Percival Vega Gull

[John Rickett 102]

The fine weather earlier this year meant there was plenty to do away from the workbench, but with the recent cool and rainy weather a bit more progress has been made. The cowl seemed a good thing to get on with as it is something that can be progressed in small stages.

The Vega Gull will have a fibreglass cowl, it’s the only material I know which will stand up to the harsh environment of the engine, fuel and exhaust.

 

 

The plug was made by using the principle shapes of the rear bulkhead, the rear of the nosebowl and the spinner. Its 16” long and 12” wide so needed to have a substantial ‘skeleton’ to hold the shape while the foam blocks were being added. Cheap, low grade plywood from sources such as B&Q is adequate for this job.

 

 

Recticel floor insulation, cut into manageable blocks and with the aluminium faces removed in a bandsaw, sands easily and being left over from a house build, cost nothing this time.

 

 

The blocks were glued using ordinary yellow modelling glue. Once dried, the glue is a bit harder than the foam so care needs to be taken not to end up with ridges.

 

 

Apart from the compound curve of the nosebowl, the bulk of the cowl is single curvature so a long length of wood with some 180 grade glasspaper glued on, made a suitable sanding block. A small Surform for the curves and for initial whittling down, soon got the foam into shape.

 

 

The plug was covered in a single layer of 80gm cloth. After hardening, the low spots were corrected with Isopon filler and sanded smooth.

 

 

Eight coats of Meguiar’s Mirror Glaze release agent (wax) were applied, buffing in between each coat to build up a good layer. The instructions on the tin only recommend three to five coats but experience from doing this quite a few times says the more the merrier; it’s a nice smell of wax polish, just like you remember your granny’s house, so not an unpleasant chore!

 

 

 

The parting board and support frame were made from various ply off-cuts, the parting board covered in parcel tape and the gap sealed with decorator’s caulk. To my mind caulk is better than plasticine or Polyfilla as the excess can be wiped away with a damp cloth and once it’s served its purpose, can be peeled off without leaving a residue.

Strips of 50gm cloth were laid -up first, tucking it tightly into the 90^O flange bend. This was followed up by 4 layers of 100gm cloth.

 

 

A day later, after the initial hardening, the first half of the mould parted cleanly from the plug.

 

 

The excess cloth was trimmed off, three coats of release agent applied around the flange area and then the plug was reinserted into the mould. A simple jig was made to keep the mould tight to the edges of the plug and then the joint was sealed once again with decorator’s caulk.

 

 

Unsurprisingly perhaps, the second side lay-up was exactly the same as the first.

 

 

One day was allowed for the resin to harden and then the other mould half was freed from the plug. My method of separating the parts is to gently work into the gap a ¼” strip of 1/64 ply which has been sanded to a chisel shape. If all is well the separation can be viewed through the fibreglass, with an opaque appearance as the parts separate. On a piece like this, I start from the rear and push the ply strip into the gap as far as it will go, without undue force, then work round the whole rear gap. Do the same for the flange. Once a clear air gap has been formed for half the length of the piece, a bit of thumb pressure under the flange will usually separate the remainder……satisfying to behold!        

 

 

The two halves had their flanges cleaned up, drilled for the clamping bolts and then given eight coats of wax. Once bolted up, the seam area was given a further two coats of wax in case any damage had been caused to the layer with handling.

The cowl was laid up with 1 layer of 80gm cloth initially, followed by 4 layers of 200gm cloth. Different coloured felt tip pens were used when cutting the cloth to help identify the edges of each piece once in position and wetted out. I’ve found that if this is not done, it can be a hit-and-miss affair to get an even covering especially with larger moulds which require quite a few pieces of cloth in the lay up.  

 

 

For good measure some carbon tow was laid up to stiffen the flat sections. Who knows if this has contributed anything, but as I had lots and the weight is negligible, it only extended the process by a few minutes.

 

 

This is the cowl as it came out of the mould after 15 hours curing. All those coats of Meguiar’s paid off as it popped out easily. In its unfinished form here it weighs 527gm, which seems reasonable to me. It now requires the back edge trimming to fit the fuselage bulkhead, holes cutting for the engine and oil cooling intakes and the carb air intake. Once it’s in position, some fibreglass ducting will be grafted onto the inside of the main air intake to align with the fuselage mounted ducting to try and ensure the all-important cylinder cooling takes place.

