Aquabird Project Diary

 

The Aquabird was designed by Laddie Mikulasko and was first published in 1992 in RCM magazine. The original prototype is pictured above.

 

We have no retrieval arrangements at our water flying site so flipovers can be particularly embarrassing. Having experienced this predicament with both my Lazy Bee and Sealand, I was looking to extend my range of waterplanes with something that would be stable on the water. When I came across the Aquabird, I immediately felt I need look no further.

 

This 60in span model was designed for .40 - .46 two stroke power, had a balsa veneered, foam-cored wing and weighed in at 7lb. The Traplet plan which I purchased shows a built-up option for the wing but in all other respects is, as far as I know, faithful to the prototype.

 

So dire was the English 'summer' of 2012 that I started planning my 2012/13 Winter build early. Indeed, by July, not only had I chosen the subject and bought the plan, but I had all the wood in stock. The urge to relieve the tedium of the bad weather by starting the build early was becoming irresistible when my other half suddenly called to say that she had located four dining chairs which matched our existing 42-year old set of four.

 

Apparently, in spite of the family having left the nest some years ago, it transpires that we now have a need for eight chairs. So for an outlay of 32 I found myself with a project to restore all eight chairs, which kept the workshop full until the 2012 Olympics got underway - and then in mid August, at last, Summer arrived.

 

After a good late Summer's flying, the equinoctial gales arrived in mid September, so the bench was cleared, the plans rolled out and the Aquabird build got formally underway.

 

Although I do want to do a built-up wing, the version drawn on the plan does look as if it was drawn without the benefit of a prototype build. The combination of the forward sweep and gull wing dihedral does result in some awkward joining angles between the wing panels, but I can't help thinking that there must be a way to tackle this without resorting to the multiplicity of large, heavy 1/8in ply dihedral braces shown on the plan.

 

Also, because of the deep chord and symmetrical section, the wing ribs are huge. As a result, this wing seemed to be a candidate for trying the Robin Fowler wing building method. Although still unsure about exactly how this would work out, with these vague ideas in mind, I started cutting ribs for the inboard panels.

September 2012
The plan shows 1/4in square spars top and bottom, with shear webs from 1/16in balsa. Nothing wrong with that but, because I wanted to use the two part rib approach, I went for a full depth spar out of 1/8in balsa with 1/4 x 3/32in basswood cap strips. This way, the rib halves could slot into notches in the spar, making the assembly somewhat self-jigging. Having cut out five rib pairs (20 rib halves), I felt the need for a break and also wanted to check that my rib/spar concept would actually go together.

When I offered up the first rib half, it looked alarmingly skinny. However, the ribs are pretty closely spaced so, once the rear sub spar is in place supporting the aft portions of the ribs, I think it will be strong enough.

Whichever way you come at this wing though, it seems that every joint and notch has to be cut at an angle so it's not something to be rushed.

Eventually though, I had a complete set of ribs and so it was time for a more comprehensive dry fit, and the inevitable fettling process.

The rear spars were cut from 1/8in balsa. Since the aft portion of the rib profile is pretty flat, there didn't seem to be any need for the rear spar to support the covering so I made it of just sufficient depth to notch onto the ribs. One more dry fit and a quick look at how the aileron servos might be mounted and there we were at the end of September.

October 2012 (more on October)
The plan shows large, heavy 1/8in ply dihedral braces joining the wing panels together at the centre. However because of the forward sweep on the wings there are substantial gaps between the braces and the spar which are marked to be filled with 1/4in balsa which seems to me to defeat the objective of using such beefy braces in the first place. After some debate on the RCG build thread, I made up a dihedral brace from two laminations of 0.4mm (1/64in) ply which were clamped to a bent piece of aluminium sheet while the glue dried.

As this dry fit of the dihedral brace shows, it has to be fitted before the R2 ribs can be glued in place. This in turn means that the top and bottom spar caps also need to be left off for the moment. I was keen to do as much as possible on the two panels before joining them together, simply for ease of handling.

Note the wing is upside down in this picture. You wouldn't believe how often I have had top/bottom and left/right confusion with this wing!

