Airbus A350 for 50 mm EDFs

[Simon Chaddock]

Having established I could print light exhaust tubes I looked around for potential planes to use it on.

I rather fancy doing another airliner as the big long fuselages are well suited to Depron construction even if the thin high aspect ratio wings are less so.

I noticed that the exhaust ducting on the RR Trent used on the Airbus A350 has similar proportions to the 50 mm EDFs that i already have.

With a small thrust tube extension and a motor tail cone you have a reasonable (for an EDF installation!) representation of the Trent exhaust.

Next to superimpose a 50 mm EDF into a scale A350 nacelle.

Although the fan is a long way back it gives an admirably simple inlet duct with an area 1.2 times the FSA..

The next question is what size would the A350 come out using 50 mm EDFs? The answer is surprisingly small at just 46" (1170 mm) span, indeed, to my mind almost too small given the A350's high aspect ratio wing!

Looking at the nacelle I wondered if 'printing' the complete inlet tube made sense. The whole duct would thus become structural meaning the outer nacelle profile would be just a lightweight Depron skin.

With my bin slowly filling up with 'trial' prints I ended up with this.

Everything is either single wall (0.4 mm thick) or hollow so the complete duct only adds 7.3 g to the 61 g of the bare EDF and what is 'not there' means it doesn't use that much filament either..

So far so good but 'under wing' EDFs require a relatively strong heavy mounting compared to 'in fuselage' and such a configuration does not make for a good 'rough field' belly lander!

At some point soon I will have to make a go/no go decision.

[Denis Watkins]

Wow, as you point out, the small wing Simon

What about an A350 configured for landing ? with all its bits out,

Flaps deployed, leading edge slats out, wheels down,

So much more wing?

[Simon Chaddock]

As I suspected making the pylon strong and stiff enough to support an EDF means it ends up as heavy as the total of all the nacelle duct work!

The Mk1 pylon temporarily glued in with the EDF with the slightly lighter 'Mk II' duct.

.On the full size the ptlon extends quite a bit further along the top of the nacelle but with the EDF so far back mine does not have to so it can be simply made up with Depron.

In the final version the EDF wires will be extended and run up inside the pylon into the wing.

I intend to complete one nacelle with its Depron skin and then decide if I really want to go any further!

If I do proceed I will have to work out how to make these.

.

[ceejay]

put it out in the rain then on a radiator should do it fine!!!!

see you sunday simon

[Stephen Jones]

Look forward to this one.

Or maybe if the pylon prove difficult you could try the Antonov An- 72/74

Steve

[Simon Chaddock]

A short vide of the nacelle under test.

The complete nacelle and ducting adds just of 10 g to the bare EDF.

It draws 10A from my rather tired 'bench test' 2s. This rises to 12A from a ;good 2s;.

[Erfolg]

The Antonov looks a very practical model.

Certainly the vulnerable nacelles are now no longer an issue. The underside of the fuz, could be made a tad more robust and Bert,Rob, Robo, Robi, or is it. Bob is the answer, although I strongly suspect his real name is Robert.

[Simon Chaddock]

The Depron outer skin of the nacelle is certainly light but it is rather time consuming to do so I looked at the possibility of printing the complete nacekke in one piece.

There are two issues in doing this. It will weigh a bit more and due to FreeCAD limitations the nacelle has to be 'symmetrical about the longitudinal axis but unfortunately the Trent nacelle' is offset being 'thicker' on top than below..

Although it is designed as a single "stl" file, including the pylon, it has to be printed in two halves in order to insert the EDF.

The two parts are created by simply 'submerging' the print by exactly the right amount below the bed surface first one way up and then the other. If done correctly the two halves exactly match.

Temporarily glued together with the EDF inside

Overall the all printed nacelle weighs 20 g or about 9 g heavier. It takes nearly 3 hours to print both pieces but this is quite a saving if the time taken to Depron plank is included and of course it is exactly reproducible, An important asset when building a twin!

The outer profile of the printed nacelle is still not quite right but it can wait until the wing and the nacelle mounting have been designed. Before that happens however the issue of wing area and weight has to be addressed.

When the nacelle is sized to mount a 50 mm EDF the wing span comes out close to 48" (1200 mm) but with the highly tapered A350 wing it would have a total area of less than 1 sq ft. At a guesstimated weight of 25 oz (750 g) the resulting wing loading would not be easy (or even possible?) to hand launch.

In simple terms it has either got to be lighter still or bigger or maybe a bit of both!

Still thinking.

Edited By Simon Chaddock on 19/10/2017 17:05:36

[Erfolg]

Simon, which printer do you have?

Another question, one of the Timperley club members has architectural models built. He said that 3d printed items weighed a ton. Is it a difference in materials being used, or technique used for printing that allows low weight items to be produced.

