The university Henry works at does have paying customers. With money they have launches on schedule. In some ways they are a company. Most of their satellites are 20cm cubes.
Here are a few lessons they have learned:
Just because the old guard does something a certain way doesn’t mean it is the right way. You can’t trim out items one by one because too often they interact together.
Making things work is hard. Good to get into the hardware soon. Good to get something simple working and then make it complicated. Don’t jump to something complicate. Learn from something basic.
Start with small. Having your first attempt fail is really, really bad for funding. It is more important to have something work that it have a lot of features.
It is really bad to have a failure which you can’t explain.
Cheap and quick is important for developing hardware. The radiation toughed electronics is bad, takes too long. Most of the radiation harden is primitive. The commercial stuff is robust. Commercial chips have more on a single chip than the radiation chips, so need fewer wiring. Henry talked about one satellite which has been in orbit for nine years with commercial parts and it is working just fine.
More important just to get something working; don’t worry about performance. Focus on making it bullet proof.
Good to design out as many death modes as you can. For example they try to make sure that even if the software goes berserk and tries to kill the hardware it wouldn’t be able to hurt the hardware.
Make sure to completely understand each failure. It could be disastrous to encounter a failure and merely put a bandaid over it.
Regulation is a long lead item. Get talking with everyone as soon as possible. For example make sure you have access to radio spectrums.
No comments:
Post a Comment