Scratch Built SE5a

[extra slim]

Not normally my main area of interest, but I have really enjoyed and appreciated the work and craftmanship which has gone into this. Hats off and thanks very much for sharing. Thoroughly enjoyable.

[David Holland 2]

Loving this thread, keep logging in to see if there are any updates.

 

David

Edited April 24, 2023 by David Holland 2

[Ace]

@Greyhead46 Followed from the beginning and just in awe of the detail achieved from usually simple everyday bits creatively combined - thanks 👏  

[Basil]

32 minutes ago, Ace said:

@Greyhead46 Followed from the beginning and just in awe of the detail achieved from usually simple everyday bits creatively combined - thanks 👏  

Seconded

Bas.

[Greyhead46]

Hello All

 

Thanks for the replies they’re much appreciated, I’m glad you’re finding the thread informative and enjoyable. As it’s retrospective I can post updates most days, don’t have to wait to do the actual building, that was done long ago and yes I don’t like spending money unnecessarily, if I can make my own that’s better than buying readymade!

 

Grahame

 

 

I’ve said before that I like to get the positions for the radio gear sorted out as early as possible in the build so as not to get any nasty surprises at a later stage, after all the whole purpose of the exercise is to produce a radio controlled model! Unfortunately, although I confirmed the principle for the aileron control very early on in the build, the practicalities of actually fitting it into the model could not be worked out then because I had no idea how much spring tension would be needed or how much “slack” the system would have.

 

This is the design that I tested to prove the principle.



I fitted the spring temporarily using the inserts from an electrical connector strip and adjusted the tension as necessary. Because of the tension I had to reduce the length of the cable joining the top ailerons slightly then all worked nicely.



I was surprised to find that as with the elevators there was no noticeable tightening or loosening of the cables and the tension required was the same, ¼" plus a bit for safety; I’d expected to have to use about double as the one cable controls all 4 ailerons, runs round 4 pulleys and there is a lot more of it.

Fitting the spring the way I have done has given me an idea; I’m going to turn the design “inside out”, that is I’m going to attach the spring outside the servo arms not between them as originally intended. Doing it this way means I’ll be able to move the servos a lot closer together, which will mean less bending of the snake inners, which in turn will make for a smoother and more precise movement of the ailerons.

Having got to this stage I couldn’t resist the temptation to fit the wheels and take a couple of photos.






Looks a bit “naked” without all the wires, Lewis gun and a pilot!

I was discussing the aileron linkage and was asked “why use the spring if there’s no slack?”. Well the first thing to say is that I said “there was no noticeable tightening or loosening of the cables” not quite the same thing, but anyway the cable needs to have some tension to keep it in the pulleys and make them rotate, without the spring this tension would have to be supplied by the servos working against each other. Stalled servos drain power from the battery very quickly and the result can be disastrous. I had considered using a true closed loop system with the cable itself providing the tension but I feel the set up would be quite critical; the spring gives that bit of leeway.

[Greyhead46]

 I had thought that the wings would be quite “wobbly” when held in the frame but in fact the assembly holds itself in position quite well. The photo also gives a good view of the underside weathering and the inspection “window”.



The interplane struts are held in position with bent pins with only the bent section epoxied to the strut; if needs be it should be fairly simple to remove them. The struts only really need the pins to hold them in place until the model is assembled, after that the struts are always in compression so they really hold themselves in position on the pegs.



The front flying wires are permanently attached to the fuselage bracket; the 2mm stud goes through a hole in the fuselage side just above the front undercarriage leg and is held tight by a nyloc nut.



On the full size the rear flying wires are attached to the fuselage inside the lower wing stubs. This is impractical on a model of this size so these wires end inside the wing stubs joined by a length of brass tube.



The SE5a has a very thin wing section but the model has sufficient strength, thanks to carbon fibre spars and LE, to not actually need flying wires, except for show, so it is perfectly feasible to just leave the wires loose. I don’t really like that idea though as I feel they might “flap about” uncontrollably once the engine is running so I’ll devise a way of applying at least some tension; either flexible “tails” attached to the wires themselves or hooks on flexible wires inside the wing stubs, I’ve not decided as yet.

 

For the last few days I’ve been experimenting trying to make the tail plane support wires from 24swg piano wire. I made various ends using brass and aluminium tube but none of them looked right; they were all too “chunky”. In the end I decided to try bending the wire to the exact length required, doing this meant there would only be 2 wires in the “ferrule”, which could also be a lot shorter as it isn’t functional. This was not my preferred method as I thought it would be a problem getting the bends in exactly the right place but as it turned out it wasn’t too difficult, of the 8 wires only 1 had to be scrapped and re made. The ends are made from a sort length of heat shrink tubing, which is very thin walled, painted silver; the overall effect is certainly a lot better than the fishing trace.

[Greyhead46]

When assembling the model I have to connect the 4 aileron cable quick links, this is a bit awkward as there is only a relatively small gap between the centre section and the wing root rib. To make things easier I’ve made a small “tool” to hold the links open, as a bonus it also stops the links disappearing into the wing tubes!



It is simply inserted into the link and turned through 45º. It’s easier to see how it works using a link that is not on the model.

 

 

The aileron linkage is now fully installed in the model and it all works fine! I still need to tidy things up a bit; I need to make new servo connectors because as I was fitting everything into the model I realised that once they’re glued into the snakes the servos would be very difficult to remove and knowing my luck one of them would pack up, I’ll also make better spring connectors,

Because I decided to modify the servo linkage by “turning it inside out” so to speak I was able to mount servos horizontally instead of vertically as I’d originally intended, which makes things a lot easier. The mounting plate is from 0.5mm galvanised steel with some 1/8th ply to improve the grip of the mounting screws.