Although quite a lot of work for a one-off cowl, if anyone ever feels silly enough to want to build a ¼ scale Vega Gull I could supply a cowl for a fortune making price!             

[John Rickett 102]

 

The cowl has been trimmed to fit the bulkhead and the supports made.

 

Providing mounting points for a fibreglass cowl, I always seem to find a bit of a challenge if it’s to be durable. The rear of the cowl stands proud of the fuselage to provide an air escape path; for the model this meant the supports would also have to stand proud. I wanted these to be as unintrusive as possible and the holes not right at the edge of the cowl which could easily split with use. The method adopted was to make the supports from best B&Q steel strip with an additional lug brazed on to provide a thicker threaded area.

 

 

Trying to get the supports in the exactly the right place when the cowl itself was preventing access required a bit of head scratching. The problem was compounded by the cowl having a sharper taper than the fuselage immediately behind. The remedy was to mount the supports on a ¼” ply base which could be worked on, on the bench and later glued into position. The additional ¼” brought forward the mounting holes.

Maintaining the position of the little lugs while brazing (silver solder) was overcome by forming a simple jig and cutting down to size afterwards.

 

 

 

 

The holes were tapped M4.

Four mounts at the side and one underneath seem to have provided rigid support.  

 

 

 

 

The intake duct was laid up using 50gm cloth over a simple sheet balsa form which was covered in parcel tape and given a couple of coats of release agent. A sharp tap with a toffee hammer (on the former not the fibreglass!) released it easily after curing for a day.

 

 

The duct was then sanded to fit in the cowl. It was quite a fiddly job getting it into the correct shape to fit round the engine and line up with the existing ducting. I ended up repeatedly taping it to the cowl and eventually, by trial and error, achieved the final shape as seen here. Once satisfied with the shape, a further layer of 100gm cloth was laid on including some carbon tow for good measure. There’s bound to be some vibration in this area so getting the ducts as stiff as possible and rigidly mounted seemed good practice.    

 

         

 

         

 

The intake duct shown here tacked into position with epoxy resin but not yet reinforced with cloth.

 

 

The duct stands off from the cowl at the centreline to prevent air escaping around the exhaust side of the engine. All Gipsy installations I've seen have this arrangement, so its a small, scale touch.   

 

    

 

A head-on shot where I wanted to show how the intake duct lines up with the ducting attached to the fuselage. Not a very good shot so you'll have to use your imagination!  

 

I thought I was being smart in opting for the reversed inlet/exhaust arrangement, that is where the carb is on the cooling air intake side of the engine, but I now think I’ve built in a problem for myself with regard to carb access. Previous deliberations on this forum have stressed the need to ensure an adequate cooling airflow, particularly over the rear cylinder and head. Heeding the warnings, I’ve tried to make a ‘leak proof’ shroud to direct the airflow fairly tightly along the carb side of the engine; I’m hoping this will be successful, what I now realise though is that access to the intakes for priming and the idle needles for adjustment is not possible with the shrouds in place. The fuel lines also now have to be positioned inside the ducting.

 

 

The narrow layout from the rear exhaust and carb has always been a strong point in Laser’s favour, however with the production of the inline version, the rear layout may cause some challenges to practical operation. Normally an idle needle doesn’t require much adjustment but priming an engine is a different matter. I’m fortunate in having quite a few models though this can lead to considerable gaps in flying sessions and sometimes to the oil congealing in the fuel line to the carb and the needles. It only takes a few months of inactivity for oil congealment to manifest itself. Pinching the fuel line for the last stop of a session doesn’t prevent this. Also, on a scale(ish) model, with an all-enclosing cowl, adjustment may have to be made with the cowl removed and then hope that refitting doesn’t alter a setting.

 

If I had thought about it properly, the less troublesome option would have been the normal inlet/outlet configuration, which would have put the carbs outside the ducting.

There is little significant problem with Lasers having the inlet on the right (engine upright and viewed from the front) in inverted Gipsy Major style cowls, so why did I give myself the headache!