False leading and trailing edges were fitted to the panels the centre joints were fettled before one last dry fit then there was nothing for it but to join them together. As you can see from the pictures, I tried to build a bit of reinforcement into the l.e. and t.e. joints as well as the mainspar. After all, this model effectively lands on its wingtips so this central joint has to take landing shock loads rather than just aerodynamic stresses.

Although the plan shows the wings fully sheeted, I opted to sheet only the leading edge on the basis that this, combined with the full depth mainspar, should give plenty of torsional rigidity.

The ailerons are shown on the plan as cut from 1/4in balsa but the false trailing edge is over 1/2in deep so this looks to be a fundamental error. So, more 1/16in balsa was glued to the trailing edge to enable the ailerons to be partly built in situ.

I decided to take advantage of the depth of the false trailing edge to make a semi concealed aileron linkage on the basis that a rear facing pushrod exit would be less vulnerable to water splashes than a slot in the underside of the wing.

I must admit that I was quite pleased with this arrangement - until the designer responded on my build thread to point out that, in spite of the significant anhedral on the wings, the whole of the trailing edge would be in contact with the water when taxiing so, rear facing or not, my pushrod exits would be letting in water!

So, the end of the month saw me still deliberating how to overcome this setback. Watch this space!

November 2012 (more on November)

I sought advice on the water ingress problem and most people advocated moving the aileron servos forwards. I must confess to being somewhat reluctant to do that, partly because I hate re-work and also no doubt because it feels like admitting defeat. I ordered some 'transom bellows' which, as you can see from the picture, fit rather neatly and which I feel should solve the problem.

Posting this solution on my RCGroups build thread elicited a less than enthusiastic response though, so I didn't feel particularly re-assured. For the moment though, the wing building has been halted so that a start can be made on the fuselage. I will however leave the underside centre section of the wing unsheeted for a while so that I still have the option of fitting a servo box should I eventually see the error of my ways.

The fuselage sides are cut from 1/8in balsa and I used 1/4 x 1/8in basswood for the upper longerons. As you can see, the sides become quite skinny over the top of that thick wing and so if the battery hatch has to extend back into this area then it could all get a bit delicate. So I took the precaution of fitting 1/32in ply doublers to the sides. The formers are made up from 1/8in balsa, but I have left the front formers out for now so that the firewall position can be determined once I can get a feel for how the model is going to balance.

With the wing offered up to the fuselage, the inboard ends of the ailerons could be trimmed and the wing seat adjusted. Once the top of the wing centre section had been sheeted in the front dowels and rear bolt fixing were sorted out and this weird bird was beginning to take shape.

Attention now turned to the tail surfaces. The fin is built as left and right halves so that the twin elevator snakes can be routed through it as the two halves are joined together. I will be deferring actually glueing them up for quite a while yet.

With the fin pinned in position, the snakes were routed through and the servo positioning finalised.

As usual, there is more detail on this stage of the build on the November diary page.

December 2012

The tailplane is shown on the plan as solid 1/4in sheet balsa. I instead opted for a built-up structure, using a 1/4in framework which will be sanded down a bit at the front and rear so that, when sheeted top and bottom with 1/16in balsa, it will still take solid 1/4in balsa elevators.

The two dihedral braces are from 1/16in ply, and the blocks on the trailing edge are for hinges and for the aluminium tube bracing struts that run from the tailplane halves down to the fin trailing edge.

Once the top skins were in place, the two halves were joined and the scale of the tailplane becomes clear, dwarfing the 6in rule in the picture. With the tailplane trial fitted to the top of the fin, the fin outline was traced onto the underside of the tailplane and the undersides sheeted to within 1/16in of this mark, to allow for the still-to-be-fitted fin skins.

With the undersides sheeted, the leading edges, tip strips and elevators could be fitted and the final profiling done before another trial fit with the fin. This time a 12in rule was needed to give scale to this impressive tail structure.

The skinning of the fin could be put off no longer so one side was fitted, then the snake routing finalised before fitting the second side.

Once the fin was skinned, the fin/tailplane joint seemed to locate well so, with the tailplane pinned in position, I set about the aluminium struts. I had made the mistake of drilling the holes before sheeting the undersides (to make sure they went through the blocks!) so there was little or no leeway in getting the length of the struts right. Setting the crimped ends at the right angles took a while too.