I modified the mounting lugs to enable both servos to be held by the same grommets, this wasn’t strictly necessary but the assembly being that bit narrower does make it easier to fit into the fuselage. 2mm mounting studs are glued into some 1/8th spruce, which also helps to keep things straight.

 

The whole assembly fits between the front spars and is held in place by the 2 studs.

This photo is at neutral.



As the servos move, differential ensures they don’t work against each other whilst the spring keeps the cables under tension; only the “pull” servo is doing any work.

Full right.



Full left.



As each aileron servo has to operate all 4 ailerons with their associated cables and pulleys I needed “high torque” servos. Because I wasn’t convinced that the system would actually work satisfactorily when installed in the model and I didn’t want to waste money, I decided to buy a pair of “cheap” Supertec servos.

These are very powerful, giving 7.4 kg/cm torque at 4.8volts and have no trouble at all moving the ailerons, but they do sound a bit “agricultural”. I now think maybe I should have had the “courage of my convictions” and gone for either Futaba or JR. Can I now justify the extra expense?

Edited April 25, 2023 by Greyhead46

[Greyhead46]

To answer the rhetorical question posed at the end of my last post; “yes I can!” In fact having given it some consideration I changed the question I asked myself to “having spent the best part of 2 years building this model can I afford not to buy better servos?”

I’ve bought a pair of Futaba S3305, they have slightly less torque at 7.1 Kg/cm but are quicker, 0.25 sec/60º as opposed to 0.33sec for the SuperTec; I’d thought that the 0.11 sec that the SuperTec were slower than “standard” wouldn’t really be noticeable but it certainly was and importantly the Futaba don’t have that “agricultural” sound. I’m a lot happier with them notwithstanding the extra cost!

I’ve also replaced the 148s with ball raced S3001Bs.

In order to rig and connect the aileron cables to the servos the model has to be held upside down; before I fitted the Lewis gun I just used beanbags for support and protection but obviously this was not a permanent solution so I've made a stand, which is simply a length of 75mm x 50mm timber and “Y” shapes soldered up from15mm copper pipe with foam insulation for protection.

The stand will also come in handy for routine maintenance etc. as it holds the model securely whichever way up it is!

 

I’ve had the wings on and off the fuselage several times sorting out the aileron linkage etc. with no problems, until today that is! The secret is to keep a small amount of tension in the cables whilst assembling the model, which I did by pulling on the linkage as I fed the wing pegs into their tubes, but I must have got a bit complacent today as one of the cables came off it’s pulley. After a considerable length of time and much cursing I managed to get it back onto it’s pulley using thin wire fed through the exit slot for the cable: I now have much more admiration for the “ship in a bottle” modellers! I’ll have to devise some sort of mechanical device to keep the tension; I wouldn’t want this to happen again as I’m getting ready to go to the flying field.

The model is now all but finished; the “teething troubles” with the rigging has been sorted out by using a mechanical tensioner to keep the cables in the pulleys at all times. The problem is that there is not much room between the wing and the centre section whilst rigging, this taxed the brain a bit until I literally used some “lateral thinking” and although I’ve only used it a couple of times I don’t envisage any more problems.

This is obviously just the prototype; when I make the rest of the set they’ll use springs not elastic bands! It doesn’t need much tension, in fact the elastic band is really too strong.

When in the storage position the ailerons are at neutral and the quick links just protrude from the wing tubes.



When connecting an aileron cable it is pulled out from the wing tube, which of course pulls the other connector into its wing tube, the metal “staple” is to stop the thread cutting into the rib.

 

As I said in my last post, the model is all but finished and there’s a long time to go until the flying season, so the question is what to do in the meantime? I thought “a Cooper bomb rack would while away the hours!”

I was a bit concerned that my SE5a was “piling on the pounds” with all the detailing work. A quick calculation for relative sizes gave me a target weight of 4.2Kg, so I stood on the scales holding the model with my pockets full of batteries, servos and a Laser 70 and got a weight of 3.8Kg. The model actually feels quite heavy but there’s a lot of wing area so I’m sure the model can take the extra weight, but whether or not I manage to sort out the complexities of designing and building the bomb rack remains to be seen!

 

[Greyhead46]

I’ve made a start on the bomb rack, or to be accurate on the bombs; there’s no point making a bomb rack if I can’t make the bombs! Here’s a photo to show what I’m aiming for. (sorry, I just couldn’t stop myself!!)



I decided that the only practical method is to turn the bombs, but therein lays the first problem; I’ve not got a lathe. If you work for Health and Safety I suggest you don’t read the next bit. I "converted" my vertical drill stand into a make shift lathe, very Heath Robinson but it works.



I started with a length of curtain pole and I’ve now got the basis for one bomb, but I don’t think I would be able to produce all 4 bombs and end up with them all being the same; they’re quite a complex shape.



I’m thinking of using this one as a master for a mould and then making a set of bombs from casting resin. I’ve never used casting resin before so it’ll be a steep learning curve!

One question that immediately springs to mind is “is it tuff enough to withstand being dropped from the model?” They will be released quite low, I’d think about 50ft and at relatively slow speed over grass. For safety I intend to mould the front black bit from rubber and the arming vanes will be removed. There was no chance of getting any casting resin until after the Christmas holidays so I thought I might as well give it go at turning all 4 bombs.

