Spectre Project Diary -Part 1

 

Selection

I know that many modellers have a long list of ‘must build’ aircraft and a stash of unstarted kits in their lofts so their ‘what to build next’ challenge is simply one of choice. My situation is different. This year’s flight log (2021) lists more than 40 models (not counting the little indoor ones), all flight-ready and competing for car space on each flying outing. So, when contemplating a new build, the chosen subject either has to bring something completely new to the fleet or represent such an improvement on an existing model that I can invoke my long standing (but less than fully successful) ‘one in, one out’ policy in order to avoid exacerbating my ever more desperate space and storage crisis.

 

So it was, as the Mustang build drew to a conclusion in early summer 2021, that the familiar question once more raised its head - what to do for a winter build? Over the last couple of years I’ve been lucky enough to fly with a group of enthusiastic glider flyers who hold regular aerotowing sessions. Although I do have a model available which has some towing capability, it has its limitations (and isn’t really mine!) so doesn’t come out very often. The fact is that, when it comes to competing with those 40+ other models for a space in the car, a dedicated tug isn’t going to make the cut unless I’m sure it’s going to be needed. So I started musing on what I might build that could justify its place in the fleet in its own right and also be pressed into service as a glider tug when the need arose.

 

I’d already decided that, following the Mustang, it was time for a non-scale project so attention naturally turned to a sport aerobatic model. The competition here is considerable - I already have the Capiche, Astrohog, Kwik-Fli and Aeromaster. As I type that list, I realise now that the Capiche, published in 2000 is the most modern design of the lot. So arguably, in the quest for something different, maybe I should have dragged myself into the 21st century and gone for an up-to-the-minute modern design. However I went the other way, gave in to nostalgia, and found myself looking at the control line models that I had lusted after in my schooldays. It didn’t take long to whittle down the shortlist to a choice between the Mercury Crusader and the Keil Kraft Spectre. Both are classic designs of the 1950s and in the end the Spectre won the day.

 

Conversion

The plans were downloaded from Outerzone - an invaluable resource. Although to me as a schoolboy, the Spectre seemed like a big model, from my current perspective at 41in span, it is rather small, so the first question was how much to scale it up? For towing gliders up to 1/4 scale, 6s LiPo power seems to do the trick. However the battery does take a bit of a pounding, with the motor at full chat throughout the climb out. I run 6s x 4000mah batteries in my Avanti so it would make sense to be able to use the same size packs in the Spectre, ideally with the option to fit two of them in parallel for towing duties.

 

Looking at the existing fleet for size comparisons, my eye was drawn to the Astrohog. At 72in wingspan, this model (admittedly built very light) comes in at an all-up weight of just 5lb 8oz, including a 5s x 3700mah battery. I’ve since stopped using 5s batteries and the Hog still performs very nicely on 4s. So I reckon that a somewhat more sturdily built Spectre of similar size could weigh 6 - 6.5lb and should fly very nicely on a 6s pack. So I had the plan printed out, scaled up by a factor of 1.75. This gives me the option of either a 72in or 73.5in wingspan, depending on which of the two wing panels I build from - unlike the C/L version, I want both wings of my Spectre to be the same size!

 

As well as equalising the two wing panels there were a few other obvious C/L features do dispense with such as the counterweight in the right hand wing and the inbuilt rudder offset, all of these of course being aimed at maintaining tension on the control lines. However, as I studied the plans over the next week or so, I built up a surprisingly long list of other modifications to be made:

 

Flaps/Ailerons

The flaps on a control line model deflect in the opposite direction to the elevators in order to tighten looping manoeuvres. I of course need to re-purpose them as ailerons but, as drawn, they have a very deep chord at the inboard end and taper strongly towards the tip. To be effective as ailerons this really the opposite of what is needed. After some deliberation, I opted to keep the inboard portion as flaps and move the hinge line forward on the outer (aileron) portion. Also, the flaps are simple flat plates hinged onto the trailing edge of the wing. I know this approach is still taken on some fun fly models but I felt uncomfortable about using it on this one and decided on the more conventional approach of building the ailerons and flaps into the wing. This led on to the next question:

