[Morten]

I will never get rocketry. I can't comprehend that you can control the thrust so accurately that the whole thing doesn't immediately topple and go out of control, like trying to balance a pencil on your finger and push it up towards the ceiling, and then to bring it back to earth again so it lands like the aforementioned pencil (well OK some Xmas tree stand legs came out) is just mind boggling.

Well, it's literally "rocket science".

Actually... that is a false comparison. The problem with balancing a pencil or broomstick on your finger is that your finger pushes *upwards*, not towards the CG of the broomstick. As the broomstick topples, you need to adjust your push.
The engine at the bottom of a launcher is, for most intents and purposes, fixed, and always pushing through the CG. Any small offset will remain a small offset - as opposed to your finger where any offset dynamically increases.
In reality, most engines (but by no means all) will have a very small one or two axis gimballing range and they are controlling the thrust orientation, and it is indeed rocket science, but the pencil/broomstick comparison is false.

[Charles Hunt]

But at the start of the pencil experiment the thrust line is through the CG . I take your point that as soon as the pencil moves away from the vertical, if the thrust remains vertical a destabilising moment is induced and that is why the pencil experiment goes horribly wrong.

But if that is the case why do firework rockets need a stick, where presumably the aerodynamic drag helps to keep the whole thing pointing upwards? If a rocket is as nearly stable as you suggest then the cardboard tube on its own should ascend gracefully vertically into the sky.

I imagine that doesn't happen, which is why a stick has been provided .

[Morten]

For sure, it will need guidance. Even if the thrust is perfectly aligned through the cg making ut neutrally stable, it would need to respond the external stimuli. A varying thrust vector, drag vector, moving cg, etc would all cause it to veer off course.
But it would be veering off course, not the tumbling your finger/pencil would get.

The infamous Proton failure a couple of years ago was caused by an assembly error in the guidance system, causing a reversal of the controls. The result was not a tumbling launcher but an erratic flight into the ground...

(and a testament to Russian engineering that it did not disintegrate until very late in that process. Sadly, that same engineering also managed to install sensors the wrong way up ) .

Try taking the stick off a fireworks or just ignite an Estes engine by itself and see what you get - from a distance

(apologies for polluting the thread with videos of failures instead....)

[Charles Hunt]

Ouch!!

I suppose I was thinking of videos of early V2 launches which didn't end well.

[Mike Tango]

Rather nice tracking video of a Falcon 9 landing back a day or two ago…

[Colonel Panic]

Stunning.

I also like watching the on board video of the return to the ship - trying to spot the ship as early as possible, and then seeing it land in the circle. On thing that isn't immediately obvious from the on board stuff is how far from the noisy end the titanium grids are - but this video helps.

[akg1486]

These videos of booster recovery look like something out of Blixt Gordon.

Elon turned 50 y/o the other day: he looks younger, but his list of accomplishments is pretty long.