I’d turned the first one using a card template and found it quite a difficult and slow job; then I had a “eureka” moment, it does happen! I measured the diameters at various points along the length of the bomb, turned the wood as a series of cylinders and then simply tapered from one diameter to the next.



This produced the other 3 bombs very easily and quickly; they’re not factory perfect but certainly near enough for me and a lot cheaper than buying casting resin etc.

[Greyhead46]

Making the bombs as a series of cylinders was simple and quick, which is just as well because I’ve had the make a completely new set of bombs. It wasn’t until I came to fit the rear fins that I realised that the rear section of the bombs was about ¼" too short. It just goes to once again emphasise the old adage “measure twice cut once”!

So what went wrong?

I first tried to turn the bomb using a single template of the wood to be removed, this meant that before the template for the widest part of the bomb got close to the surface the deepest cuts had to be made and this made the work piece quite weak and impossible to work with safely. I decided to “split” the bomb into 3 sections; I made templates for the front and middle sections, the rear section is just a straight taper so didn’t need a template. To ensure that the curves “flowed” nicely I decided to make the templates overlap by a ¼" but somewhere I got the measurements wrong! Having turned the front and middle sections I simply measured the overall length of the bomb and tapered the rear section accordingly not realising that the front of the bomb was in fact ¼" too long. When I decided to turn the other 3 bombs I used the master, not the photo, to set the callipers and measure the distances so duplicating the error. When I discovered the error I remounted a bomb in the lathe and tried to correct it but that didn’t work at all so 4 new bombs it had to be. It’s not all doom and gloom though; at least I now have some “spare” bombs to practice on.

I found that cutting 2 slots as a cross to fit the fins made the rear section very fragile and I was afraid that I might break the “legs”, so I cut one slot first, glued in opposing fins made from a single piece of 1/64th ply, then cut the other slot and fitted the other 2 fins once again as a single piece.



The bombs have been filled and sanded but I’m not too concerned about a perfect finish, after all if everything goes to plan they will be dropped from a great height and are certain to sustain damage or even get lost in the grass!

I turned a plug for the nose of the bombs from which I hope to make a mould to produce them in rubber.

I intended to make a mould for the nose section from Plaster of Paris but a trip to our local chemists drew a blank; they even tried to order some from their supplier with no luck. During a recent decorating episode, which involved a fair amount of repair work to the walls in preparation for emulsion painting, we used Polyfiller, it seemed very similar to Plaster of Paris so thought I’d try it.

The plug was prepared by dripping molten candle wax all over it, heating it up with a hot air gun re-melts the wax and most of it drips off leaving a very thin layer. I mounted it on a plastic base with a wall of Plasticine and then gave the lot a spray with WD40.



The Polyfiller was mixed quite thin and carefully poured in to fill the mould. It’s not as quick setting as Plaster of Paris; it took about 2 hours to harden and be ready to remove the Plasticine and the base, then another couple more to dry before carefully easing out the plug.



It has actually turned out surprisingly well; there are a few places where the Polyfiller has broken away in between the lugs but to be honest I expected a lot worse. It remains to be seen whether or not I can use the mould to produce acceptable rubber nose sections; I still have my doubts.

I have not idea what sort of rubber I’ve used; I bought it years ago from the model shop, I think its intended use was for making moulds for casting Plaster of Paris figurines. It’s very soft with a “natural” colour; you melt it over a low heat then pour it either into a mould or over a plug dependent upon whether you’re making a male or female item, and then just let it cool and set.

Being thermo-plastic it was easy to “touch up” the mouldings using a precision screwdriver heated over a candle flame. They’ve been glued to a 1/64th ply base with cyano and given a coat of black paint.



The finished parts are not perfect but I’m not going to too much trouble as they are “throw away” parts and as such will be attached to the bombs with double-sided tape for easy replacement, I’m sure they will need replacing after just a few “bombing runs”. I might make a set of better finished parts for static display.

 

[Greyhead46]

The SE5a was developed well after the time of dropping hand grenades or bombs by hand from the cockpit or shooting at the enemy with a pistol. The load it could carry was limited so 4 bombs of 20lbs each was the maximum, either as 2 sets of 2 bombs, one under each wing or a rack of 4 under the fuselage just aft of the undercarriage legs. The bomb release, operated by the pilot, is the out of focus black lever mounted low down on the cockpit side.

 

Pulling it back released a bomb, pushing it forwards reset the ratchet system, pulling it back again released the next bomb and so on until all 4 bombs had been dropped.

 

The bomb release mechanism on the SE5a is very complex, no surprise there then

 

 

Although of course it is a proven design there is no way I could duplicate it in miniature so I’ve had to design from scratch. To release a bomb is simple, just a pin connected to a servo. But to release bombs sequentially is not so simple. I needed something that would be both reliable and practical at this scale; the full size Cooper bomb rack design would be neither!

 

To control the release mechanism I only have either the retract channel or, if I invest in a new 8 channel receiver, a proportional channel. A proportional channel would be the easiest to work with but I don’t like the idea, if I turned the knob a bit too far a bomb might almost but not completely release then with vibration it could release itself at an unplanned time, which could be dangerous, or at the very least loose me a bomb.

Having decided that the “bang / bang” retract channel it will be, I need some way of “stepping” the release mechanism. I got my thinking cap on and came up with an idea.