 

Wing Section

The Spectre has a fully symmetrical, very thick wing section, no doubt partly to try to prevent excessive speed building up during the downlines of looping manoeuvres. Having just done the first few flights with the Mustang and been very happy with the way it handled, I opted for the devil I knew, traced the section from the Mustang plan and scaled it up to match the root and tip chords of the Spectre, in the hope that the intermediate ribs could be cut out using the sandwich method.

 

Wing Fixing

The Spectre is designed as a one piece model but my enlarged version would need detachable wings. A 72in wing fits nicely in my car so I simply drew in an additional fuselage former in front of the leading edge to take a dowel peg and marked the position of a ply plate to take two nylon bolts through the trailing edge into T nuts. All very straightforward. Apart from a revised trailing edge structure, the addition of cap strips and fitting shear webs to the main spar, the remaining question on the wing structure was dihedral. The C/L version has none at all but I reckon an inch or so under each wing tip might be a good idea.

 

Tail End

The tail surfaces are shown as solid sheet balsa so, in enlarged form, it may be worth using a sheeted built-up structure instead. Whilst the fin is on the small side, the tailplane is generously proportioned. However, the tail moment arm looks a bit short. I don’t want to lose the character of the prototype but longitudinal stability is important for a tug so I decided to lengthen the rear fuselage a bit. Since I need to build in a tow release behind the cockpit, I reckon that’s the best place to perform the ‘stretch’. Other mods needed at the rear include a steerable tailwheel and possible provision for rear mounting the elevator and rudder servos, depending on how the cg works out.

 

Front End

This will be an electric model of course and my selected motor is a lot smaller than the enlarged drawing of an i.c. engine on the plan. A new firewall was drawn in to mount the motor and the details of the cowl structure were left to be considered later.

 

Midships

While making these mods to the fuselage side view, I had a look at the question of battery placement. If I really was going to have the option of fitting a second battery just for towing duties then, whereas battery no.1 can sit happily in the ‘tank bay’, the second battery really ought to be as close as possible to the cg. With a slight adjustment to the cockpit floor it looked like there was just room to fit this optional second battery over the wing. To give easy access to both batteries the hatch logically should extend from just behind the firewall to the back of the cockpit. It was all looking good until I suddenly realised that, with the wing and the hatch removed, the front and rear fuselage sections would be joined only by the upper fuselage sides which, in spite of the slightly thinned wing section, would only be about 1/2in high over the thickest part of the wing!

 

This realisation sent me back to the wing again, re-drawing it in the form of two plug-on panels so that the integrity of the fuselage sides could be preserved. It adds a bit of weight and complexity but, by way of compensation, this arrangement also frees up a bit more space in the fuselage for the optional second battery.

 

There are lots more minor issues and modifications to be tackled as the build progresses but at this point I reckoned I had a sufficiently stable baseline to make a start - except that it was only mid-August and I have a self-imposed rule of not starting a ‘winter’ build until September. This sometimes seems like a singularly pointless rule - after all, nobody else cares if I start cutting would in August! However I have to admit that, but for this rule, there is every chance that I would by now have been well on the way to building a one piece wing before the above structural problem dawned on me, so this time at least, a bit of delayed gratification has paid dividends by hopefully avoiding a lot of frustration later on.

September 2021

So, as September arrived, I cleared a space on the bench and made a start. Usually I build the wings first, on the basis that it’s easier to fettle a wing seat to fit an existing wing than the other way round. However, since the Spectre is going to have plug-on wings, it seemed that this time it might be worth getting the basics of a fuselage built first.

