Rocket Science With The Other SpaceX

When you say that something’s not rocket science you mean that it’s not as hard to understand or do as it may seem. The implication is that rocket science is something which is hard and best left to the likes of SpaceX or NASA. But that’s not the hacker spirit.

Rocket science with BPS.Space[Joe Barnard] recently had an unsuccessful flight of his Falcon Heavy’s second stage and gives a very clear explanation of what went wrong using those two simple concepts along with the thrust, which in this case is just the force applied to the moment arm.

And no, you didn’t miss a big happening with SpaceX. His Falcon Heavy is a homebrew one using model rocket solid boosters. Mind you, it is a little more advanced than that as he’s implemented thrust vectoring by controlling the engine’s direction using servo motors.

And therein lies the problem. The second stage’s inertia is so small and the moment arm so short that even a small misalignment in the thrust vectoring results in a big effect on the moment arm causing the vehicle to deviate from the desired path. You can see this in the first video below. Another issue he discusses is the high drag, but we’ll leave that to the second video below which contains his explanation and some chart analysis.

So yeah, maybe rocket science is rocket science. But there’s no better way to get your feet wet then to get out there and get building.

29 thoughts on “Rocket Science With The Other SpaceX

  1. Just in case one or two of us have not discovered it, I should mention again that John Drury Clark’s book “Ignition!” gives a view of the hazards of rocket science. For liquid-fueled rockets some of the best contenders for fuels were simultaneously explosive and poisonous.

    Over in the Soviet Union the 1960 Nedelin catastrophe took out a great number of people. Rocket science isn’t brain surgery.

    1. That book was amazing!! I was really hard to put down. Some of the chemical combinations were insane. I kept wondering where they were getting that stuff from, only to be greeted with an even crazier source the next paragraph. I think that’s what kept me reading.

  2. The “rocket science” proverb came of couse into existence in the timeframe when those rocket scientists did not seem to do much more than make very spectacular and expensive fireworks.

    1. Not quite.. it was when what’s now called ‘optimal control theory’ was still new.

      Classical control theory uses negative feedback loops that self-correct, but it’s hard to make those work with systems whose dynamics change. A rocket converting mass to thrust as fast as it can, and experiencing disturbance from an atmosphere whose density changes with altitude, is a “yeah, good luck with that” problem for classical control.

      Optimal control theory assumes you can crunch numbers much faster than the system you’re controlling can respond. The control system sets its own parameters, then statistically analyzes the response to find the set of control values that produce the smallest average error. An optimal control system doesn’t just control the output, it automatically adapts to changes in the system it’s controlling.

      The language of optimal control is a mix of differential equations and statistics, and at the time, people were doing it with pencils and slide rules. It wasn’t for the casual or faint-hearted.

    1. Bearing in mind that efficient staging requires a mostly-constant ratio of mass from one stage to the next, which means that the size of the rocket grows exponentially with the number of stages, yes in principle. The numbers are rather comical if you work out the math. It gets even more impractical if you want to start from Earth’s surface, since giant mountains of crappy rocket engines don’t play nicely with drag.

      Xkcd did some math on this:

      1. In xkcd that you linked, author claims 65000 model rockets are needed to lift 60 kg to 100 km (suborbital).

        So it appears to me that around 65 would be needed to lift 60 grams to 100 km. Interesting.

        1. Unfortunately, aerodynamics don’t scale down well. I don’t have his math handy to figure out how much of that thrust was spent fighting drag, but that fraction will be larger for a smaller rocket.

  3. Hi,
    From my point of view, you should use an adaptive PID to control the flight, with different parameters, one set of parameters to be used with slow speed and a different ones for high speed. Also could be more interesting to use a fuzzy control or a combined Fuzzy-PID.

    The PID parameters for the beginning with more mass and low speed should be changed to high speed an lower mass.

    Best regards

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.