 

I took my inspiration from a pendulum clock mechanism, these have been reliably sequencing one-second releases for over 350 years; the reliability comes from the fact that the “catching” plate is in position before the cog is released, but of course this produces a rotary movement whereas I required a linear movement.

 

 

The “Eureka” moment was when I realised that staggered indexing pins could give the linear movement required.

 

 I made a mock-up to see if it would work in practice.

The main release bar has 5 indexing pins spaced at 5mm intervals and will eventually have 4 bomb release pins each 5mm longer than the previous.

 

The release cam has a slot cut out from the centre; before an indexing pin is released from one end plate by passing through the slot the other end plate is in a position to "catch" the next indexing pin.



This is a series of photos showing the sequence as the servo moves the release cam arm backwards and forwards from one extreme to the other; the release bar steps by 5mm with each operation.



At the moment I’m using an elastic band, the spring tension will need to be worked out when it is actually releasing bombs, I hope!

 

Here’s a schematic showing one full cycle of the servo, 2 bombs released. Don’t ask me why I did it "stepping" right to left, must have been feeling a bit Chinese!

 

Edited April 27, 2023 by Greyhead46

[Greyhead46]

Before I could start on the bomb rack itself I needed some flat tinplate so that family size tin of Quality Street had to be finished off. As a special favour, to help the better half keep to her new year’s resolution, I volunteered to take on the task myself!

The main body is made from the tinplate with a brass tube for the rear support. The release arm is made from 2mm threaded rod, the threaded section acts as the bearing in a length of tapped snake inner cyanoed in a brass tube. It will extend up into the fuselage through the large opening just behind the undercarriage legs to connect to the servo.



The release mechanism works a lot better now that a bearing supports the release arm.

 

The bombs are now finished, not without a few problems along the way!



I painted the rubber nose sections with Warbirds paint; I thought it would be OK because it is water based although it does have a couple of drops of a catalyst added to make it fuel proof, but it turned sticky after a few hours so obvious there was some reaction. I stripped the paint off but the rubber was still sticky, I’d got nothing to loose so I tried the old “trick” of dusting them with talcum powder, this time I painted them with “genuine” acrylic and there was no problem, I dusted them again, just to be sure, before the second coat.

I’d intended to attach the nose sections with double-sided tape for easy replacement but this also didn’t work very well so they are now epoxied in place.

The arming vanes are from tinplate soldered to a pin so these are easily removable.

They will be removed before dropping the bombs because not only could they cause damage they would also be damaged themselves.

 

Even this simplified bomb rack is quite complex and it still has the safety stops to be added.

 

[Greyhead46]

I mounted the bomb rack on some scrap balsa whilst I adjusted the bomb cradles. This adjustment turned out to be quite critical; a balance between having the bombs loose and able to swing about or too tight to allow the release bar to move under the spring tension.



In fact it was so critical that I have had to number the bombs to ensure they are fitted in the same positions, the slight variations in the diameters of the bombs where the cradles support them was enough to cause problems.

The bombs are held in position by different length pins that pass through loops on the bombs, which themselves pass through slots in the base of the bomb rack. This photo was taken before any adjustments were made; the bombs are now equally spaced!

 

The bombs are loaded in the reverse of the order that they are released. With all the bombs released the release bar projects 20mm from the end of the bomb rack, I don’t think this will be too obvious “in flight” but if it is I’ll shorten the release bar and extend it with a length of piano wire. The first bomb is held in position and the release bar pushed in to engage the release pin through the bomb’s loop, the slot in the release cam is such that the previous indexing pin can pass back through it when the release bar is pushed in. The cam is then moved (at the moment by hand but once on the model by flicking the retract switch on the transmitter) to the other position. The release bar can then be allowed to spring back against the cam whilst the next bomb is put into position and the process repeated until all 4 bombs are in position.

 

The bomb rack is bolted to 4 small brackets fitted between the undercarriage legs; the release arm has a ball-joint soldered to it just inside the fuselage bottom and the control rod passes forwards between the undercarriage cross members and the fuselage bottom.



A mini servo to operate the bomb release is located in the engine compartment attached to the firewall by an “L” shaped servo mount.



Commercial “L” servo mounts seem to me to be very expensive for what they are so I make my own from “L” shaped plastic extrusion from B&Q; I bought a 6ft length for a couple of quid years ago for a DIY job and I’ve still got half of it left.

[Greyhead46]

Weight is always a consideration with scale models; because it isn’t a property with a “tipping point” it’s tempting to say, “it can handle another few grams.” The trouble is all these few grams add up and can end up as a few kilograms!

Of course given enough power anything will fly but as the weight increases for a given wing the flying speed has to increase to generate the extra lift required and it can get to the point where the model can no longer be flown in a scale manner, which really defeats the object of the exercise.

One advantage of these early biplanes is that because they didn’t have powerful engines available at the time, they were designed lightly loaded with lots of wing area, unfortunately the aerofoil sections are not as efficient as more modern ones but in practice weight isn’t usually too much of a problem, drag is what really saps the power.

Personally I don’t over concern myself with the weight, I just build as light as I can, allowing for sufficient strength, certain details have to be added to be scale and there’s nothing that can be done about it. The bomb rack on the other hand isn’t essential so if needs be it can be easily removed with 4 bolts, but I hope it’s OK as I think it will be a real “talking point” and increase the appeal of the model.
 

Something I’ve never done before is to make a “static” prop, after all the object is to end up with a flying model not a static display, but as I’ve got time on my hands I thought I might as well give it a go. An added incentive is the fact that I’ll be using a 13" or 14" prop for flying whereas the scale diameter is 18¼" so the flying prop does look very small.

