Printing with Light Weight PLA

Let's title this on Repairing with LW-PLA.  One arrival on my not-as-calm-as-i-hoped ended with a ragged wing tip. Good excuse to explore how to repair LW-PLA.

 

My idea was to reprint the end section of the wing and the splice it in.  I was thinking that I could meet up the shear webs of the wing and the replacement to do most of the heavy lifting. I used a stair-step approach to engage multiple shear webs.

 

 

I initially considered some PLA straps glued to the surface but then I had the idea - I have the underlying CAD for the wind prior to adding in the shear webs - why not just print a the end of the wing as a hollow and then glue it over the patch to spread the stresses.

 

 

Although the sleve weights very little I printed its mirror and added that to the other wing tip.  Having an extra layer near the tip seems to me to be a rather nice way to give a bit of extra abrasion resistence at the cost of a couple of grams.

Oh, I used thick CA on the sleeve, kicker on the wing and then held it in place for a half hour. I did not bother to sand down the join.
 

jon
I wonder if rather than try to salvage any of the existing damaged tip area it would have been better to have just glued on the new tip with a straight join in line with the airflow. Then sanded the join smooth and applied narrow strip of single wall PLA, not LW-PLA, over the join. You could always apply the same strip on the other wing for symmetry.
I would also worry that your 'double skin' wing tip, apart from any weight increase, would encourage any damage to be transferred elsewhere on the wing in a similar event. 

I am a firm believer on delicate structures to sacrificially absorbing impacts rather than trying to resist them.
An example is that despite the total 'compression' destruction of over 30 cm of the bigger Skyray in a 'lawn dart' nose plant it left the rest of the airframe totally undamaged.
 

Thanks for the comment.  My thinking was that if I cut the wing off that would leave two 'open' ends with minimal surface area to glue to. I figured that weak point (no continuous shear web) would come back and bite me. Remember, my "almost-vase-print" does not allow for the filament to run along the span.  The other reason is that at my skill level - catching a wing tip on a landing is not a rare event especially since calm days are a rarity here.. Yes my approach shifts the stresses elsewhere.

I might do a test splice of a couple of wing sections with your idea of a single wall PLA strip at a join and see if that resists bending in a similar way to the LW-PLA sleeve.

I think a compromise would be to look at running a CF spar to the end of the tip - it might be a smaller diameter than the main spar but would ensure that there is no sudden drop off in the structure where stresses could build up.  

I do seem to have a habit of being conservative on the design and construction of wings.  Granted, I did not attempt the 1kg span test on the printed wing.  I will certainly try testing some alternative joins in the future.

On the flying thread I asked for advise on a LW-PLA fuselage that did a better job at cradling the wing. I took the earlier fuselage and stretched it to match the chord plus a bit, shifted the cut-out for the elevator servo back a bit, widened the fuse just a bit to allow a bit more space for the receiver and integrated in a bit of a platform at the back. The wing profile was subtracted and set 10mm into the fuselage..

I will create some PLA straps to accept the hold-down bands at the leading and trailing edges at the bottom of the fuse.  Here is what the CAD looks like thus far....

 

I think the answer is yes.  The bottom of a Clarke-Y is flat so the inset platform is flat (parallel with the boom), with the leading edge of the cut-out matching the shape of the leading edge of the wing at the root.  My understanding (could be wrong) was that the camber of that airfoil is still positive. The original flat-wing-platform was also parallel with the boom.  Here is a side-view. The green is the boom....

 

If the flat surface at the base of the wing should not be parallel with the boom what other orientation would folk suggest?

 

A 'standard' Clsrk Y section  with an 11.7 T/C ratio has 2.5 degrees effective incidence with the airflow parallel to the flat under surface.
For the best lift to drag for slow flying the incidence can go up to 7 degrees in other words the wing mount can be +5 degrees to the boom. The  CoG would have to go forward some...
In addition you would need down thrust on the motor (or compensate with the elevator) to stop it climbing positively as you increase the power..   . 

I have printed out the LW-PLA fuselage for the Dragonfly (shall we call it version 3)?.  I is designed around the LW-PLA wing. It is three bits which share the CF boom. Here is a side view with the open top of the battery compartment in line with the platform for the wing.

 

There is a PLA motor mount which will be glued to the front.  I will glue a 1mm x 5mm x 140mm PLA to each side of the top inner edge of the fuse (the LW-PLA is rather thin along the top edge.  The dimensions of the oval have been upsized so I can use an 850mAhr battery.

 

The fuselage and motor mount and CF spar adds up to 44g.  Note - I will probably brush on some resin to resist grass staining and also there are PLA hold-down steps to add.

And here is the side opening for the push-pull servo for the elevator.

 

Here is a top view.

 

 

jon
Can't tell from the pictures but I would make sure the carbon rod itself is well glued to the motor mount. The carbon will easily take the motor torque but the open top "U" section of the LW-PLA nose less so even with PLA stiffeners.
Just an observation.

That crack in the fuselage with the prior version was at the point where the CF stopped. The ~2-3g of CF needed to carry it fully forward is, for me, well worth it. It also ensures that all the bits stay in alignment during gluing.

And yes, I do have it well glued into the motor mount. And I also added a 1mm x 5mm base skid from the motor mount to the front band hold-down (see image).  I just designed and printed the aft hold-down strap.  I have now embedded the servo for the push-pull elevator and given the outside a coat of clear resin.

I did try a test session of the front open-top fuse with embedded CF rods near the top edges - but there wasn't really enough space in the cross-section to accommodate the size of CF rod that I had in stock. We shall see how well it holds up.

 

Even with the original PLA oval fuselages and round 'franken' fuse I always ran the boom from the motor mount back to the tail.  It is very good at making sure stresses are well distributed. Even in lawn-dartings the fuse held up with only minimal patching.