This page details the two year construction and initial flights of Ian’s fabulous 47.5% scale model of the Bristol Bulldog. It was produced from a set of articles Ian wrote for the LMA Journal, issues 98-100. The Journal is automatically sent to all members of the association.
The Bulldog was developed for the Air Ministry’s Interceptor Fighter Specification F17/24 1924. Designed by Captain Frank S Barnwell, the specification called for a Rolls Royce Falcoln X inline engine. The project was shelved until 1926 when Barnwell resumed work on a single seater day and night fighter to Specification F9/26. Bristol Type No. 102A.
In the meantime the interceptor requirement was revived and at the beginning of 1927 Barwell submitted two schemes, one with the Rolls Royce engine, the other with Bristol’s own radial Mercury engine. Both layouts were similar, but early on in the project stage the lower wing was reduced in chord and span which was shown by wind tunnel tests to be advantageous.
Bristols were asked to revise the scheme to comply with Specification F20/27 with the geared Mercury III. At the same time the Air Ministry ordered from companies Gloster, Hawker, Vickers and Westland, prototypes to the same specification. The Bulldog and the Hawker Hawfinch emerged as the winners, the Hawfinch being superior only in spin recovery.
An attempt to improve the spin recovery of the Bulldog by fitting a larger fin and rudder resulted in a tendency to weather lock into wind during cross-wind take-offs landing. A different, more successful solution was to increase the length of the rear fuselage. This aircraft became known as the Bristol Bulldog MkII Type 105A and was sent along with the Hawker Hawfinch to squadrons for evaluation. The Bulldog was declared the winner. A total of 445 Bulldogs were produced including 59 two-seat versions. The most numerous version was the Bulldog Mk IIA fitted with a Jupiter VI IF engine rated at 500HP at 10,000ft attitude and a maximum speed of 178mph.
Bulldogs remained the foremost front-line fighter until 1937 when they were replaced by Gloster Gladiators. Many aircraft were exported to Australia, Finland, Sweden, Latvia, Estonia, Siam and Japan. Bulldogs were the standard equipment of 10 of 13 fighter squadrons forming the Air Defence of Great Britain. One famous Bulldog Pilot was Douglas Bader.
Only two Bulldogs survive, one in Finland and the original Bulldog Mk IIA demonstrator. This was restored in the 1950s by Bristols but was completely wrecked during a low level loop at Farnborough in 1964. What remained of the aircraft was stored until Skysport Aviation were asked to reconstruct the aircraft to static condition. The rebuild took five years.
Ian hadn’t intend to build a model of a Bristol Bulldog, but says it came about in a roundabout way. He had been offered, a few years earlier, a JPX 425cc twin two-stroke petrol engine by a friend. At first he had declined thinking it was too big, but curiosity finally got the better of him. He was immediately impressed, it was superbly built and very robust. An examination inside the barrels and the crankcase indicated that the engine had been run for less than one hour. The price was right and so he bought it. Ian contacted the JPX factory and they supplied very useful information. Before going any further the engine was test run.
It was calculated/estimated/guesstimated that a propeller of 44″ diameter and 12″ pitch would hopefully be a good starting point. Ian designed, laminated and carved a propeller from beech. Unfortunately, he couldn’t get the ignition system to work and on close inspection (after removing the pull cord starter which he hoped he could dispense with) found that the coil/magneto had a hole and hence an electrical short in it. This was heavy anyway so instead the electronic induction system from his spare 342cc Weslake target drone engine was modified and fitted. The carburettor was also replaced as the original looked a bit like an early motorcycle one complete with float chamber which would be unsuitable for aerobatics, Instead a Mikuni pumper type carburettor, again from the spare Weslake engine, was fitted. The ignition system was set up to give about 7° advance at tickover and about 27° at full throttle.
The engine was bolted to a very heavy metal test stand and tied to a tree, just to be safe! With one of the cylinder decompressors pushed-in the engine started
on the fifth flick. It ran exceptionally smooth and the draught produced was awesome, nearly blowing over the conifers in his garden. It wasn’t quiet though and
a larger volume silencer was required. The engine speed was 4100rpm. The JPX factory had said that a speed range of 4000 to 4500rpm was the best speed to aim for. With some running-in and unloading in the air Ian thought it would gain at least 200rpm. The engine was run for only a brief time so as not to annoy the neighbours
Now that the engine had been shown to work came the decision of what to build for it. It would have to be both big and strong and it was felt that a biplane would be the most suitable.