As I said I’ve never done this before so it’ll be a case of trail and error and hope for the best; to make things easier I decided to go for a 2 bladed prop. I thought of making it laminated but it would be quite costly to buy the wood and as I’m not certain of the outcome I decided against it. From a front view photo I produced a template, glued it to a piece of knot-free pine and cut it out with a jigsaw.



I was very surprised that it only took about an hours work with a plane and a rasp to produce a reasonable looking prop, and a large pile of wood shaving!

 

A working prop is in a completely different league from the one I’m making; a static prop just has to look right, it doesn’t have to produce thrust or withstand any of the forces that implies. I certainly wouldn’t want to trust my prop to fly the SE5a!

The prop blades have been thinned down considerably, initially using the Dremmel but finishing off by hand. One coat of stain has been applied, several more to go.



The props had various things done to them and as I have no evidence as to what, if anything, was done to the prop of E5808 I can choose the variant I like best.

The choices I know about are:

1. Metal reinforced leading edges.

2. Metal reinforced tips.

3. Both of the above.

4. Outer half of blades fabric covered.

5. Blades bound with several turns of cord about mid point.

6. Plain stained wood with no additions.

At the moment my preferences are 1 or 5, I’ll probably try litho plate for reinforced leading edges first, if that doesn’t work I’ll go for the binding.

 

I’ve got the easy bits out of the way first; it certainly looks better than the “standard” Laser prop nut.



For flying I normally use a 13" prop with a 6" pitch, but having seen how much better the model looks with a bigger prop I’ve been thinking, if the pitch was reduced to 4" then the diameter could be increased to 15" or 16". This has the advantage of increased static thrust, which is useful for overcoming drag, also the model will fly slower for any given revs so the engine can be tuned for maximum power without having to fly like a pylon racer.

Searching through the catalogues I found that APC do make a 16" x 4", the downside being that it is about 3 times the price of a 13" x 6"; do the looks really matter that much?

  

I thought that getting the litho plate to bend around the prop blade following the curved leading edge would be a problem; one reason why I got the hub etc. done first! As it turned out it was no problem at all, the most awkward thing to do was embossing the rivets as this had to be done last of all and it was quite tight to get the embossing wheel in. To make the job easier I “opened out” the litho plate somewhat then carefully refitted it to the prop blades using just my fingers and a cloth to ensure none of the rivets got flattened.



Of course it all looks very shiny and ex-works at the moment, a bit of weathering will soon change that.

I have to say that I’ve thoroughly enjoyed the last few days making the static prop; obviously it won’t improve the model as a flying model but all in all I think it’s been worth it, hope you agree.

Edited April 28, 2023 by Greyhead46

[DaveB1]

Amazing work - really enjoyed following your summary of this incredible build - thank you for sharing

[Greyhead46]

Hello Dave

 

Thank you for the reply, glad you're enjoying it.

 

Grahame

 

I’ve finished the construction section of the thread, not that the thread itself has come to the end. From now on, as well as the more mundane task of actually fitting out the model ready for flight, I’ll be adding / improving some of the detailing that I didn’t bother with earlier. Then of course there’s the flying!

The Lewis gun is very prominent, mounted as it is on top of the wing, so I thought that the magazine strap, which I’d made from masking tape, let the job down a bit. I still had some of the "thinned down" leather I’d used to make the cockpit edging so I decided to make a leather strap, complete with brass buckle.

 

The buckle was very easy to solder up from thin brass rod, but I think it really finishes the job off and was well worth the little time and effort it took. It was so easy in fact that I did the same thing to the "spare" magazine, which is just about visible through the fuselage window if you look very carefully.

 

I’ve added a bit of “bolt head” detailing to the wings but nothing of note, however, a couple of other things may be of interest.

I attached "looped" tails to the rear flying wires that thread easily through the lower wing stubs and into the fuselage; these are pulled tight and held by a pair of aluminium clamps. When prepared for flying the tails will be positioned down the side of the servos to ensure that they can’t snag the elevator or rudder controls.



With all the flying wires under tension the wings no longer need the grub screws, which lock onto the wing pins, to hold them securely to the centre sections, but a bit of extra security never goes amiss so I’ll still use them!

One major drawback of having a fully enclosed engine is that regular checking / maintenance is awkward and the temptation is to stretch the times between checks. For this reason it is important to make sure that the fuel tubing can’t slip off the tank tubes etc. and the best way I’ve found to do this is to solder a ring of thin wire around the ends of the brass tubes.



To make doubly sure a commercially available fuel tube clip can also be used but I haven’t found this necessary in the past.

 

The “plumbing” on most scale models is somewhat more involved than on sports models; with the engine being fully enclosed it’s not easy to get at the fuel pipes and we don’t what to have erroneous holes in the cowl or pipes protruding where they shouldn’t.

My preferred method is to have the filler pipe going to the bottom of the tank so that it can also be used to empty the tank, this means that to prevent the fuel siphoning from the tank the filler pipe has to exit the fuselage somewhere above the top of the tank and my original intention was to use the radiator cap to hide the filler pipe. This idea was abandoned because it would have made the fitting and / or removal of the cowl very awkward, other openings also presented problems and I was beginning to think that I’d have to resort to removing the cowl for filling and emptying the tank.

The answer was in fact blindingly obvious; it just took me a long time to see it!

