During the building process many issues came up - like the stability of the whole structure. The durability of the joints. Any residue on and in the joints, like leftover wax in the rifle of soem joints. Also finding out in what order the framesections had to be put together. As I collect all issues in this section it can maybe happen that the one or the other problem was mentioned before - I want to apologize for this in advance.

For the prototype and the final run 3D printed joints were used and 3mm bench wood rods for the prototype. So far so good, the prototype went so smooth and showed that my intuition for the structure proved to be very reliable. It was quite light and again very stiff. Even with some pressure, that I put on it with my hands didn't broke the structure.
As I found out later, it was the natural flexibility of the wood that prevented the joints to break. And this was also one of the major differences comparing to the carbontubes. A meter of a benchwood rod and a meter of the cfk can be bend nearly for the same degree, but the cfk gets more stiff and more unwillingly to get bend the shorter it becomes, while the benchwood keeps his flexibility to a certain degree.

When I started to rebuild the motor suspension with the carbon fiber rods I quickly realised, that the handling of the cfk differs a lot from the benchwood rods. Thanks to my habbit, always to build first something up without screwing or glueing, I realized that I had to rethink the build up of the structure with the cfk. The order on how the joints and the rods were put togehter was/is crucial

The first set of joints I have printed on a Projet 3500 HD Max. It has an impressive quality and precision and also it is quite fast when it comes to print a lot of parts in the same time. For example it took just 6 to 7 hours to print about 120 joints.
The technology is based on liquid resin that is cured by UV light, and for overhangs and any support structure wax is used. This is a factor that has also to be counted in - to clean the parts - like melting the wax first and cleaning the leftovers in a ultrasonic bath also can take several hours. Unfortunately it didn't worked out perfect for me. Although I did everything as decribed some leftovers still remained - and this caused later a lot of problems within the stability of the structure. The the rods and the joints kept losing connection altough I used superglue to stick them together.

What's more there were in some joints wax left within the rifle, so when I fit them together differences in length appeared and as result the structure wasn't as stable or symmetric as supposed

I ended by printing the joints all over again on a FDM 3D printer in ABS and PLA - the only reason why I used two materials was just because of time. Of course I tried in the meantime somehow to enforce the resin printed joint by covering them by using a hot glue gun - This worked astonishingly good enough to do some test before the new joints were ready for the exchange.
To that stage it was clear that my spaceframe looked different and somehow 'beautiful' as some people said - but that was not the goal - it has to fly!

During one of the test flights a part of the arm-ring structure broke of - the joints in that section couldn't take the load peaks that build up during the start. This was the proof of what I already knew, and what's more I realized that also the structure as at the moment designed was to instable to fly.
This was an experience I made during the build-up process due the lack of any previous knowledge with multicopter. And my attitude anyway is that a failure is something good as long we can learn something from it and make it next time better. So I redesigned on the fly the structure - I made it about 30% smaller in diameter - so that it became more stable and compact what I thought would have a positive effect on the flight behavior

Density helium 0.1785 kg / m³, density air 1.225 kg / m³ i.e.. 
Buoyancy per cubic meter is equal to 1.04 kg 
(actually convert in Newton = 1.04 * 9.81)

Return to Index

Resumee

---

At the moment my allover resumee is that the weekness and the fail of the first printed joints led to misspent much time for fixing these issues instead of bringing my project to a satisfying end. But the good thing on the other side is that I learned a lot about aerodynamics in the multicopter world, the settings and the configurations that had to be done, the weak points in different electronics, basic safety handlings when working with multicopter ( Thanks Daniele) and a lot of other issues that I didn't know before because I had no experience with Multicopter in any way.
This new knowledge will be integrated as good as possible in my redesign of the structure.