Some years ago Ian had a 33% scale Hanriot WWI biplane and he thought about another larger version, but it did seem like repeating oneself. He is well known for his “early”aircraft, but fancied a change. His thoughts wandered to a 50% model of the 1920s fighter aircraft, the Gloster Gamecock, but it would have been too tall to fit in his van. Ian did really like the aluminium dope and gaudy squadron markings though and spotted the Bristol Bulldog which had the same colouring but was not so tall. From some calculations and measuring he worked out that at 47.3% scale it would fit in the van, just! At 16ft wingspan and 68 sqft of wing area it should provide a good platform for the JPX engine. At this scale, 14” cycle wheels were just the right size.
Ian had the Nexus drawing of the Bulldog and John O’Donnell supplied him with two further drawings plus the profile on the Bulldog. Using the Nexus drawing it was modified into the MK IIA version which has the advantage of a wider undercarriage, steerable tail-wheel and a slightly larger rudder. The drawing was enlarged until 1/10th of an inch on the drawing represented 1” on the model.
When designing a large model Ian likes to spend a lot of time determining the structure and sizes of components. He has a list of various materials, ie pine, ash, beech, balsa, ply, stainless steel, aluminium etc with an average weight per cubic inch. Using this information and the model’s dimensions he tries to decide on a suitable weight. This value is divided into component weights ie, wings, fuselage, tail, undercarriage, engine plus accessories, covering, pilot, details, glue, struts, rigging etc, and then further reduced to wing panel weights and eventually down to weights for the wing spars, leading and trailing edges, wing tips, mounting blocks, reinforcements etc. Then for each component, using an estimation of its dimensions and volume, the approximate weight can be determined. If it seems to heavy he will consider reducing the dimensions of the component. This may seem a tedious exercise, but is one Ian is comfortable working with.
Once the pieces have been cut and machined they are weighed and if necessary their size reduced or replaced with something that is stronger. Despite all these efforts Ian says that he has still never manage to build an aircraft to the target weight, but he cheats a littler by making the target weight say 10% less and then usually ends up with an acceptable weight. With large models the weight is within reason not too critical, the thing that seems more important is to have more than adequate power plus a strong and durable airframe and this he certainly achieves.
Traditional model building methods were used. The full-size construction as detailed on the Nexus drawings was copied, but using wood. The wood used was clear Columbian pine (“clear” means knot free) joined with Evostik Resin W exterior PVA glue.
The longerons were tapered towards the rudder and the front end of the longerons were slit on a bandsaw and glued with a filler strip whilst being clamped to a curved piece of wood to replicate the side view contour. Diagonals were added and lots and lots of corner gussets helped to reinforce the structure/joints. Ash strips were added to the inside edges of the longerons at the forward end plus some of the uprights and cross braces were from ash.
The Bulldog has an “under structure” which supports the lower wing and rear undercarriage legs which are replicated. To help tie the front-end together 3mm and 4mm stainless steel tie-rods were fitted at key points using stainless steel brackets.
The engine bulkhead was produced from 1/4″ aircraft ply with two smaller rectangular sections clamped and glued together. In addition, two vertical beech rails were fitted on the front plus two horizontal beech rails on the rear of the bulkhead. The black circle in the picture opposite is the intake for the engine carburetor.
All of the above, whilst time consuming, especially all the gussets, wasn’t that difficult, but the next stage of determining cutting and fitting the formers to achieve the outside shape of the fuselage was!
The fuselage starts off round, then oval, but with a flat bottom, changing to oval, but with a pointed bottom, then more squarish before the almost rectangular section stern post. A lot of cut and try plus trimming was required to get the right shape and to get the cyparis wood stringers in the right place and to curve smoothly.
The forward part formers, which were to be covered with aluminium panels, were made from 1/4” plywood, with lightening holes, whilst the rear part formers were made using 3mm liteply with the edges reinforced with a 2nd layer of liteply.