The top of the tank becomes the bottom of the tank when the model is upside down and, as I have to invert the model to de-rig it, I simply have to wait until that time to empty the tank. The filler and vent pipes both go to the top of the tank and are accessed through the prop shaft opening just below the crankcase



Both pipes face forwards to provide slight positive pressure in the fuel tank and are set back somewhat into the fuselage so will be very unobtrusive.

Edited April 29, 2023 by Greyhead46

[Greyhead46]

I’ve not as yet decided how I’m going to do the exhaust. The standard silencer fits inside the cowl, of course the outlet will have to be extended to exit the fuselage via one of the many holes in the bottom pan but I’ve had problems with silicone exhaust tube in the past with my Parnall Elf.

To be fair the silicone exhaust tube in that case was fitted between the engine and the silencer so had to deal with more pressure and a higher temperature than normal



Or I could use flexi pipe to a custom silencer mounted under the engine plate similar to the method I eventually used for the Parnall Elf.



Obviously using the standard silencer is the easier option but it might be “spoiling the ship” etc.; I’ll see.

The tank has been sealed to the firewall using “bathroom” silicone sealant to ensure no fuel / oil gets into the fuselage innards; when I came to seal the tank in place I realised that had in fact prepared the wrong tank (see the photo in my previous post!)

If at all practical I do prefer to use the standard silencer rather than make my own because the manufacturer has tested it so you can be sure it will work as an efficient silencer and also won’t have a detrimental effect on the engine’s performance.

The standard silencer fits OK inside the cowl but extending the outlet is not straightforward as it involves a tight 90º bend which could cause a heat / pressure build up and possible failure of the silicone tube.

To eliminate this bend I’ve made a 90º connector using standard copper pipe fittings silver soldered together; a 15mm Tee and 15mm to 10mm Fitting Reducer. I’ve removed as much copper as possible from both fittings in order to save weight and sealed one end of the Tee with a flat plate. I used a “solder ring” Fitting Reducer because the ring formed to hold the solder makes a nice lip, which will help to retain the silicone tube.



By good fortune the silicone tube is just the right size to form a flexible, gas tight seal when the connector is pushed onto the silencer outlet.



I’ll extend the connector with either a short length of silicone tube then flexible stainless steel or silicone tube all the way to the exit.



I’ll soon be ordering a new Laser 70 for the SE5a so I’ll use the running in period as an opportunity to thoroughly test the system before committing it to the model.

[Greyhead46]

The exit pipe once again uses a 10mm copper pipefitting and in this case also some 10mm brass tube; all silver soldered. The exhaust will exit via 2 of the holes in the engine under pan, not the scale exit points but at least I’ll have 2 exhaust trails following the model.



The complete exhaust system.



I ordered a new Laser 70 today so it shouldn’t be too long before I can start to test everything. If all is OK then I’ll trim the ends of the exit pipe to be inline with the under pan when attached to the model so in practice they will not be seen.

If the system isn’t satisfactory then it’s back to the drawing board

 

My new Laser 70 has arrived and of course it ran flawlessly after starting “first flick”, not that you’d expect anything less from a Laser. I should point out at this juncture that I have no commercial interest in Laser engines; it’s just that they are such good engines that I can’t stop “singing their praises”!

Although the excuse for getting this new engine is the SE5a, in fact I’ll use one of my older, well run, engines. This one will free up one of my other Lasers by being mounted in the AcroWot and have a good few flights, first without the cowl to ensure really efficient cooling then after a couple of hours flying time with the cowl fitted. A fully enclosed installation such as the SE5a isn’t the place to bed-in a new engine.

I have modified the crankcase breather on all my Lasers to make it easier if the engine is mounted close to the firewall. A short length of brass tube is soldered to the nipple; without it the fuel tubing extension would have to make a sharp 90º bend, which has a tendency either split of restrict the breather.



I make it a “T” so it’s just as good if the engine is upright or inverted, the extension tube is fitted to the appropriate side and the other is blocked off.

I ran the engine both with and without the SE5a exhaust extension and the only difference I could tell was that the exhaust note sounded a bit quieter with the extension fitted. One thing that I did find out was that I’ll have to incorporate some form of strap because when I opened the throttle above about half way it blew the extension off. Before I do any more testing I’ll run a few tanks of fuel through the engine to get it settled down and fit the strap.

For the next tests I will get more “scientific” results on the effect of extending the exhaust by using a rev. counter, which I don’t own, but I know a man who does!

 

Although I’ve not yet done any “scientific” tests, I should be getting a loan of a rev. counter next week, I have carried on with the installation of the exhaust.

Having run the engine both with and without the exhaust extension I applied my usual method of decision making; if it looks / feels / sounds OK then it most probably is OK!

I silver soldered mounting brackets to the exhaust and squashed the ends of the pipes somewhat to fit through the holes in the under pan. 2 hardwood blocks epoxied to the fuselage sides to screw into completed the installation.



I trimmed the pipes so that with the under pan fitted the exhaust doesn’t protrude and isn’t visible when viewing the model normally.

 

Edited April 30, 2023 by Greyhead46

[Greyhead46]

I said in my last post “A fully enclosed installation such as the SE5a isn’t the place to bed-in a new engine “. Of course it isn’t ideal for any air-cooled engine new or old so extra care has to be taken to prevent over heating; the full sized was water-cooled. To explain how I hope I’ve overcome the problem I’ve used some old photos as well as a couple of new ones.

The SE5a has some positive points; there is a hole directly in front of the cylinder head for cooling air to enter.



