By continuing to use this site, you agree to our use of cookies. Find out more
Forum sponsored by:
Forum sponsored by CML

A printed but tissue covered wing

An 'assembly' of printed parts

All Topics | Latest Posts

Search for:  in Thread Title in  
Simon Chaddock28/05/2020 13:24:04
avatar
5717 forum posts
3034 photos

I considered using printing to make a bi plane version of my 3/4 size Depron Ballerina but had to call a halt as it looked like the weight would reach un acceptable proportions.

Trying to reduce the weight I adopted a part tissue covered solution but the weight was still unacceptable but the concept of building a complete wing from just 3D printed parts looked interesting and worthy of some further development.

So here goes.

The new target was to build a printed wing to replace the 40" (1000 mm) Depron version used on my 3/4 size Ballerina. This would allow a direct comparison between the two although the result could be easily predicted. The Depron wing would be lighter and more robust!

As this would be an all printed structure the first question was could something like a spar be printed at all?

Although PLA might be sufficiently strong it would involve two issues:-

1. The printer bed would limit the size of any part to a maximum of 220 mm.

2. What shape of spar, within the limits of 3D printing, would give the best strength to weight ratio.

Like wood a single layer of printed surface has a 'grain' being significantly stronger along the line of the print bead rather than across it.

My proposed 'spar' would thus be a rectangular box with one or more print layers top and bottom with a lightweight 'fill' pattern between them to provide the necessary shear resistance.

Indeed it would be a printed version of the balsa/Depron/balsa spar actually used on the Depron Ballerina wing. wink 2

Whilst printing such a shape is easy enough it does require a bit more work on the CURA print parameters to get the required print directions.

The first trial spar.

outsprtest

200 mm long it weighs just 4.9 g yet can easily carry a centre load of 510 g (18 oz) with no visible bending.

So far so good but the final spar will have to be 5 times longer and with its parts just glued together. smile o

Edited By Simon Chaddock on 28/05/2020 13:24:56

Doc Marten28/05/2020 13:40:44
585 forum posts
8 photos

Do you mean H section spars?

Edited By Doc Marten on 28/05/2020 13:42:41

Simon Chaddock28/05/2020 16:40:25
avatar
5717 forum posts
3034 photos

Doc M

An H section would be possible but it is a difficult shape to print with the H lying horizontally. There is no support for the upper flange during the print process. A box beam with an suitable infill ensures the top flange is adequately supported particularly if it is very thin.

In the above example both the top, bottom and side walls are just 0.25 mm. thick. The internal structure has to be very carefully positioned to prevent any surface from buckling when under compression.

This is actually the prototype for outermost spar section. It thus carries the smallest bending load so it is printed as light as possible. The inner spar sections will have to be significantly stronger.

Peter Miller28/05/2020 18:38:05
avatar
11125 forum posts
1309 photos
10 articles

That looks very interesting

Simon Chaddock29/05/2020 13:13:49
avatar
5717 forum posts
3034 photos

The best way to 'join' two rectangular box spars together would be for one to slot into each other.

Such an arrangement would also suggest each section would be slightly larger than the neat which would actually be beneficial as far as beam stiffness was concerned.

I had in previous builds I had successfully used printed ribs. Very little heavier than even 3 mm Depron and considerably more rigid.

rib2big

A 'test' print of 3 'plug in' spar sections with the associated ribs.

spar1

Each section has a 10 mm deep close fitting 'socket' to accept the next spar.

spar2

Using super glue I judged that a 10 mm overlap would allow a full transfer of the tensile and compressive loads between the spars sections.

As it would require a total of 5 spar sections to achieve a 1000 mm span and allowing for a 10 mm overlap will mean the centre section needs to be 220 mm long, the mid one 210 mm and the outer 200 mm. The centre section would also have to incorporate any dihedral angle but at least this would mean no joint at the point of maximum bending. smile

With the test piece giving the exact dimensions for fit it should in theory just be a case of extending a single dimension to give the required spars but as each part takes nearly 2 hours to print any errors will represent a significant wast of printing time. smile o

Simon Chaddock30/05/2020 23:05:47
avatar
5717 forum posts
3034 photos

Nearly 10 hours to print but the full 1000 mm spar can be assembled.

fullspar2

At this stage it is not glued together as it will make life much easier to add the ribs to the centre section before adding the next spar section. when those ribs are added finally the out spar section will be added.

Slowly going forward. wink 2

Simon Chaddock01/06/2020 18:17:47
avatar
5717 forum posts
3034 photos

The ribs on the outer spar sections are one piece and are threaded onto the spar.

A 'dry' run of the components of an outer panel.

outdryfit

The aileron spar and the leading edge are a single skin and fit over the rib which is locally stepped down to accommodate them.

The spar centre section is the full depth of the wing so the ribs are divided and simply glued on to the front and rear face of the spar.

cntrsctn

The trailing edge is a simple "V" with the ribs stepped to accommodate it.

