Mail the Knewt?


News Photos Planes The Bwlch People Jokes Knewt? E-mail


Estrella Review

by Bill and Steve Haley as seen in QFI 52


After a very frustrating season in 2000 Team Haley (Bill and Steve) decided that it would be wise to assess the situation with regard to the F3B models being used worldwide in order to determine if we were still using the best model available for F3B.

We have been using Stuart Blanchard Cobras for the last four years with a fair amount of success. Remember that it was an ex Team Haley Cobra that won the last World Championships in South Africa, albeit in the hands of Daryl Perkins! In fact there were more Cobras in SA than any other model, Steve's own model, this time with him flying it, ended up seventh.

We sent e-mails to all of our friends on the continent and also asked the question on the Internet. To complete the picture we also looked at what is being used by F3F and F3J competitors in this country.

In F3F most of the top models tend to use RG15 which we don't think is so suited to modern F3B with its 10 minute Duration task and the high number of Distance laps being flown. RG15 loses out on the launch to sections with higher lift, not really a consideration in F3F! We therefore eliminated super models such as the Pike and the Tragi 702. These models are also a little short on span for the tasks in hand. We feel that we probably need a span of around 3.1 to 3.2 m in order to get the aspect ratio up for efficient soaring.

We visited a friend who was doing well flying an Eraser from the Czech Republic in F3J. Steve had a little play with it and it launched well but we felt that the model, quite understandably being primarily an F3J model, didn't have the all out speed that F3B demands.

The other thing we were looking for was something with a section other than the standard MH32 as used on the Cobra and indeed the Eraser. You may remember my report on the F3B World Championships in South Africa where my favourite model was the Warp 1 by Dieter Perlick that uses a 7.5% thick MH32 wing section. This model launched and did every task very well, indeed it came very close to winning the competition.

The result of all our investigations was that everyone gave us the same answer as to what was the only production model to have to be competitive... the Estrella, manufactured by Fischer in Germany.

The Estrella won the F3B Euro Tour in 2000 flown by Reinhard Liese and captured many of the other top spots. Decision made!


What's In The Box?

With this type of model you don't get a fancy box, so no comments on the quality of the picture on the outside, anyway it's the quality of what's inside that counts.

So what do you get for your money? A glass fuselage, a pair of moulded wings, a one piece moulded V tail, a carbon wing joiner, one moulded nose weight, a screw-in tow hook and 8 pieces of 21 mm square solid brass ballast weighing in at 1050 g.

Quality overall was good with only the odd bubble in the gelcoat, mainly on the fuselage. The fuselage definitely qualifies as thin. Sub C batteries will simply not fit, AA size cells will be required. In terms of construction it seems to be a little complex in that five mouldings are required together with another two for the flimsy, non-structural, outer nosecone. The fuselage is moulded in glass and Kevlar with carbon for reinforcement.

The excellent Website of Graham Woods ( has some useful information on the Estrella, and we are indebted to Graham for allowing us to use his observations and photographs. One particular point of interest is that Graham feels that part of the fuselage could be of sandwich construction, I am not sure about this but it certainly is very light and very strong. Carbon tows are used along the fuselage sparking fears about radio reception but we have never had this problem using a Graupner MC24.

The fuselage has a number of features that make it unusual when compared with normal trends. The radio installation is very tight with the two ruddervator servos just behind the battery area and the receiver behind the servos in the enclosed area in front of the wing. Access to the receiver is from the ballast opening on the bottom of fuselage. We decided against this in favour of altering the installation to make the receiver more accessible, more of which later.

As you've probably realised from the shape of the ballast, the "tube" is square. This is a first for us. As the fuselage is elliptical in shape the square tube leaves a gap between the tube and fuselage bottom that is filled with hardwood into which the tow hook is screwed. The ballast is held in place with a brass rod across the fuselage.

The wings are plug on, which we haven't used for a number of years, and are located by two torque pins and a substantial wing joiner (20 x 15 mm) going through the fuselage. The one-piece tailplane sits on a tail seating that is two moulds added to the rear end of the fuselage. The tailplane is held in place by one bolt from underneath and its surfaces are actuated by two pre-fitted snakes using carbon inners and ball joints.