And there is a large opening in the fuselage behind the engine for heated air to escape.



However, things are not as straightforward as they seem; hot air wants to rise but the exit hole is in the bottom of the fuselage. Once the model is moving forwards the air flowing over the hole should produce a venturi effect and suck the hot air out, but this is where the SE5a designers throw another spanner in the works!

Below the engine shaft there is an even bigger air entry hole, air entering here will not help with the cooling but will have the effect of “pressurising” the bottom half of the engine compartment and restricting the outflow of hot air.

The easiest solution would be to mount the engine inverted but because of the position of the engine shaft this would mean either a much larger model or the cylinder head sticking out the bottom of the fuselage, so this is a non-starter.

The engine plate has a hole corresponding to the one in the fuselage, so the hot air can transfer to the bottom half of the engine compartment.



A template, this is where you’re glad you paid attention in all those geometry lessons, is drawn and stuck to litho plate.



When cut out and bent up it produces an air scoop that forces air entering below the engine shaft upwards into the top half of the engine compartment where it can help to cool the engine.

 

Also with little or no pressurising of the bottom half of the engine compartment the venturi effect can work and suck out the hot air; a win / win situation!

[paul devereux]

@Greyhead46 Beautiful model. I've just thought, is this build on YouTube?

[Greyhead46]

Hello Paul

 

Thanks for the reply, no Youtube, just photos not video!

 

Grahame

 

I’ve made a clamp to stop the extension being blown off the silencer; it’s similar in concept to the Laser carburettor and exhaust clamps.



It’s made from tin plate and I’ve used soft (lead) solder. I don’t think it will get hot enough to melt the solder but even if it does it won’t matter because once it’s clamped in position it holds itself together.



I’ll use a cable tie around the silicone tube.

I’ve done the exhaust tests and the results are very promising. The clamp has held the extension in place and the silicone tube used as a seal has worked well. I’m using the AcroWot as a test bed and now I know that the basics work I’ll cut the silicone tube to the correct length and attach the exhaust extension well enough to fly with it to give the system a thorough test before fitting in the SE5a.



I tested the revs using 5 different props altogether, both with and without the exhaust extension; for scale models such as the Parnall Elf I have in the past used a 13x6 but I’m thinking of using a 16x4 for the SE5a

Prop 13x6 Revs without extension 9700 Revs with extension 9400

Prop 16x4 Revs without extension 7000 Revs with extension 6800

The revs have dropped somewhat with the extension fitted but not to any significant amount and although the revs are considerably less with the 16x4 prop the static thrust is increased. I have no way of measuring this but just holding the model it was clear that the “pulling power” with the larger diameter lower pitch prop was much greater. The theoretical flying speed will be less of course but then it will be better able to overcome drag, which the SE5a has in abundance, so the end result could be the same!

 

The exhaust did get hot enough to melt the solder joints for the strap, which used soft “lead” solder. It didn’t affect the working of the exhaust but I decided to silver solder it to eliminate any chance of “metal to metal contact”, which could cause radio interference.

 

Edited May 2, 2023 by Greyhead46

[Greyhead46]

Well I was a bit hasty silver soldering the strap.

I took the AcroWot out today to test the exhaust in the air and basically everything worked OK but the strap broke so I’ll have to replace it with one made from thicker steel, I’d used tin plate and obviously the vibration proved too much. I only had 3 flights but the silicone tube used as a seal shows no sign of deteriorating and the reduction in revs had no noticeable effect at all.

I’ve decided to include the prop size / revs data in this thread as it is relevant to what follows.

For the first 2 flights I used a 12 x 7 prop, which gives a theoretical top speed of 64 mph. I then tried a short flight with the 16x4 prop, being a new engine I didn’t want to run it for too long with such a large prop. With a theoretical top speed of only 26 mph I expected to see a vast difference but in fact there was very little difference when flying into the fairly strong wind, with the wind the difference was noticeable but not excessive. As the AcroWot isn’t a particularly “draggy” design I’m very surprised by this but it bodes well for using a 16 x 4 prop on the SE5a.

 

Positioning the RC switch is often not the simplest of tasks on a scale model because it needs to be easily accessible but at the same time we don’t want it to be visible; there are really only 2 alternatives, either hide it or disguise it.

The most practical way to hide the switch but still be able to easily operate it is to use a scale hatch and the SE5a does have one but it is behind the cockpit so not ideal, disguising the switch seems to be the way to go but how to do it?

 

This photo provided the answer.

 

The photo shows the interrupt pump driven by the prop shaft, of course this isn’t included on any of today’s restored aircraft because the Vickers is inoperative. But it does explain why the front has such a large cut out below the prop shaft!

The switches, I’m using a twin power pack, 2 batteries with an electronic switch over for added security, are operated by pulling / pushing the pump gear wheel, from a defunct servo, which is connected via a “snake”.

 

Not only does it make the perfect “invisible” switch, it adds a bit more interest to the front end,

[Greyhead46]

Because the basics were worked out well in advance the radio gear has all fitted in with no unexpected problems. The throttle servo, bomb release servo and onboard / remote glow driver are located in the engine compartment, their leads have to pass through the firewall so the resultant hole needs to be sealed to prevent the ingress of fuel / oil into the main fuselage.

A standard grommet won’t work very well because the leads have their plugs permanently fitted, so I made a 2 part seal from a rubber wine bottle “cork”; another of those useful Christmas necessities!



An appropriately sized hole is drilled through the firewall, one half of the seal fitted in place, the leads / plugs threaded through, then the second half fitted, the rubber compresses around the leads forming a good seal.