To be able to transport the Bulldog the tail and rudder had to be removable. With the centre mounted tail it was decided to make it in two halves, extending the front and rear pine spars into deep ply and ash boxes which were glued across the fuselage. The spars were reinforced with ply and ash strips in addition to each spar being laminated from two pieces of wood. A jig was used to enable the ends of the spars to be trimmed so that they would interlock inside the fuselage boxes.
The wings are based on a two spar construction; the spars again from pine made-up/laminated from two or three pieces of wood. The benefits of laminating include extra strength plus the piece is much less likely to twist or warp. Pine was also used for the leading and trailing edges and each of the curved outlines were produced from 6 strips of 1/16” ply. These were glued together whilst being bound to chipboard formers with electricians PVC tape via windows in the formers.
The ribs were made from hardish 3/16” balsa. They were all reinforced and some of the key ribs were further reinforced with ply and balsa box sections between the spars.
Due to the large chord (36” on the top wing) balsa spars were added between the main spars and also between the rear spar and trailing edge. These stopped the ribs twisting and stiffened up the wing. The full-size top wings feature wing fuel tanks which are accommodated via a deeper section protruding below the wing. Balsa and pine strips were used for the dummy wing fuel tanks.
The ailerons have a Frise action, with the hinge line some distance from the leading edge of the aileron. The up-going aileron “dips” its leading edge down into the airstream and deflects the air through the formed slot. The idea is to help provide a more balanced turn and this was duplicated in the model.
Ash blocks were fitted to house the 3/8” x 1/4” diameter titanium tubes for the 1/4” diameter piano wire wing joiners. Here you can see the outer strut fixings. Stainless steel brackets are fitted to the dural struts. A 5mm allen bolt inserted into a tapped hole can also be seen. A 5x7mm tube goes through the ash blocks in the wing.
The fin and rudder are of a similar construction to the wings, all the surfaces have ply capstrips, 1.2mm for the wings and 0.8mm for the tail.
The Bulldog features a sprung undercarriage, this was replicated using a variety of metals. The top of the front leg fits to a point slightly below the top longeron and on the model this was achieved with stainless steel plates fastened to the ash uprights and ash filler piece. A universal coupling was made from 1 1/4” diameter stainless steel rod. The top part of the front leg was made from 1 1/4” diameter stainless steel tubing which had a top piece welded on, again from stainless steel. At the base of the leg, was fitted a phosphor bronze bush for the lower leg which was 3 1/4” diameter tube. At the top of the leg was another bush/piston which slides up and down inside the top tube leg. For springing two springs were fitted above the bush/piston which gave a semi-progressive action with a travel of nearly 3”; 450 lbs force was required to “bottom” the leg.
The picture opposite shows the front leg attachment. Outer ash diagonal struts with seamless steel brackets can be clearly seen. Note the cross rods stainless steel bracing fuselage structure.
A rebound spring was fitted below the bush/piston plus to provide additional damping. An aluminium bush at the base of the top leg was fitted which had four ‘O’ rings to grip the outside of the lower leg.
The cross axial has an outer made from ¾” diameter titanium tube with an inner axial of 12mm made from EN24T high tensile steel rod in aluminium bushes. The protruding ends of the inner axial accepts the cones/bearings for the 14in diameter small cycle wheels.
For the rear legs, ¾” diameter titanium tubing was used. 5/8” light aluminium tubing was bound to these legs plus the lower part of the front legs and the axial to achieve the correct cross section. The rear legs fit into stainless brackets fitted to the underside of the lower wing centre section. Additional support is provided by struts which are bolted to the fuselage longerons and top surface of the lower wing centre section.
The steerable tailwheel has a ¾” diameter titanium tube leg with a stainless yoke for the tailwheel. Metal bearings support the leg and a compression spring is fitted.
Shown opposite is the stainless steel “box” with rubber stops which fits around the tailwheel. This limits movement to approximately 12 degrees.
The outer struts were made from 1¼” diameter aluminium tubing which was progressively squashed in a vice to produce an oval section. The ends were fitted with stainless plates tapped 5mm for allen bolts. The centre section struts were made from ¾” inconel tubing again slightly flattened.
Inconel is an exceptionally tough and strong material, the downside is that it is very difficult to cut or drill. To cut one piece of tubing you need a new high speed steel hacksaw blade and cutting oil. After cutting one tube in half, you throw the blade away!