 

As described in the above preamble, The rear fuselage had been stretched by adding in a bay behind the cockpit to accommodate the tow release so it seemed to make sense to mount the servos immediately in front of this bay, beneath the pilot, leaving the area around the cg free for the optional second battery pack. Once the fuselage sides had been cut out from 1/8in balsa and the doublers and longerons fitted, I started with this all-new area, installing the servos and tow release linkage and then went on to sort out the control runs to the rear end.

As you can see from that last picture, sorting out the control runs led naturally onto the mounting of the tailwheel so it seemed a good idea to fit the main gear as well so that the ground stance could be checked. The undercarriage is from 4mm (8 SWG) piano wire. I was lucky enough to be given a K&S Mighty Wire Bender several years ago which makes short work of this kind of job.

The firewall should arguably have come next but I wanted to keep open the option of adjusting the nose length depending on how the cg works out, so the fuselage was put to one side and a start made on the wings.

When cutting out wing ribs using the sandwich method, it simplifies things if the ribs are evenly spaced and all of the same thickness. A set of blanks were therefore cut out of 1/16in balsa, on the basis that they could be doubled up and/or reinforced where necessary later. The ribs were notched to accommodate 1/4in sq. balsa spars. Before gluing the ribs in place though, a second set was cut out, wiring holes bored where appropriate and apertures cut to accommodate the aileron and flap servos.

I’ve used this ‘through the rib’ method of servo mounting for many years now and find it to be simpler and lighter than fitting ply plates between the ribs or mounting the servos on removable hatches.

As luck would have it, I had enough brass box/steel blade joiner material left over from previous projects for this job so ply boxes were made up to accommodate the joiner blades and impart a little dihedral. These boxes fit between the spars and will be bonded to shear webs front and back - I just have to remember to stagger them appropriately so that their mating brass boxes can be fitted across the full width of the fuselage.

All the above preparatory work seemed to take forever but at last I felt that it was time to get the glue out and assemble the basic wing structure.

Rather than use solid sheet balsa for the wing tips as on the original, a template was made (from two layers of 3mm depron) so that they could be laminated from 4 layers of 1/16in balsa.. The laminated tip outlines are pretty strong on their own but it seems a shame to waste those depron templates so I’ll probably use them to provide a bit of support to the wood.

So the first month of this build draws to a close with a reasonable amount to show for my efforts. Maybe next month I’ll get to see how the wings actually mount to the fuselage.

October 2021

As anticipated, a fair bit of time this month has been devoted to sorting out the wing joining arrangements. First, the joiner boxes made last month were built up with additional laminations of 0.4 and 0.8mm ply to match the 6mm wide balsa spars, taking care to offset the steel bars so that their mating brass boxes could sit alongside each other across the fuselage. General alignment was checked before the fuselage got in the way. The boxes were then glued in place with 0.8mm ply shear webs front and back.

Two 3mm ply 'collars' were cut out to fit over the pair of brass boxes and, after a few more dry fits, this assembly was epoxied in place.

The root ribs could now at last be fitted. I have to admit to being caught out here. The wing chord is so great that it extends back onto the tapering section of the fuselage, so the false trailing edge had been trimmed a bit short - hence the 6mm square reinforcing piece you can see in the picture! All in all though, I was pleased with how this all turned out, even though the wings suddenly do seem a lot heavier!

After a lot of eyeing up from all directions, holes were made for the incidence pegs - steel pins, sliding into brass tubes.

Finally captive T nuts were fitted to receive 5mm nylon bolts from the inside of the fuselage. I just need to remember to make sure that these remain accessible when I work out the battery stowage details.

By the end of the month a start had been made on the control surfaces. These will be simple structures, skinned top and bottom with 1.5mm balsa.. The underside of the wings have also been sheeted and cap stripped, and I'm ready to make a start on the hinging and linking of the ailerons and flaps.

I think there’s at most two more weeks work to do on the wings so there’s a real danger that by the end of next month this airframe will be nearing completion. It certainly looks like the model could be finished by Christmas, with at least half of the building season still to go. Which begs the obvious question. . . . . Click here for part 2 of this diary.