 

Routing a couple of pipes to enable squirting some fuel directly into each intake could be a possibility for priming but it won’t solve an oil congealed situation, for that the fuel lines will have to be disconnected and some neat fuel blown through. Time will tell if trying to get the engine started, particularly after a lay-off, and keeping it cool is going to be an awkward affair.

 

To try and get some data on what is an acceptable cylinder head temperature, I had fitted Powerbox temperatures sensors to a Laser 200v installed in a DB SE5. The engine (until I broke it!) ran fine with no inclination to overheat.  The only convenient mounting points for the sensors were the rocker box retaining screws – not ideal but hopefully would provide comparable data. My Futaba transmitter gave a maximum temperature of 180⁰F (I’d have preferred degrees C but height is measured in feet and the user has to choose metric or imperial measurements throughout, there is no mix-and-match with Futaba). Armed with the knowledge that 180⁰F won’t cause problems, I will fit the sensors in the same position to the 200i and see what’s recorded, that’s still a long way off though.       

[Ron Gray]

You’ve made a lovely job of that ducting John but I wouldn’t worry about the tuning or priming aspect of the in-lines. I had exactly the same doubts with mine as when fitted in the Sportster, Mustang and Hurricane nothing is accessible. I tuned them without the cowls and have found that I then only had to make slight adjustments after the first few flights, mainly to lean them, now I leave them alone. If I was being extremely picky I should  tweak the top end before each flight but I don’t as I don’t want to keep taking the cowls off. For priming I just spin the motor at WOT with the starter for about 10 secs then connect the glow (my plugs are connected in series) and they usually fire first time. The Hurri usually fires up on one then a couple of seconds later the second pot kicks in, I suspect that a further bit of fine tuning would correct this but I haven’t got round to doing that yet!

[John Rickett 102]

Thanks for the reply Ron and the suggestion, its reassuring to know that you haven't had difficulty with priming. I don't usually use the wide open throttle method to draw the fuel through, but will remember to try that when the time comes.

[Jon H]

That cooling setup looks great and is exactly what i had in mind. The carb access thing is a total non issue as you can set the slow runs with the cowl off, and then put it all together. 

 

Priming is not required if you are electric starting, but in the rare event of an oil blockage just open the throttle fully and block the tank vent pipe while filling the tank with fuel. The pressure of the fuel pump will be more than man enough to clear it, although it might make a mess. 

 

Having the carbs on the cold side is a good idea as the alternative is blowing hot air from the exhausts over the heads. This is hopeless for cooling and reversing the heads is the only option. Going forward i am likely to drop the standard head version as the reverse head type is suitable for all models and the standard one only for WWII stuff. 

[John Rickett 102]

Ok Jon, thanks for the reassurance that engine operation isn’t going to be a problem.

 

Just to finish the coverage of engine cooling, these pictures show how the side and rear baffles close the gap at the bulkhead end. The major path for the air is now over the heads and through the gap between the cylinders. That gap would have been closed as well but there isn’t a convenient fixing on the engine, hopefully the forward silencer will block some of the air and help redirect it through the cooling fins.

 

 

 

 

 

[John Rickett 102]

Beryl has paid a visit to the Percival works for a trial fit. Not much space left but that was probably the case in 1936.

 

 

 

Once she has been tidied up – white overalls, hair dyed and a spot of make-up I think she will look splendid.

 

 

Sean of Real Model Pilots has created her from photos found on various sites covering her life and exploits. Sean’s idea is that once satisfied there are no adjustments to make, she can be returned for final clean-up and painting…..that’s not a bad service. As it is, she has dropped in there nicely. The seat can now be covered and perhaps raked back a few degrees to make her look more comfortable.   

 

Plan A is to make her, the seat, the column column and the floor as a removable one-piece unit held in with magnets (as are the tanks), that way seat belts and their fastenings can be added without having to work in a very cramped space. Having everything in the cockpit removable will help considerably in gaining  access to the receiver, batteries and perhaps gyro which will sit in the wing well below the instrument panel. The wing centre section will be permanently fixed to the fuselage so the only access will be through the cockpit.   