A wood jig was made to cut and weld the centre section struts. The mounting plates were from stainless steel, the top ones being tapped for the 5mm fixing allen bolts.
The full-size Bulldog doesn’t have a full cowl, but a formed exhaust collector ring acts as the shaped front cowl to which are fixed aluminium plates, these are cut away to clear the Jupiter cylinders. For the model the challenge was to provide a mount for the 9 dummy cylinders which at 4” diameter are quite large plus trying to hide, as far as possible, the JPX engine which is 16” across in width (not including the spark plugs and plug caps). Additionally, the large silencer needed to be positioned and concealed.
It was decided to make a cowl. For the male former two MDF rings with a ply cross plus a pink foam front were used. The sides were covered with 1/16” ply. From this a two piece female former was made as shown opposite.
The cowl itself was made from glass, carbon fibre and epoxy resin. Two large circular cutouts (5 1/2” diameter) were required for the JPX cylinder heads, which protrude slightly through the cowl. These cutouts were reinforced with carbon and kevlar cloth. Stainless steel straps were added on the inside of the cowl and at the rear of the cutouts. The rear of the cowl was additionally reinforced with kevlar cloth for the cowl fixing screws. On a trip to visit the full-size Bulldog the fixing screws for the cowl and fuselage aluminium panels (with the permission of the RAF Museum staff) were measured. These were about 32mm in diameter and on the model 6mm countersunk posidrive screws with a head diameter of 14mm were reasonably close in scale. These were used for both the cowl fixing screws (18) and for the fuselage panels. Ash blocks glued to the ¼” ply formers accepted the screws.
For the dummy cylinders two attempts were necessary. The first ones were made from thin ply rings with the inner rings from 3mm liteply. These were glued together, but were not aligned very well and they to be sanded to even up the outside rings. Ian had also incorrectly measured the diameter of the cylinders and coupled with the sanding looked too small. So Ian re-measured the cylinders on all three drawings and studied his photos and came up with a cylinder diameter of 4”. A better method of gluing together the rings was used by making a jig which could accommodate the rings and keep them aligned. Opposite can be seen one cylinder clamped in the jig.
A careful study of the photos of the full-size cylinders enabled accurate templates for the various rings with 4 x 3mm holes used for both alignment and for fixing each cylinder to the cowl. A mixture of 1.2 and 1.5mm ply was used for the fins and 3/16” balsa for the inner core rings. These were jig drilled and cut with a fretsaw in batches of 8 or 10 to produce a total of 369 rings (not exactly a five minute job).
For each cylinder the appropriate number and order of rings were slid onto the 3mm stainless steel alignment rods, the sliding top fitted and the stack of rings pushed to the top of the jig. One by one the rings were pushed down to the baseplate and glued together. When all the rings were assembled the top was pulled down and a large clamp used to hold them tightly together whilst the glue set. It was a long and tedious job which Ian found best done whilst half watching the television.
Another jig was used to hold the top of each cylinder at the base, the front and back of the jig was shaped to suit the diameter of the cowl and its taper and the cylinders were cut to suit on a bandsaw.
The cowl was wrapped with some paper and the cylinder positions determined. Then using a long 3mm drill the cowl was drilled for the 3mm fixing rods.
The two cylinders that clashed with the JPX cylinder heads had to be cut away and faced with thin ply. On these two cylinders the rear 3mm holes were enlarged to 4mm for two 4mm fixing rods which engage with tapped holes in the cowl stainless reinforcing straps. Top detail for the cylinders was made from balsa, cyparis and ply. 3mm stainless rod plus springs, nuts and washers for the valve/springs were added. The rocker arms were made from thin aluminium.
For the intake and exhaust manifolds the use of blue foam was considered, but Ian was concerned that they could be vulnerable to damage. Instead laminated balsa was used to give the correct width. With the addition of dummy pushrods made from thin gauge aluminium tubing plus dummy spark plugs fitted with cables and braided outer Ian said the result looked a reasonable facsimile of the original Jupiter engine. To most onlookers it is a piece of master modelling.
The pictures show the massive effort Ian put into creating one cylinder head plus ignition leads, sparkplugs, valve springs and washers. The long rods (left picture) represent the cylinder head long bolts which were silver soldered on inside “washers” at angles to suit taper of cowl. The short rods are the “valves”
The silencer was more difficult than it had to be due to trying to achieve a large silencer volume. This made the silencer a tight fit inside the lower part of the cowl.