The onboard / remote glow is located in the engine compartment in order to keep the glow plug power leads as short as possible, the control unit and battery are wrapped in foam and held in place by a cable tie. I’ve complicated the wiring somewhat by including a stereo, switched socket, which enables me to use a remote connection for starting and also, using a different jack plug, to charge the battery.



Previously I've used a homemade onboard glow driver, which simply switched on the glow at a preset throttle setting. But this one is an “intelligent” glow driver; it continuously monitors the glow temperature (actually the resistance of the element) and connects the battery only when required, whatever the throttle setting happens to be. In theory it should be a lot better, but I’ll let you know how it actually works in practice at a later date!

 

I ran the Laser up to test the plumbing and adjust the onboard glow. The driver has an LED, which shows when current is being supplied to the glow plug; this is very useful as you can see exactly what is happening.

Having adjusted the driver following the instructions, one thing became quite obvious; with a Laser you really don’t need an onboard glow driver. I’ll still use it if only as extra security just in case a plug goes faulty when I’m sure it would help.

 

[Greyhead46]

I’ve altered the exhaust slightly by incorporating a 10mm copper 90º bend, I found that as the silicone pipe heated up with the engine running, it tended to collapse slightly at the apex of the bend so restricting the gas flow. I have to admit that I didn’t notice any change in the engines performance but decided that it’s better to be safe than sorry.

Connecting the remote glow lead to the plug can be a problem with Lasers, along with most other four strokes, because the plug is recessed quite a way into the head; there are many methods used to make this connection, this time I’ve used the threaded section from a 3mm “quick link”. A very short bolt holds the solder tag and the rest of the 3mm tapped section pushes tightly over the plug post, being made of “springy” metal the threaded section can be adjusted slightly if necessary to ensure a tight fit by squeezing with a pair of pliers.



The power lead is sheathed in silicone fuel tubing for heat proofing.

 

I was chatting in the model shop recently and as is often the case when I’m involved the subject turned to biplanes. The consensus was that they’re more trouble than they’re worth, just too fiddly to set up and awkward to takeoff and land in anything but ideal conditions. I pointed out that my Elf, with its folding wings, is easier to rig than most sports models; but I had to admit that it is the exception rather than the rule! I also had to admit that the flying characteristics of biplanes can be “challenging” but in my opinion the inconvenience of rigging is far outweighed by the appearance of the assembled model and that it doesn’t have to be all that time consuming with a bit of forethought.

My decision to use the scale closed loop control system for the ailerons on the SE5a has made the assembly somewhat more difficult and as I explained the procedure it did sound horrendous, which in reality it isn’t. It’s one of those things that sound complicated in words but when you see what is involved it isn’t, so I’ll take some photos as I assemble the model.
 

Edited May 5, 2023 by Greyhead46

[Greyhead46]

I’ll start with a photo of the “tools” that I’ve made to make the rigging easier and a brief explanation; it will be much clearer what they do when I post the assembly photos.

 

From left to right:

One of a set of 4, which keep the cables tight during storage and whilst connecting the first quicklink.

The third version of a tool to hold the quicklinks open

A spring-loaded quicklink

Tensioning tool for the rear flying wires.

 

So here it is, the method of assembling the SE5a with closed loop aileron control!

The fuselage is held inverted in the assembly support and the underpan and wheels removed.



The aileron control is not strictly a “closed loop” because there is in fact a gap between the two servo arms, although in operation this gap stays the same effectively making it a closed loop. Moving both arms outward in essence makes the loop longer enabling the quicklinks to be accessed with the wings not fully butted up to the fuselage and centre section. Notice the spring inside the fuselage pulls the servo links inwards.



During storage, with the ailerons at neutral, the quicklinks just protrude from the root ribs and they are kept under tension by button thread attached to springs.



The top quicklink is pulled out from its wing tube and is held open and in a protruding position by my third generation “special tool”, at the same time the spring tension on the lower quicklink is eased to allow it to retract into its wing tube but still maintain tension in the cables. As you can see the ailerons are now deflected up.



The spring-loaded quicklink is attached to the link on the opposite side of the centre section.



The wings are then located loosely in position and their supporting frame removed; the bottom wing grub screws are tightened slightly to hold the wings in place whilst allowing the top wing to slide somewhat.



Pressing on the spring-loaded quicklink pushes out the centre section link, which is attached to the wing quicklink and the special tool removed, tension in the cables being maintained by the spring-loaded quicklink.

At the same time as pushing on the spring loaded quicklink the bottom wing quicklink is pulled from its wing tube and held in position by the “special tool”. The servo link is pushed out from the fuselage against the internal spring tension, engaged with the wing quicklink and the “special tool” removed. Tension being maintained by the spring loaded quicklink and the internal spring.



Both wings can now be pushed fully home and the bottom wing grub screws tightened. The tails of the rear flying wires are threaded through their anchor block and hooked to another of my special tools. Pulling on the ring applies equal tension to both wires and the bolt is then tightened; this is the most awkward part of assembling the model and is a procedure that would be simplified if I had three hands!



The spring-loaded quicklink is removed and the centre section link held in position with a pair of forceps, tension now being maintained purely by the internal spring.



The whole procedure is repeated for the other wings except this time the internal spring replaces the spring-loaded quicklink.

Finally the wheels and underpan are fitted, the model turned the right way up and the top wing grub screws tightened.

 

Edited May 5, 2023 by Greyhead46