The heart of any soarer is the wing and this wing is different to anything I have handled to date in both construction and shape. The section by Martin Weberschock is the MW3, which is also new to us. Wing shape is high aspect ratio with tips that curve both back and up. Whilst this is more common on F3J models it can be seen on the odd F3B model. It uses polyhedral and quite long ailerons and short flaps. The flaps and ailerons are bottom hinged with silicone and are very free. Personally I would have preferred slightly longer flaps to help with launch and crow braking. The aileron and flap pushrods exit the top surface through small shrouds to the control horns. The wing appears to be produced in normal moulds but the internal construction is very different in that normal spars have given way to a multi-spar system. This is how Graham describes it.

"Each wing uses some 33 m of carbon fibre braiding and does not appear to have a conventional spar. My guess is that seven or more lengths of this braid are placed around an equivalent number of inflatable tubes. These are wetted out and inflated inside the joined moulds taking the cloth with them (to the skins) to get the effect that you see in the photographs. Alternatively the tubes could of course be held at normal air pressure and the whole wing mould put in a vacuum bag to create the same effect."

There is no end rib as such, the gaps at the root are filled with epoxy and microballoons. The skins are of sandwich construction although it is difficult to see whether the sandwich material is balsa, foam or something entirely different. There is a 40 mm band of carbon main spar, top and bottom of the joiner boxes. One of the consequences of this type of construction is that the wing of the Estrella is exceptionally rigid in torsion. Probably the best we have come across.


Putting It Together

Except for the fuselage the rest of the model went together the same as most moulded models so I will not spend too much time explaining every detail.

The fuselage layout is designed for the receiver to go behind the servos that use pre-fitted ply servo mounts. We were not happy about this as getting to the receiver to change crystals etc. would be very difficult, so we decided to see if we could change it. We removed the pre-fitted nose weight and made up a four cell battery pack with one cell fitted right up to the front of the nose and the rest of the cells positioned to fit the available space around the first cell. We were then left with enough space to fit the receiver and then the servos. Whilst this layout makes crystal changing a lot simpler it does make the addition of nose weight very difficult. A mould was made and a lead weight cast to go around the battery as far forward as possible. This has worked very well even though we had to file away the leading corners of the receiver in order to get the nose cone to fit! Fitting the remainder of the electrics was then straightforward. We used Multiplex plugs for the wing servo connections.

As mentioned earlier the supplied tow hook was a simple hook that screwed into the bottom of the fuselage. We like to set our hook to suit ourselves and so set out to design an adjustable hook that could be used on this model. The layout of the ballast tube meant that the height of the hook was very restricted, so we had no option but to encroach into the ballast tube a little. A mould was produced and new lead ballast was cast that could slide over the tow hook, as shown in the photos. This system has worked extremely well; the new lead ballast weighs approximately the same as the original brass, which means that we have gained an adjustable towhook at no expense of ballastability. Big sigh of relief!

The wing on this model is very straightforward to put together. We normally just fix servos with 5 minute epoxy, which makes them easy to remove if required. This time we used the new Volz servos that have plywood servo mounts available for easy removal. The servos and mounts have so far worked very well. Indeed in two months of flying the servos have not had to be re-centred once. The aileron and flap horns are supplied (alloy) but you have to drill the mounting holes for these. No problem, just work carefully and check the results making sure all the horn geometry is appropriate and equal.

Servo covers are supplied and have a small bubble on them as with the thin wing the servo arms tend to stick out of the bottom of the wing; a neat touch. The only small problem that we have found with the wing was that when the model was first flown it needed a very small amount of aileron to keep it straight. On checking we found that the wings were not 100% in line with each other. This can happen with plug on wings and was easily corrected by moving the torque pins a fraction. This can happen to any type of plug on wing panels, so if your model needs a little aileron to keep the model straight check that the panels line up.



