3D Printing - Geodesic Internal Structure
On more than a few occasions I have wanted to build the internal structure of my models in a similar way to the Vickers Wellington bomber. It's a beautiful structure, and exceptionally strong, spreading the applied forces in all directions.
I have seen it up close and personal as I live only a walk away from the Museum at Brooklands, the site of the old Vickers factory.
This structure would have the skin riveted on top resulting in a "stressed skin" fuselage. Any forces applied to the fuselage are spread through the skin and stringers.
In the Wellington Hanger at Brooklands is the last remaining Wellington that was built at Brooklands. In 1940 during a training sortie over Scotland the aircraft suffered engine trouble and the crew were forced to ditch into Loch Ness, all 8 crew managed to bail out, however in the rear gunners parachute failed to deploy and he sadly died.
The aircraft was located by an American team in 1974 whilst they were looking for the Loch Ness monster and salvaged in 1985. Despite the crash and 45 years underwater it remained in remarkable condition, even the tail light worked when hooked up to a battery. You can see how well the structure stood up to the dictching
So, I have started another project (which I'm keeping under wraps for now) but it gave me another opportunity to try and build the geodesic structure. At the moment, I make the walls on the model 1.2mm thick, then tell Cura to make the walls 0.4mm thick and fill the gap with a geodesic structure. Its quick, and results in an airframe that is strong enough, and light enough.
I started by shelling the fuselage to give a 0.4mm wall then, in the middle of the fuselage I sketched a grid. I spaced the lines roughly the same distance apart that I would get with 3% infill in Cura. The function used to speed things up where rectangular pattern and mirror.
I projected this pattern onto the internal surface of the fuselage in the z direction, this gives us roughly what we are looking for. I had to go back over my first attempt and put the lines slightly closer together to make sure they met up on the left and right hand sides of the4 fuselage, otherwise the structural strength would be lost.
Now, the tricky part. It is annoying that you can't use the web or rib function to get the stringers in - If you can I don't know how. So what I did was to construct a plane along path at a point near the bottom of the fuselage and on that plane sketch a 1mm by 2mm rectangle, making sure it was intersected by the sketch curve and overlapped the fuselage. Then I used the sweep function to create the ribs as new bodies. I had to do each one individually in this way, and this wasn't even the tedious bit.
I couldn't sweep the profiles as a join as at various points on the curve part of the rib would stick through the outer skin so I had to make them new bodies. So with each rib, I used the split body tool to trim away any excess, then combined the ribs with the fuselage.
The result looks great. Time to run it through Cura and see how this affects print time and component weight.
Here is a screen shot for the original pice. Double walled with 3% infill. It comes in at 72g and 6 hours 14 minutes to print. No too shabby.
Here is the fuselage with the geodesic structure and a single wall. It takes almost 4 hours longer to print and weighs roughly 20g more, an increase of 25%. I could try and make the ribs smaller to shave off a little weight but I doubt I will be abel to get a big saving on the normal piece using this method. I might print out small section to see what the strength is like but would be surprised if it was 25% stronger.
There is a potential for this method though, and especially so with LW-PLA. More investigation is needed. For now I will carry on using the double walled technique as in terms of the design process it is so much quicker.