The mid section of the spar are now glued on.

Although an all printed structure the construction is actually very conventional so not only is there considerable printing time but careful glue assembly too.

It could be argued that as a concept is this method adopts the he worst features of both worlds! wink 2

Edited By Simon Chaddock on 01/06/2020 18:19:11

Simon Chaddock03/06/2020 13:13:10
avatar
5717 forum posts
3034 photos

With all the parts printed the assembly is fairly quick.

For light weight both the leading and trailing edges are just hollow printed skins.

hollowle

The ribs have an local 'indent' so the surface of the leading edge lies flush with the rib.

The complete wing in front of the small Ballerina that it will be fitted to.

wing1

The bare printed wing weighs 90 g so is likely to be about 120 g when covered, doped and the servos included.

It is worth noting the complete balsa/Depron wing ready to go, including servos, only weighs 79 g and is probably stronger too! smile o

Simon Chaddock06/06/2020 22:36:20
avatar
5717 forum posts
3034 photos

With the Depron wing carefully cut away from the fuselage suitable wing mounts for the M4 nylon bolts and threaded fuselage mountings could be designed and 3D printed. Printed mounts for the aileron servos were added.

Took the best part of a day of trial and error to get it exactly right.

wingon1

There are two bolts just ahead of the spar and a single central bolt at the wing trailing edge.

The servo wires run through holes in the ribs 'melted' with small soldering iron.

This allowed the tissue covering to start.

tissue1

Simon Chaddock08/06/2020 14:33:13
avatar
5717 forum posts
3034 photos

Testing the completed wing with ailerons added.

Although quite adequate the tissue does not stick particularly well to the PLA even after it has been doped so it gives a rather blotchy appearance over the red framework.
As there is no foam in the wing it could safely be painted with a 'rattle can'. Even a light application of "chrome silver" gives a very 'solid' covering.
silverwing2
The wing ended up weighing 132 g so adding 51 g compared to the Depron original but at 464 g (16.4 oz) ready to go and with 160 Watt available it should still have a 'spirited' performance.
Simon Chaddock09/06/2020 08:33:02
avatar
5717 forum posts
3034 photos

With the particularly calm weather yesterday it has it maiden and it flies.

Sorry about the video I did not centre the camera correctly so it was always pointing a bit to the left.

Plenty of power and it did loop and roll without breaking.

Does this method of construction have any particular advantage? Not really unless you are desperate to build something, have a 3D printer but no balsa or foam! wink 2

It has been an interesting exercise.

Now if the printed structure could be made from a plastic that could withstand the heat involved for a shrink film and with a carbon spar the result might be rather more practical.

Edited By Simon Chaddock on 09/06/2020 08:33:30

Simon Chaddock06/07/2020 12:19:37
avatar
5717 forum posts
3034 photos

Although the first "printed & tissue" wing worked well enough it ended up a bit heavy at least compared to the Depron original. Could it be made lighter?

A tapered wing with a tapered spar better follows the shape required to resist the bending loads. The spar was still going to have to be in 5 sections limited by the printer bed size but the structural design of each section could be adjusted to better handle the loads involved.

As before the root spar section incorporates the dihedral.

sparroot

Given this particular plane with a Clark Y wing section is likely to be able to pull more positive g than negative I decided as a further experiment to incorporate some carbon tow into a notch printed on the surface of the lower spar flange.

spartow

Otherwise the construction was much a before with printed ribs that were fed in sequence over the spar.

ailerons1

The only difference was the addition of hollow printed wing tips not for any aerodynamic or aesthetic reasons but to support the outer rib against the tissue tension.

The wing is held on with the same three bolts so both printed and Depron wings are interchangeable.

taper-depron

As before it can be rattle can painted after being doped.

taperwing

To within one gram it is exactly the same weight as the Depron wing although it does have very slightly less area.

I am sure it will fly well but the weather will have to calm down some before it can have its maiden.

Still quite a time consuming task to actually put the parts together but atleast it equalled the Depron in weight.

The next question is keeping to the same 1000 mm span would it be possible, using no Depron or carbon, for a complete plane to weigh less than 250 g?

Edited By Simon Chaddock on 06/07/2020 12:23:09

All Topics | Latest Posts

Please login to post a reply.

Support Our Partners
Slec
electricwingman 2017
CML
Sussex Model Centre
Advertise With Us
Sarik
Latest "For Sale" Ads
Has home isolation prompted you to start trad' building?
Q: The effects of Coronavirus

 Yes - for the first time
 Yes - but Ive bashed balsa before
 No - Ive existing projects on the bench
 No - Im strictly an ARTF person

Latest Reviews
Digital Back Issues

RCM&E Digital Back Issues

Contact us

Contact us

Magazine Locator

Want the latest issue of RCM&E? Use our magazine locator link to find your nearest stockist!

Find RCM&E!