However, with the struts mounted and a couple of pins through the tailplane, the assembly feels fairly solid so I am contemplating making the tailplane removable.

Fin leading edge was next and, once the fin-to-rudder fit was finalised, there seemed nothing for it but to glue the fin in permanently. Well, perhaps one more trial assembly and alignment check will put that job off into the new year. . . !

January 2013 (more on January)

The prototype Aquabird was finished in a very 'feathery' colour scheme which I didn't fancy trying to replicate. Instead, I wanted my model to be more aeroplane-like, so the first job of the new year after finally plucking up the courage to glue the fin in place was to fit a cockpit of some sort. I had saved an old detergent bottle whose width was similar to that of the aft end of the fuselage so a shape was drawn on it and cut out.

The first fit was very encouraging, so I pressed on and sheeted in the fuselage top with 1/16in balsa, then fabricated a fairing at the base of the fin.

More pictures are on the January page.

The model was then assembled for another balance check and also to check that the fuselage hadn't twisted during the sheeting process. Then there was nothing for it but to start on the wing tips.

Again, there is a bit more on the wing tip panels on the January page, but the end of another month soon came round so I couldn't resist putting it all together again.

February 2013 (more on February)
The first job of the month was the sponsons. These are fairly straightforward in construction but of course they fit to a wing which has anhedral, forward sweep and taper. Also, because my Aquabird looks like being quite a bit lighter than the original, I reduced the width of the sponsons by an inch or so. All of this meant that the drawings of the sponson sides on the plan could only be taken as a rough guide, so paper templates were cut out before taking scalpel to balsa.

Once satisfied with the fit, the sides were glued in place, with a couple of bulkheads to keep them square, then the undersides were sheeted in. A 1/64in sole plate was applied to the front section for 'gravel protection'.

With the sponsons complete, the model was assembled for another balance check to finalise the position of the firewall.

I was keen to avoid having a detachable cowl so, after much deliberation, decided to try mounting the motor through the rear of the firewall. As you can see, there isn't a lot of firewall left to hold the captive nuts, but I think it will be strong enough.

The hatch has a compound curve so was planked from 3/32 sheet. Although I thought the framework was pretty substantial, it was a struggle to stop it distorting at times. Wire retaining clips were incorporated and a securtity tether will be fitted - losing a hatch in the middle of a lake is always embarrassing!

The last construction job was the top of the nose, then it was time to assemble the complete model, sit back and contemplate colour schemes. As usual, there is more detail on this month's diary page.

March 2013
The colour scheme was determined largely by the covering film I had in stock. There wasn't a lot of silver so the underside of the wings are just red and blue. Once the wings and fuselage were more or less covered, the model was put together to decide on the colours for the tailplane.

When the tail had been covered it was epoxied and screwed into place and I set about finalising the elevator linkage. Here I hit a snag. Although the snakes lined up well with the horns and the elevators moved okay, there was a significant dead zone around neutral due to slop in the curved snakes running up inside the fin.

After contemplating scrapping the snake linkage altogether and fitting a couple of micro servos in the top of the fin, I eventually opted for the slightly less drastic option of fitting larger control horns.

This sequence of pictures shows the swap.

The larger horns have double the effective length so the impact of the slop in the snakes is much reduced. They are a bit ugly though and have backing plates on the top surface of the elevator which is a shame.

Anyway, once all the controls were working okay, the receiver and motor were installed and the model assembled once more to finalise the battery location before making the battery tray.

So, with a re-cycled and heavily hacked about battery tray fitted, cg and motor checks performed, it is now just a question of sitting back and waiting for this interminable Winter to relent.

April 2013
well, the 'bird has flown! Some permutation of nervous pilot, bumpy air, rearward cg and/or dual centring elevator led to a rather jittery first flight. However, the model got away well, had ample power and, after just under four minutes flying including one ceremonial loop, alighted safely back on the water. There was no evidence of any water ingress and apart from a bit of peeling trim tape, nothing fell off.

So, a successful maiden. I'll file another flight report when I have got to know the model better and have the controls harmonised to my liking.