[John Rickett 102]

A tumble off a pushbike led to an enforced building board sabbatical. Well on the mend now though so what better physio for a poorly hand could there be than making a slight deviation from the planned order of things and starting on a wing panel.

 

LMA Member Bob Maltby had made an excellent job of cutting the ribs from a design drawn up in Compufoil 3D. Not having learned anything about CAD design, I find Compufoil a simple way of achieving the same thing when it comes to plotting and lofting a wing.

 

 

The picture shows the earliest stage of discerning that the few bits of wood are actually forming a wing. The large gap is where the ribs have to been cut off to allow for the wing fold supports to be added.

 

 

Its crucially important the hinges are positioned in the right place as three things are determined by their position:

·         the dihedral (5⁰) is properly set

·         the wing folds without pinching

·         the relative heights of each panel are correct

The centre section is being sheeted in 1/8” balsa with the outer sections in 3/32” so this adds a bit of a complication to what is already a hit and miss affair. I think that much playing with thin plywood shims will get the correct heights but there can be only one attempt at drilling the spruce/plywood spars. To overcome making a mess of it. the plan is to devise a method of holding the ¼” ply pieces in position without gluing so that if mistakes are made a new piece of ply can be substituted without having to cut out the previous piece.

 

At least progress is being achieved once again after a 4 week delay, I’ll just have to live an extra month to make up for it.        

[Ron Gray]

Sorry to hear about your accident John but good to see you back in harness again.

[John Rickett 102]

Thanks Ron,

All's getting better by the day so in these dark months hopefully some meaningful progress will be made. It's just as well as Beryl is getting impatient to be airborne!  

[Richard Crapp]

Test from Richard

[cymaz]

Stunning work sir. Glad you’re on the mend

[Richard Crapp]

No worries with priming, just open throttle 1/2 and spin with starter. Laser carbs are so well made that If you close the carb no fuel will pass.

 

All my recent cooling ducting is done with firm black foam sponge, off the type Mick Reeves sells for tyres, Cyno'd to the inside of the cowl. if it touches a real hot spot it just melts back a little then stops.

The foam touches the fins and all gaps are sealed. I was just thinking that you could use some blocks to seal some  gaps.

Power is a Laser 180.

 

My Gull was a  airframe started by Auther Searl, perhaps 30-40 years ago so it has scale of 3.7 and a few oddities .

 

My wing now has  a M/R spitfire wing tube, but  still has hinges to fold the wing, and retain it, This means i have to pull the hinge pin, remove the wing and tube then replace the hinge pin. 

It is just a party piece as I did not want a true folder, to difficult to transport.

 

All surfaces are complete with a little finishing.  here is a picture of the flap box, It has a locating dowel into the wing root as dose the wing.

 

 

 

[John Rickett 102]

It’s good to see that another example of Edgar Percival’s talents is underway and that Arthur Searl’s creations live on.  There seems to be so few Percival designs flying - the Mew Gull has been kitted before and available as a plan, but little else I can think of, what have modellers got against them!

 

The Vega Gull was a bit larger than the Gull 4, so at 1:3.7 scale the wingspans will be much the same at 117 - 118”. I’d expect a Laser 180 single and a 200 twin to have similar power outputs, so if the Gull 4 gets to be airborne first, it will be interesting to learn if (power wise) it was a good choice of engine.

 

This will be the third model I’ve built with folding wings, though the first cantilever wing. I’ve found it’s a speedier job to pack a model back into the van, or awning, when the heavens open than the alternatives of tubes or bolts. I accept they are more complicated to build and with all the fittings and screws required, the final weight tends to be greater.

I’m optimistically hoping for a weight of 25lb. I have a much flown Chrislea Ace weighing 25lb and powered by a Laser 180. The power is……adequate, but it is quite draggy. Percival designs were more streamlined so a small increase over 25lb should still be ok.    

 

Thanks for the tip about using black foam as a method of directing cooling air to where its required. I’ll see how the motor performs before adding more baffles but hadn’t considered a material as simple as foam glued into position - something to remember.        

[Richard Crapp]

Thanks John,

For the bits that were missing on Auther's construction, that's nearly everything, I am using Dennis Bryant's 1/4 plan.