The Estrella is a very quick model to assemble with only two wing panels and a one piece tailplane to fit. The plug on wings need to be taped in place at the root as is normal with this type of fixture. The tailplane is bolted on with a single bolt through the bottom of the fuselage.

As is normal we gave it a hand throw to centralise the elevator etc. and it glided for approximately 150 yards, very promising! It was then put on the winch and the first launch initiated. There was little or no wind and the model was very soggy on the line. We had used a set up very similar to that used for the Cobra and it was very obvious that this model needed a very different approach.

At this time we had no set up details from the manufacturer and therefore set about re-programming the model ourselves. Slowly we got the model to launch reasonably. As any experienced flyer will tell you, a good launch depends on many different factors, one of the most essential of these being the CG so a long time was spent on trying to get this somewhere near correct. At the end of this first session we were getting reasonable launches but still not good enough, so we obtained some set up details from a good German friend of ours (thanks Armin). On checking the settings we had arrived at, we found that ours were not too far away.

It took us a further three or four sessions to get the launch to a state that we were reasonably happy with so that we could start getting the model to perform the various tasks. We still find that it can be prone to tip stall if everything on the launch is not spot on. This, I assume, is down to the narrow tips.

Speed is generally the next task to perfect and again it is all down to experience and the type of turns that you want to achieve. Some flyers like very tight knife-edge turns at base B, while others prefer a reversal or rounded turns. Steve very rarely does reversals at base B and tries to keep his turns reasonably tight, but not quite as tight as a lot of the European flyers. This allows the model to fly around the turn without scrubbing off too much speed. At the end of the day this to a great degree determines the amount of elevator and snap flap movement that is required. We think that we've got this sorted out pretty well as we had a speed get-together before the F3B season started and with Steve at the controls the Estrella won this easily, averaging 16.5 seconds for the 6 rounds that we flew. The weather on the day was very cold and quite windy, far from ideal. We still feel that we can get even more from it in this task.

Turning to distance we again found that with our Cobra set-up the Estrella's distance performance was poor. It was probably 3 to 4 laps down. So here we go again. Reset all functions. Remember that not only has the model got to go forward with minimum loss of height but it also needs to be able to turn correctly at each end of the course and so time needs to be spent on getting this right. At the end of the session the model was performing satisfactorily.

Duration is not a task that we practise very much because if you are a thermal soarer by nature it comes naturally, and providing that the CG and flap settings are correct any model will go up in lift. The secret of thermal soaring is the ability to pick out lift and then work it. Suffice to say that the first time this model was asked to do 10 minutes was at the first F3B comp of the season at Peterborough in late April. No problem was found in this department. In fact the model was almost out of sight on both duration flights, having said that there was good lift around for most slots. Steve ended up second at Peterborough, not winning was nothing to do with the models performance it was more down to poor flying and missing a turn in speed and losing some laps in distance down to bad tactics (or bad chatting!).

So what do we think of the Estrella? Well in our view it is a superb model. It's probably not a model for the beginners to F3B because of the time and experience needed to set it up correctly. It is not an easy model to fly in that you have to be on top of it at all times to get the best out of it but if you are a fully committed competition flyer who is after the ultimate performance to win, then this is a winner. Make no mistake this is a model that the more you put into it the more you will get out of it.

Besides F3B I am sure that it will make a fine F3F model. Dave Woods who flies an Estrella in F3F tells me that it is the fastest model in a straight line that he has flown. I believe him. Despite the rave report we still feel there is even more to come from this model, more practise is required, but doesn't that apply to any model!

Estrella Specifications:

Wing span: 3150 mm
Wing area: 60.95 dm2
Aspect ratio: 16.4
Wing section: MW3 (2.3 camber, 8.4% thick)
Weight: 2000 g to 3800 g
Wing loading: 33 g/dm2 to 62 g/dm2
Tailplane section: HD808
Price 2032 dm plus VAT (around 675)

Supplied by:
Flugmodelltechnik Fischer
Backhausgasse 3
D-64832 Babenhausen
Tel 06073 3677
Fax 06073 3030


Back to top?