The original silencer for the engine had a volume of 3.25 litres and Ian’s was almost 6 litres. It was made from 0.6mm thick stainless steel sheet with a curve and taper to suit the cowl. The silencer has a central separator and each side has two baffles. As with the undercarriage and centre section struts, the silencer was TIG welded together.
The wing rigging was something Ian spent ages considering. The full-size aircraft uses flat bracing wires 14mm wide. Whilst this could have been replicated with say ¼” wide stainless steel strip, it does dramatically increase the time taken to assembly or dismantle the model.
On his other models round wires permanently fitted to the fuselage and outer struts had been used. For transport the outer struts and rigging are temporarily tied to the outer sides of the centre section struts. Flat bracing wires prevent this easy route as flat wires need to be unbolted at both ends for transport. In the end, round wires of 2mm diameter 7×7 strand with a breaking strain of approximately 900 lbs were used. These are tensioned with home made stainless steel tumbuckles.
Before covering the model, the structure was given a coat of Clearcote lacquer to seal and protect the wood. The covering material was Ceconite lightweight polyester fabric as used on microlight aircraft and gliders. It comes off the roll at 61in wide. Ceconite supply a special adhesive which is used to glue the fabric to the outlines of each structure. Washers plus thin nylon cord lacing was glued to simulate the lacing and eyelets. As with traditional model coverings, an iron is used to shrink the ceconite. The material is very tolerant to iron temperature. The shrinkage is phenomenal and you can hear the structure creaking as the Ceconite tightens.
Three coats of thinned dope were then applied to seal the surface. For the aluminium dope Ian was lucky since Skysport supplied some of the actual dope left over from the full-size Bulldog restoration. Somewhere Ian had read that to get a better depth and light reflection it is better to give the fabric a coat of white dope prior to the aluminium dope, so that is what he did.
The markings and roundels were painted using enamel paint which was matched to the colour details supplied by Skysport. Several days were necessary for just the masking and the end result is a set of bright gaudy squadron markings.
Twelve panels were required and these were made from 0.55mm thick aluminium. They were formed with a variety of items including his wife’s rolling pin! As with the cowl mounting blocks, the blocks for the fixing screws for the metal panels were countersunk and with the screw tightened up it produced a reasonable representation of the full-size fasteners.
Tor the ailerons, thick wall brass bushes were set into the ash rails which supported the ailerons. Similar ash strips built into the ailerons also used the same brass bushes. Thin-wall aluminium tubes epoxied between the hinge bushes provided a guide for the 2.5mm diameter piano wire hinge pin which is inserted via a drilled hole in the wing tip. This provides slop free hinging and if required, the fabric patch on the wing tip can be detached and the rod removed completely releasing the aileron.
The tail used 3 brass cabinet makers hinges; 1½” long for each control surface. These were screwed and glued to each 1/2″ depth spar before gluing on the top half of the spar.
Once the structures were built, the hinge pins were drifted out and piano wire rods used similar to the ailerons. For the removable rudder a tapped bush was used which is attached to the top of the rudder. The piano wire hinge pin was silver soldered into a hole bored inside a 5mm allen bolt which engages with the tapped bush.
The elevators and rudder have horn balances plus glassed-in brass rods to the inside of the leading edges of all the control surfaces to provide some mass balancing. This helps to prevent flutter and also reduces the load on the servos.
Opposite is shown the elevator and aerodynamic balance. Note the ¼” brass rod mass balance fitted to the insider of the leading edge.
The rudder and steerable tail-wheel are operated by closed loop via 3mm thick epoxy glass cranks mounted near the servos and connected with short ¼” titanium pushrods and metal links.
Initially it was thought that two receivers should be fitted in the base of the pilot’s seat, but looking at the full-size why not use the scale radio compartment behind the pilot? This allowed easier access plus the receivers would be directly below the fuselage aerial post. The receivers are held in a balsa/ply tray and a tongue engages with a formed slot with two 3mm allen bolts to hold it in place.