It is nice not to strain the brain to mutch. I can trust DB plan, just needs 153% enlargement on the printer.

I think the Gull will fly on the 180 but with no excess of power. I am relying on the clean design. We will see.

 

BTW, i have a story about DH86 but hesitate to clutter your thread.

[Ron Gray]

12 hours ago, John Rickett 102 said:

I’d expect a Laser 180 single and a 200 twin to have similar power outputs,

The 200 has noticeably more power than the 180. I used to have a 180 in my first ESM Hurricane which I then changed to a GA30 (Laser petrol 180), there wasn’t much difference in performance. My latest ESM Hurricane weighs quite a bit more and is powered by the 200 in-line and it gets hauled around the sky much easier than the 180 powered one did. I also have a Great Planes Giant Super Sportster and that has had a Zenoah 38, an NGH 38, Laser GA30 and now a 200 in-line. The 200 provides far more performance than any of the others. Of course the 200 also sounds so much nicer too!

[Richard Crapp]

If the 180 is not enough i would look for a 50cc 4c. The Laser 200 , The Large Enya's and Saito at around 40cc don't seem a big enough increase for the extra £s

 

Unfortunately the rang of 50 cc 4c is small.  Valach have one at £1500. Gaui have one at £1000 but there are questions , Enya have a one at .... a lot!

 

Is Laser's www sight still down, or is it me?

[cymaz]

4 hours ago, Richard Crapp said:

 

 

Is Laser's www sight still down, or is it me?

laserengines.com is working fine with me 🥳

[John Rickett 102]

Richard,

Perhaps its early days to be concerned about engine substitution, although keeping the options open during a build is a wise precaution.

 

The smallest Valach is 60cc, so could possibly be in excess of the requirements for a Gull 4, which I doubt had a sparkling performance fitted with a Gipsy Major 1 in the 1930s. Kolm produce a 50cc single, I’ve got one in a 1/3 scale Leopard Moth which weighs 34lb, the power seems about right. The engine is one of the early ones and for some reason is now on its third set of front bearings in just under 200 flights. Kolm are now up to Version 4 of the engine so I imagine they've found the need to introduce some improvements along the way.

 

Apart from the bearings in my example, its a nice engine. Machined from solid it’s a slim design with the inlets and outlets at the rear in the same manner as a Laser. Its also shorter than a Laser 200i so may need less butchery to install.

 

I’m very interested to see how you get on with completing the Gull 4, you need to start a thread on this forum………

 

Edited December 9, 2023 by John Rickett 102

[Richard Crapp]

I meant Kolm, not Valach. I have one of those at 70cc and it dose not fit,  by some way. the 60 is only slightly less tall.

Unfortunately  the Kolm is eye wateringly expensive!

 

I have a full detailed thread on RCSB    https://www.rcscalebuilder.com/forum/forum_posts.asp?TID=35651&PN=1&TPN=7

 

Is Laser's www sight still down, or is it me?

[John Rickett 102]

Ok, I can see the build thread, thanks.

 

Laser's website has been up and running again for some time now, you could trying accessing it here   

[John Rickett 102]

The right hand wing hinges have now been lined up, drilled, the dihedral set and all permanently screwed into position.

 

     

 

Besides the main spar webbing (1/8" balsa), webbing has also been put in at the roots where I imagine the highest stresses will be.

 

Some 1/8" balsa, 3/8" wide has been added to the first two (liteply) ribs to provide a better gluing area for the sheeting.

 

 

The wing folds as it should but at this point I've no idea whether the other one will be level with it at the tips, so perhaps shouldn't get too carried away with the sheeting and making the screws unreachable. The leading edge will be sheeted to form a D box as well as the first bay of the outer panel. This won't be scale as there doesn't appear to be any sheeting in this area on the real thing, but practical is sometimes better than sticking tightly to scale in my opinion.

 

The distance between the upper and lower hinges is 55mm, I was concerned that there could be some flexing with the wing hung off this arrangement, but am relieved with the rigidity, even without the sheeting.

 

 

 

 

A start has been made on the folding section, this also carries a split flap and will house the servo. 1/2" x 1/4" spruce strips will carry dolls house brass hinges to enable the section to fold right over onto the wing.  