[Simon Chaddock]

Erfolg

My printer is the cheap Anet A8.

I agree '3D printing' tends to be 'heavy' unless you go to some lengths to minimize the amount of filament that is used by using the sort of techniques used for the 'printed' RC planes..

The all printed nacelle uses PLA (the lightest printable plastic?) with walls that are 0.3 mm thick which is just about as fine as my printer will go and still ensure each layer sticks together properly.

In addition the internal structure and its pattern are carefully chosen to try to place the elements only where they are required to obtain adequate rigidity.

In this case 3D printing appears to be a valid option because:

1 The nacelle is a complex but 'pure' geometric shape so it can be mathematically generated relatively easily.

2. Compared to the rest of the plane the nacelle is relatively small so any weight penalty is limited on the all up weight.

3. The nacelle requires an accurate and smooth inner surface for the EDF to function well..

4 The nacelle and its pylon can be printed as one piece eliminating the need for joint in a high stress area.

[Erfolg]

Simon. I never had any doubts that the weight penalty was acceptable, if you chose the option. My difficulty was reconciling an opinion that 3d printing was heavy, from some one else i know to be reliable.

I have been tempted myself to buy a printer. Just because i am interested, particularly.

As the person who brought in the first Autocad set up into the company where we both worked for ( and threatened with being sacked), 3d printing intrigues me. Like CAD, I am not convinced it is the answer to everything, yet it does seem to have a place. Now that we are both retired, it is principally of interest to us both, I am guessing.

It has annoyed me that the IMechE recently had a lecture on all the forms of 3d printing, currently in use, not just the thermo plastic types, in that I now live to far away to conveniently get back to UMIST.

I will be intrigued as to how the mechanical property aspect stand s up to everyday flying.

My Natter still looks at me asking to be completed.

[Simon Chaddock]

Erfolg

I tend to agree the current 3D printed planes are a technology marvel but not really a good solution for the 'rough and tumble' of RC flying for mere mortals!

For the A350 its projected high wing loading is forcing me to think again!

The first solution is to simply make it bigger from 48" (1220 mm) to say 60" (1830 mm) and if mist of the thicknesses remain the same it will be proportionally lighter.

Of course the engine nacelle will be bigger as well with an internal diameter of 3" (77 mm) but with such a light weight it definitely will not need the power or more important the weight of two 77 mm EDFs!

This shows how big the Trent fans are in relation to the fuselage.

The fan is in the correct relative position.

What I need is a 3" EDF that is about half the weight of even a 'light' conventional one and with only half the power it would only need half the battery as well to get a wing loading closer to 12 oz/sq ft than 25! .

We shall see!

Edited By Simon Chaddock on 23/10/2017 15:24:45

[Simon Chaddock]

I think I have a solution for a 3" super lightweight EDF but the bit are coming from China.

In the mean time I will build the centre portion of the fuselage (the bit over the wing) of the 60" span A350 as a test to find out how practical the 'printed' formers are.

.

One advantage of.generating formers mathematically is the ease with which they can be modified to give the flat underside required over the wing.

The 7 printed formers required for the section of fuselage covering the wing weigh just 7.4 g in total.

It will be planked in 2 mm Depron.

At this stage still just experimenting.

Edited By Simon Chaddock on 28/10/2017 00:47:44

[Erfolg]

Are the structural properties of the formers similar to Depron? I guess at best similar to lightply.

It has been the formers that have dissuaded me from a number of models such as B57 Canberra, although i have been toying with the idea where the fuz is a cylindrical tube, making a blank, around which either balsa or very thin ply could be wound.

In the case of these 3d printed formers, how will you (or with what adhesive) stick/join the Depron to the formers together?

[Simon Chaddock]

Erfolg

The printed formers are considerably stronger/stiffer than Depron but being just a channel section they twist fairly easily. Of course when covered with a skin this flexibility is not a issue.

To keep a reasonably smooth circular surface each plank is only 10 mm wide so there are quite few but on a parallel section of fuselage this makes it repetitive rather than difficult.

The test fuselage planked in 2 mm Depron with 'flat' section covering the wing.

Built vertically or at least it was until it has sufficient planks in place to be handled.

As UHU POR is almost impossible to sand this is only used to glue each plank to the formers which it does very well. At the same time PVA is used to glue each plank to its neighbour.

The POR holds each plank securely in place straight away so several planks can be positioned one after the other although handling limitations ultimately limit how many.

Left to harden for 24 hours the Depron surface can be safely (but gently!) sanded.

5.5" (140 mm) diameter and 14.5" (370 mm) long the 'test' fuselage section weighs 26 g.

