous progress within just two years!
This early success was the breakthrough for molding technology and accelerated the development of new models. The next step was the use of electric winches for launching, because one could gain superior heights in the zoom.
At the World Championships in Sacramento, USA, most competitors used electric winches. But there were no limitations for the win
ch, therefore they became more and more powerful. The most powerful winches had 4,5 kW of power and they catapulted the planes violently into the air. One could also install the winches freely on the competition field, so some clever pilots catapulted the planes directly into the speed course. The Canadian team used extremely powerful winches that were run at 36 Volts (also known as Gorilla winches) just to accelerate the planes. Due to the enormous forces, it could happen that a turnaround pulley was ripped out of the ground or that planes were destroyed during launching.
As this became more and more dangerous, the FAI had to limit the power of the winches by rule. The first idea was the so-called weak link, an element (typically a piece of nylon line) that would break if the line tension exceeded 40 kg. In 1989 it was decided to limit the internal resistance of the winch and the battery, a procedure that has lasted till now.
Despite this power limitation, the performance in the three tasks increased from year to year and the rules had to be adapted to this development: In 1983 the speed task was increased to four legs and in 1987 an unlimited number of legs was introduced in the distance task.
As the launch heights also became higher and higher, the duration task was adapted to seven minutes in 1993. With the introduction of high-lift airfoils in the mid-nineties, the launch height could again be significantly increased, consequently the flight time in the duration task was increased to 10 minutes in 2001.
Although the progress in F3B happens in small steps, over decades, the development is quite amazing. Today's F3B planes achieve around 250-350 meters launch height, a value that is better than the height obtained with a “Gorilla” winch, which had 4 times more power!
F3B is the
Formula 1 of model flying, because progress results from the competition situation. In thirty years, F3B has brought a lot of new technologies and normal modelers also benefit from this development, as better airfoils and excellent models become available.
Memories of the RG 15 Airfoil Development
The RG 15 airfoil is one of the most successful RC sailplane airfoils - for more than a decade it was the de facto standard in F3B, it was also widely used in F3J and even today it is used for fast slope soarers. But most people don't know its designer, Rolf Girsberger.
At the end of the seventies, Girsberger and colleagues decided to design their own F3B planes. Girsberger studied turbo machinery at the Federal Institute of Technology (ETH) in Zurich and he has good knowledge of fluid dynamics. He realized that the available Eppler airfoils were not optimally suited for the speed task in F3B. So he decided to develop his own airfoil that would perform better in the speed task but would also have enough lift for launching and thermal flying.
From scientific publications he realized that a computer program developed by Professor Eppler, was very promising for the airfoil design. The program calculated the airfoil via a conformal transformation to predesign the behavior at different angles of attack, a procedure that is known today as inverted airfoil design. The performance of the resulting airfoil was then verified by performing flow simulations using a boundary layer model and a panel procedure code.
Girsberger wrote to Eppler, who was working in the States at that time, asking if he could have the program. But Eppler was unable to send him the code, because of a strict US policy to prevent the export of know-how.
Fortunately Girsberger found a NASA Technical Report in the ETH library that had Eppler's code published in the appendix. So he decided to copy this code and implement a new programe. The technical report was copied several times, so there were finally a lot of bugs in Girsberger's code.
Therefore Girsberger was forced to understand the code down to the smallest subroutine in order to debug it. At the end, this paid off because Girsberger understood what was happening in the code and how the input parameters influenced the calculations. Finally, the code compiled and ran successfully, so Girsberger could start his airfoil calculations and he iteratively optimized his designs.
One has to imagine that the computers at that time were completely different from today's PC - they were programmed with punch cards and the results could not be directly visualized on a screen! So iterative optimization was a very tedious and costly task at that time!
The first successful result was the RG 8 and club mates quickly constructed a test model, that performed as predicted in the high-lift regime, but the speed performance was worse than predicted.
From this experience, Girsberger designed the RG 12 and the RG 14 which he considered as the most promising airfoil for F3B. Additionally he designed a more conservative airfoil as he had realized that a computationally over-optimized airfoil can lack performance in reality. This airfoil was the RG 15 and he published the results in 1983 for the first time.
Some top F3B pilots were interested in Girsberger Airfoils, among them Ruedi Binkert, who was designing the new plane named Quasar. To Girsberger's surprise they decided to try the rather conservative RG 15. In a short period of time, the RG 15 had spread successfully in the F3B scene.
One of the reasons was that Girsberger's airfoils were represented by a sufficient number of coordinate points, so its contours were quite smooth. So the RG 15 was ideal for direct machining, while other airfoil data sets featured less coordinate points and therefore required interpolation, an approach that could yield bad results.
But it would be unfair to reduce the success only to the available number of coordinate points, because the RG 15 was a really high-performance airfoil and a lot of competitions have been won using the RG 15 (won World Championships: 1993 Denis Duchesne with Ellipse 1 and 1997 Daryl Perkins with V-Ultra plus). Also the first commercial F3B planes, like the Ellipse 1, Tragi, Calypso or the V-Ultra, all employed the RG 15.
The RG 14 was also quite successful. Nic Wright won the World Championships 1989 in Melun, France with his Electra that employed a thinned RG 14. Despite all the success, Girsberger did not continue his airfoil development because he found other challenges. In the nineties, programs for airfoil computation (like Xfoil by Professor Mark Drela) became available and a lot of people started to design airfoils. Nevertheless, the RG 15 could last a long time against these new designs and was also considered as the benchmark.
The fact that Girsberger was able to design such a perfect airfoil with relatively limited computer power is quite remarkable and is evidence that he really knew what he was doing.
In summary, the RG 15 was a breakthrough in F3B and a lot of excellent planes were built using this airfoil.
Acknowledgements
The Author Reto Fiolka is grateful to Rolf Girsberger, Emil Giezendanner and Rudolf Schaub for fruitful discussions.
It was a great day with the BERG field in superb condition. There was plenty of lift about and the light breeze helped with the launches. 