 

[John Rickett 102]

While the section was accessible prior to sheeting, the servo supports were installed. The servo plate is 1/8” birch ply with a couple more strips where the holes are drilled to provide a bit more support with negligible weight penalty. The servo plate slides into position along spruce rails and is then retained by the cover. All this puts to good use the scrap bits that have been collecting in a drawer.

 

 

The servo is a Hitec HS 5245MG, 4 in total for the flaps. Programmable, digital servos are possibly in excess of what’s needed here as with modern transmitters being able to reverse the output, adjust speed, fail-safe and throw, the programmable part of the servo amp is redundant but with metal gears and a torque of 5.5 kg.cm at 6v, this little unit fits in easily and will certainly do the job.

 

 

 

The flap is simply a triangular box made from 1/16” sheet. A length of ½ x ¼” provides initial stiffening and support for the hinges. Recesses for the hinges were chiselled out prior to gluing into position. The ribs were then glued in with a few extra, scrap ¼” sheet ones as a horn support.

Kavan hinges will be adequate for this job and have the benefit of the pins being removable, should separation of the parts be required later on.  

 

  

 

With the trailing edge of the wing getting quite thin, there is not a lot of depth for all the sheeted bits. The ‘roof’ of the flap bay was cut from 1/64” ply with a strip of ¼” square balsa added to help keep straight the otherwise unsupported front edge.

 

 

The folding section taking shape.

 

 

Underside of folding section showing the flap resting in position.

 

 

Flap installed and connected to the servo, ready for sheeting.

[Richard Crapp]

I like your servo arrangement with the horn on the inside where it should be. I will move mine. 

The four flaps you refer to must be the inner ones on the centre section.

The Gull four dose not have those but it dose have an air break between the front undercarriage struts. 

A fascinating detail that i will be unable to resist.

[John Rickett 102]

I deliberated for a while on how to retain the folding sections in their flying position. I wanted a quick method of releasing them (as in when it starts raining and the model has to be packed away smartly) and decided that spring loaded latches would be a secure and positive method. This method was used on the Leopard Moth and has given good service. On that model the pins were home-made but I found these brass and steel ones were still available so thought it would save a bit of time - not to scale of course but shouldn’t be obtrusive once painting is complete.

 

 

 

 

The latches were epoxied into ply supports, including 1/32" ply rib doublers, and set into the folding section as close to the trailing edges as possible – actually quite a distance from the trailing edges as the flaps and ailerons each get in the way.

 

 

I'm (fairly) confident the latches will hold ok, it all seems rigid enough on the bench - it will be disastrous if one lets go in the air!    

 

The ailerons were progressed at the same time as the folding sections; they form an integral unit and the trailing edges have to align.

 

 

After slotting the ribs for additional spars which would become the corners of the ailerons, the hinge recesses were formed with a small sanding block the same width as a hinge and then the aileron was cut free from the wing and faced with 1/16" sheet.

 

  

 

 

There's not much to an aileron, here capping strips have been added.   

 

A decision was taken to install Hitec 5245’s for the ailerons and well as the flaps, they pack quite a punch for such small units. A few years ago Futaba S148’s would have been the standard servo for ailerons, yet only had a torque of 3.0 kg.cm whereas these little units provide 5.5 kg.cm (at 6v), so with nearly twice the torque available there seemed no reason to fit larger and heavier servos.

 

 

When first studying photos of the real thing, it wasn’t obvious why the shroud for the wing fold gap was kinked. In trying to incorporate a folding wing on the model the penny dropped, the wing join has to be staggered to allow for that part of the structure which is behind the pivot point.

 

 

 

I won’t attempt a shroud, just make sure that the gaps have been closed as much as possible.      

 

 

Folding section just about complete, showing access to the servo. A 1/8" balsa cover will be made which will be glass cloth covered to add strength and stiffness.  

 

 

The wing as near complete as possible but without sheeting. Once the other wing is made, there may have to be some adjustments so don't want to cover up anything which there may be a need to access later.

 

A start now needs to be made on the left wing, hopefully it will be straightforward but probably not as satisfying as progressing the first one.