Extrapolating this to the full 62" (1575 mm) long fuselage suggests it would weigh about 100g (3.5 oz). As at this size an A350 will have a wing area close to 2 sq ft such a fuselage weight is on course to achieve my target wing loading of 12 oz/sqft - but a lot will depend on the weight of the 3" fans!

Overall the 'printed' fuselage formers work very well creating a stiff but extra ordinarily light structure that looks almost scale from the inside as well!

**LINK**

[McG 6969]

You really are amazing, Simon.

Following with great interest.

Cheers

Chris

[Pete Collins]

Hi Simon

Just discovered this Thread. Referring back to your earlier posts: When you printed the duct structure for planking with depron, I assume you printed it with the duct vertical, so what sort of support structure did you use to support the webs/formers while printing?

Pete,

[john stones 1 - Moderator]

Yep watching as well.

[Simon Chaddock]

Pete

Yes the EDF duct was printed vertically as just a single wall tube. The formers were printed separately but sized to be a good sliding fit over the duct so they could be simply glued in place. UHU POR sticks PLA very well.

The complete nacelle and pylon were thus made.up of 8 individual printed parts. The inlet ring, the duct tube, two formers, the short exhaust tube, the out runner tail cone, the pylon and a pylon bracket.

It takes a bit of trial and error printing to get all the parts to fit exactly but once done the second nacelle is easy. .

This is almost certainly how I will make the bigger nacelles with their 3" fans for the 60" A350 although they will likely have 3 formers rather than 2, however the final design is '/on hold' until the bits arrive from Hong Kpng.

As it will no longer use 50 mm EDFs I will also have to get the title of the thread changed!.

[onetenor]

Talking about sanding POR glue . I made a trial fuz with 2mm depron . Circular section with seam underneath and POR glued. Not being ready to proceed further I kept it on the radiator near my chair. It was there for a couple of weeks.When I picked it up again there was a cracking sound and the joint opened. On examination the POR had gone crystal clear and brittle. Tony B must have seen this before as he advised I must sand the brittle glue off before attempting to re glue the seam again.Obviously the heat had accelerated the glue drying out like this but will it eventually happen over a period of time anyway? Time will tell if our Depron models fall apart I suppose.

[onetenor]

Talking about sanding POR glue . I made a trial fuz with 2mm depron . Circular section with seam underneath and POR glued. Not being ready to proceed further I kept it on the radiator near my chair. It was there for a couple of weeks.When I picked it up again there was a cracking sound and the joint opened. On examination the POR had gone crystal clear and brittle. Tony B must have seen this before as he advised I must sand the brittle glue off before attempting to re glue the seam again.Obviously the heat had accelerated the glue drying out like this but will it eventually happen over a period of time anyway? Time will tell if our Depron models fall apart I suppose.

[Pete Collins]

Thanks Simon,

​ ​I thought you'd come up with some magic method of printing it all in one piece!

Pete

[Simon Chaddock]

OT

Yes POR does harden over time but my oldest 'stressed skin' wing (in other words it relies on the glued joints for its strength) is still flying after 6 years.

I found that Depron itself becomes considerably more brittle over such lengths of time.

What I try to avoid is relying on the glue to hold the Depron into shape as this in effect stressed the joint (or rather the delicate Depron immediately adjacent to the glue) before any 'flight' load is applied. I either preform the Depron (and use its grain where appropriate) or use many individual pieces (ie narrow planks) to keep down the load required to bend each one.

[Simon Chaddock]

I am afraid the thread title is now a bit misleading as what it looks like it will definitely not be using 50 mm EDFs in fact it will not be using fans at all but ducted props, four blade 3x3.5 for a racing quad! .

As the test fuselage has gone together well it will be used as the basis for a 62" (1570 mm) span Airbus A350 -1000.

First it is extended forward with a parallel section to the point where the fuselage starts to narrow towards the nose.

Next it will be extended rearwards to the point where the fuselage narrows for the tail section.

In preparation the former 'set' for tapered rear fuselage has been 3D printed.

The resulting shape is rather complicated. It starts off round but then significantly narrows to a vertical ellipse in the area of the tail plane and returns to a true circle at the beginning of the 'tail cone' itself.

This continuously changing section will make it a bit of a nightmare to plank!

[Simon Chaddock]

The nose and tail sections under construction.

The little printed cone is the APU exhaust that goes right on the end.

When each section id fully skinned (planked) the base former is carefully cut out so the ends of the skin can be carefully sanded to make sure it fits 'true and square' on the the preceding fuselage section.

To give more glue area on what would otherwise be a butt joint a 2x2 mm flange is glued to the former.

All five sections glued together.

61.5" (1560 mm) long and at 96 g it is just within my guesstimated target of 100 g.

Next is to start thinking about the wings although I really need the motors and fans (props) before I start building iti