วีดีโอ

Possibly yes, but it would depend on various factors, including whether I have enough time and what aspects of regulations the course would have to cover. If you Google my name + TU Delft you should be able to find my staff page with contact details, that's easier for communication than TH-cam comments (which as you can see I don't monitor regularly)

Thank you!

Yes indeed. At the moment the state of the art is that if you want accurate fatigue predictions, you need to generate data on exactly the laminate of interest. This becomes an expensive proposition if you want to use multiple different laminates in a structure, or to use fibre steering to effectively have hundreds of different lay-ups. In my view the way forward is to develop models that can correctly capture the effect of the fibre orientations on the damage mechanisms, which is one of the things we're trying to do in the D-STANDART project (www.linkedin.com/company/d-standart/). As you say, there's plenty of interesting research left to do :)

Thanks for the useful feedback, good luck in your new role!

Indeed there are no specific regulations for hydrogen yet. It is likely those will be developed as more hydrogen powered aircraft are developed. In part 23, the regulations are already written in a more 'neutral' way so they can apply to different kinds of fuel or even non-fuel energy sources (e.g. batteries). But it's likely specific regulations for hydrogen will be needed. For the first hydrogen aircraft it's possible regulators will use the mechanism called 'special conditions', which you can think of as specific additional regulations to deal with certain novel features that are not covered by the normal regulations.

@@drj.a.pascoe2606 Thank you for your response Sir.

Thanks for the kind words. Indeed, lots of complexity to understand, but that's what makes it fun, right? :)

To be honest, I have no idea. Considering to my knowledge no electric helicopters are in service yet, it would likely take several years at least, and I wouldn't be surprised if the final cost would be in the millions.

There definitely are! In principle type certification is handled at a national level, so all countries that have aircraft manufacturing industries will have agencies handling type certificates. So for civil airliners (part 25 in FAA / EASA language) that includes Brazil, Canada, Russia, and China for example. What often happens in practice is that the 'local' authority (FAA in case of Boeing, EASA in case of Airbus, and so on) will take the lead in the certification process and other agencies will take more of an observing role, maybe adding one or two specific points. Then when the 'lead' agency issues the type certificate, the agencies from other countries will accept that and only have a very short process to grant type certification in their own jurisdiction.

That's great to hear! Good luck with your homework!

Glad I could help. Good luck with the essay!

Good point. The simulations tools will of course need to be validated to show that they can produce reliable predictions of part performance (i.e. strength, stiffness, etc, depending on part and load cases). The key thing is that the simulations should not take drawings / CAD data as input, but data from the actual as built part. So you would need to show that your simulations can accurately predict the effect of defects in the part, and that they can accurately simulate a part with a internal microstructure / defect distribution that has not specifically been tested. My feeling is that this effect of defect simulation may still need work before we can rely on it fully. Indeed the idea is that once the simulations are validated (which will of course require testing), then no further experiments are needed. For example at the moment if you have a 3D printed part, and you want to print it on a different location on your build plate, you need to redo all the qualification testing. The same if you want to make that part on a new machine (even if it's the same model and type of machine). I believe it should be possible to account for those kind of differences purely through a combination of process monitoring and part simulation.

@@drj.a.pascoe2606 in my opinion 3D printed parts have the (very) big advantage they are not operator depende t as most of the traditional long fiber composites are (excluding processes like ATP,ATL, etc. ). Of course to be still "machine-dependent" doesn't solve the problem, but a sort of standardization of manufacturing tools would probably be a way to easily comply with certification specifications. But all this would hold for for 3D printed components. How to transfer the approach to the evaluation of performance of traditional composites ? Would, in this case, a microscale approach be practically possible and economically convenient to allow that design freedom that is the real strenght of composites ?

​@@CP-zi3eg Unfortunately, to my understanding, standardization of manufacturing tools for additive manufacturing is still very tricky. Machine-to-machine variability is still so large that a part qualification is only valid for a specific machine. So currently you need to redo each part qualification for each new machine you install. For traditional composites I don't think you'd have to go all the way down to the micro-structure; but if you're using e.g. ATL / AFP you do need to check whether your actual fibre paths are deviating from the design, and what is the effect of any gaps or overlaps. In the current approach you would probably do that statistically (so have a knockdown factor based on the assumed presence of a defect size that will occur with a certain probability), but I think it is possible to replace that with knowledge of the *actual* defects in a part, and then modelling the effect on part performance. That said, if you already have a composite part that is certified, it's probably not worthwhile to change to a simulation based acceptance approach. For traditional composites, the current approach is to fully qualify a limited 'menu' of standard lay-ups, and then only allow those lay-ups in the structure. That is an approach that works for certification, but it limits your design freedom to fully optimise a structure. For a fully optimised structure you may wish to continuously vary your fibre angles, and effectively have hundreds or thousands of different lay-ups throughout your part (because at each location you want different combinations of fibre angles and ply stackings). In that case it's not cost effective to first qualify each individual lay-up, which is why a new certification approach is needed.

Thanks, glad you enjoyed it!

Thanks for the kind words!