Here can be seen the servotray that houses the 4 litre plastic bottle fuel tank and rudder/steerable tailwheel servos. The tank retaining "hoops" also support the scale foot duckboards and scale rudder bar. The tray is fitted with 4 x 3mm allen bolts for fixing, making it easy to remove.For the ailerons, rudder and steerable tail-wheel, four Multiplex jumbo servos are used. These have approximately a 24kg on 5 battery cells. Metal gears and programmable electronics make them very reliable. For the elevators the slightly smaller Futaba 5801 servo was used. The short pushrods for the ailerons and elevators were from ¼” diameter titanium tubes with metal ball links fitted to 2-piece stainless steel horns.
The flight batteries are two five cell packs of Sanyo 5000mA capacity and the ignition battery is a six cell Sanyo 3000mA pack. Two 8A diodes provide battery separation and two double pole 8A switches with the two poles "ganged" together to give 16 A switch capacity/extra redundancy.
S.M. Services supplied two 5 cell multi-LED indicators plus a 6 cell unit for the ignition. The diameter of wire used was 0.5mm throughout with heavy duty connectors. To provide more redundancy, each receiver has three battery inputs spread along the input rail/connectors. This arrangement also reduces the current flow into the receiver input tracks.
On an aircraft like the Bulldog possibly two of the features that people look at first are the dummy engine and the pilot. Pete Richardson of “Pete’s Pilots” made a superb pilot to the correct scale (47.5%) with a very authentic, fur-lined flying suit plus scale Sutton harness.
Ken Forty, an LMA member and friend supplied the instrument panels and printed dials to the correct scale. When mounted on a ply dash with ply bezals, sprayed black and with small 10BA brass screws and acyclic faces it looked right for not too much effort.
The two machine guns were produced from 7/8” diameter aluminium tubing with a ¼” diameter tube for the centre barrel, drilled and painted matt black. Ian says they look OK.
The oil cooler on the full-size comprises seven cooling plates, tubing, plus tube stays. Initially, seven pieces of 0.7mm aluminium were cut, but when assembled they looked awful. Instead, a recessed jig was used into which was formed 0.3mm thick aluminium into a dished/recessed shape using a small spoon. Two of these with a filler piece of 1/8” balsa were epoxied together and looked much better and also much stiffer than the original effort. Stainless steel 4mm rods plus ply spacers were used to make the oil cooler with thin wall inconel tube stays.
The generator on the lower wing is based on two ply circular formers and a wrapped ply “ring”. Blue foam forms the front and rear. It is fitted with 6mm sealed ball bearings and a stainless steel shaft for the beech impellor.
The dummy wing tanks have 0.3mm aluminium plates on the lower side of the top wing epoxied on and also screwed on with 300 plus miniature screws! Several more smaller plates replicate access panels on the wings.
The dummy exhaust pipes were made from 1 1/4” diameter ABS, 41 pipes cemented together, plus stainless steel brackets. Balsa and ply was used for the scale lower tank fittings with a thin wall aluminium tube inside streamlined to replicate the fuel lines from the tanks to the fuselage.
Two aerial posts are fitted on the top wings plus another aerial post above the radio bay. The rudder has another aerial fixing point which is a metal tag on the top of the rudder hinge post. Non-operational dummy lights are fitted to the top wings and rudder. The lights and aerials are removed for transport.
The wheel discs were made using fibre glass, Two female moulds were turned from blue foam and sealed with a thin layer of glass and epoxy resin.
With wire wheels it’s not practical to fasten the wheel covers to the rims due to the flex of the wheels during take off and landing. So plywood hubs are used to which the wheel covers are screwed. This allows the rim to move/slide past the wheel covers without causing any damage.
The model finally weighed 170 lbs some 20 lbs (13%) heavier than my (low) target weight of 150 lbs. This may seem heavy, but with 68 sq ft of wing area the wing loading is a very modest 2½ lbs a sq ft.
The model was started in the autumn 2000 and finished by Christmas 2002. There were times when Ian had a break from it (as you do!) when interest was lost. A big project starts full of enthusiasm and progress appears to be easy, but halfway through you realise that you have a huge amount of work to do to complete the model. It’s at this point that you think wouldn’t it be easier to build something else. Maybe totally illogical, not practical but somehow appealing.
The most time consuming part was the final finishing of the structure, covering, painting, panels, making the dummy engine and the other scale details. This took a concerted effort over several months.
For models weighing over 20kgs the LMA operate a scheme on behalf of the Civil Aviation Authority. For the Bulldog, John Townsend who was the LMA Safety Officer for many years, acted as the inspector. Ian discussed with John the construction and various details and John inspected the basic airframe at one of the LMA meetings. John made a second final inspection when the model was completed and in due course Ian received an exemption certification to allow flight tests.
To gain an exemption certificate, which allows you to fly an over 20Kg model at public events, it’s necessary to complete a flight test programme. A flight log sheet is required to be completed. These flights have to be witnessed by a person authorised by the Large Model Association. Ken Wooten, the owner/operator of the airfield near Withernsea, East Yorkshire, was asked to witness the flights.
The first attempt at a flight was a non-attempt! The engine started and ran well and was taxied around to get the feel of the model. When a take-off was attempted the engine faltered. A considerable amount of time was spent fiddling with the carburetor, but the engine would not run smoothly. A decision was made to abort and come back another day.
Back home Ian rang the JPX factory (in France) where they very kindly answered his questions regarding the correct spark plugs to use and timing details. Fitted with these plugs the engine ran much better and so back to the field for another go…….
The engine was gradually opened-up and the swing to the left was easily held with a touch of rudder. Immediately, Ian says he had the impression that the model would be a good flyer. A variety of turns and figure of eights at various throttle settings were made and the model felt right. When carrying out turns Ian automatically mixed aileron and rudder, so he thought he should try a turn with ailerons only and rudder only. He came to the conclusion to carry on as started!
The model felt a touch nose heavy, but not unduly. A stall turn was attempted which was very safe and predictable. The rudder was very effective.
Normally, Ian said he wouldn’t do any aerobatics for the first few flights, but felling confident he tried a loop. The model tracked around without any tendency to “screw out”. At a safe height a half roll followed by half loop looked good. Next was a complete roll. Ian claims it does not roll as well as expected. The first half of the roll is OK, but the remaining part not so good. Ian rapidly came to the conclusion, that like most biplanes with only two ailerons, the Bulldog did a better barrel roll. (Maybe the scale 5° dihedral didn’t help). Before a roll you need a fair bit of height, the Bulldog looses height when inverted during a roll and using some ‘down elevator’ tends to slow it down too much. Probably the model replicates the full-size rolling characteristics. Douglas Bader’s accident, when he lost both his legs, was the result of a low roll in a Bulldog hitting the ground when inverted.
A spin was tried next and this turned out to be a very sedate affair. Wingovers looked good, really showing off the red and white chequers on the top wing on the turn. For the landing it needed a fair bit of throttle and it seemed as though it needed to be flown all the way down which happily resulted in a very good crosswind landing.
For subsequent flights the main 5A batteries were moved back giving a centre of gravity of 29% average chord, which felt better. A three pointer landing has been tried, but resulted in about six bounces/landings; not to be repeated!
Eight flights and nearly two hours of flying were carried out for the flight test programme which was accepted and in due course an exemption certificate from the Civil Aviation Authority was received.
On the 10th flight there was a near disaster. On a cross-wind take-off the engine when almost airborne, lost power and at first Ian hoped it would pick-up, but it didn’t and with a real lack of power Ian couldn’t hold it straight against the cross-wind and it ended up touching down heavily with quite a bit of sideways drift. The port wheel was wrecked along with the fibreglass wheelcovers plus the axle was bent, but fortunately no other major damage.
The spark plugs were black and it was concluded that the second-hand carburettor was the problem. It had been difficult to set-up and so it was decided to order a new one. A priming tube was also fitted to the intake bend between the reed block and carburettor. This had an allen screw to seal the end. Previously, the engine was primed by squirting fuel in through the cylinder decompressors. However, neat fuel on the plugs is not the best situation. By injecting the fuel before the reed block, by the time it gets into the cylinder, it’s a vapour. With this modification, plus the new carburetor, the engine ran much better, cleaner and with a touch more power, plus the starting was much easier.
A new 12mm diameter axle was made and phosphor bronze bushes fitted to the wheel hubs. With better hardening and tempering it was perhaps 33% stronger. The dummy ABS pipe exhausts had also come apart and so the joints were reinforced with internal metal straps.
The engine problems were not fixed though. Sometimes it would run very well delivering ample power, especially for the first few flights, then at other times it would run rough with a lower power output. New spark plug caps offered no improvement so a new Tillotson carburettor was obtained and fitted, again with no improvement.
Ian now suspected the ignition system. Regrettably, the engine seized-up at the worst possible moment at an event at Longhorsley and with limited height the model suffered a lot of damage. One lower wing panel, the lower centre section and one elevator were wrecked and new ones had to be built along with a new cowl and silencer. The undercarriage and dummy engine suffered and needed extensive repairs.
Tony Collins very kindly offered to re-align the crank on the engine. He reported that the cylinders/pistons had seized/gummed-up and at first it seemed as though the fully synthetic oil may have gummed-up the engine. He stripped the engine down and rebuilt it, the barrels and pistons were cleaned and new piston rings and cylinder gaskets were obtained and fitted.
Regrettably, once refitted in the model the compression, especially the port cylinder, was low along with the power output. It was hoped that with more running it would improve. The oil was changed to Castrol Super TT mineral oil, but still there were intermittent faults which at first didn’t seem to fit any pattern. A complete three piece identical ignition system was used one piece being added at a time to try and identify which part may have been faulty, but with no success.
One evening Ian was reading through the comments in his Flight/Log Book when suddenly he thought heat build-up, not on the engine, which had been repeatedly checked with a temperature probe, but at the ignition advance/retard unit which is mounted around the rear end of the crankshaft inside the “bell” shaped engine mount. A call to the ignition designer/producer followed. The components were rated at 80°C. This explained the problems experienced. The optical device fitted on the advance/retard unit had to be shielded from light and oil fumes which could prevent it working properly. A lithoplate cover had been fitted which probably increased the heat buildup.
Three or four attempts were made with modified versions of the advance/retard unit, regrettably without success. Each attempt plus the previous changes were a major effort. The removal of the propeller and spinner, followed by the forward fuselage panels, two of the dummy cylinders, then the JPX cylinder heads was necessary each time. The cowl with its 18 scale fixing screws plus the inner filter cover had also to be removed to gain access to the engine fixing nuts followed by the removal of the engine.
If you add the time spent setting up the unit together with ground and flight testing you can imagine how much time and effort was expended. Possibly, with more time, the unit could have been made to work, but after spending every spare moment of the previous eleven weeks repairing the model and trying to get it working correctly, Ian gave up. Especially, as he had missed displaying the models at several events including La Ferte-Alais and Cosford.
Instead, Ian borrowed a Becker ignition unit from a King twin cylinder engine. This has the advance via the electronic control unit. The magnet was fitted on the propeller driver with the sensor above it on a simple adjustable bracket. Additionally, the metal spark plug caps had to be extended with thin metal sleeves to fit the longer spark plugs. The Becker system worked perfectly first time and no problems whatsoever have been experienced since.
Regrettably, the poor compression did not improve and although the Bulldog was flown at the Rufforth and Much Marcle shows, the lack of power really limited the aerobatic performance and general flying of the model.
After Much Marcle, the engine was stripped down and the cylinders and pistons carefully measured. They were outside the recommended limits in the JPX engine manual, especially the port side. Enquiries were made to the JPX factory and Ian was almost put off by the very high cost of the parts, but if he wanted to fly the Bulldog again there wasn’t much option. New cylinders, pistons, rings, gudgeon pins, small end bearings and gaskets were obtained and carefully fitted. Following over three hours of running-in the engine delivered ample power with total reliability.
Since then the model has been flown successfully many times, albeit with a touch more oil content, and a slightly richer setting. There is now ample power to loop from straight and level flight and the climb for stall turns and wingovers is now impressive. Given more running, tweaking the carburetor and a touch less oil it can get only better. Tony Collins has suggested that a larger bore carburetor may also release more power.
Ian now enjoys flying the Bulldog. Ian says it is involved and you have to think about what you are doing, but is not too difficult. It has no real bad habits or characteristics and it does look attractive in the air especially when you have a bright sky.
Was it worth all the effort? The answer must be yes, but maybe for the next model Ian says he will probably pick something much simpler and easier!
With the increased size/complexity and possibly more than one model it can sometimes be difficult to remember everything that is required with regard to packing/transporting and flying our models. Ian has been using a check list for some time which he finds invaluable.
Below is the current Bristol Bulldog checklist.