There's a great video on repairing a severely damaged tuba, which includes the use of the magnet-ball for straightening the walls of the tubes, as well as disassembling parts of the tuba to gain access to individual parts:

https://www.youtube.com/watch?v=wflildRmcJw

The business of forming pipes with curves typically involves mandrels. A mandrel, in this case, is a solid body that prevents the pipe from deforming in undesirable ways.

https://en.wikipedia.org/wiki/Mandrel

Musical instrument repair is clearly more complex than this—it's a skilled, improvisational profession that would probably welcome this kind of trick. I feel that, outside of Mega City One, those specialists just would not exist, their skills wouldn't be sustainable as a source of employment. That's the economic side of this—it's cool to come up with new techniques for instrument repair, but try being an instrument repairer in a city on the periphery. It's scary that this kind of (fairly mild) specialization is so precarious.

Speaking of hammering from the inside: that's how you make "jumping beans" using a steel ball wrapped in foil:

https://www.youtube.com/watch?v=Y9sl1jLlekk

Made lots of these as a kid.

https://www.youtube.com/watch?v=3BwIOSpD6ro

I am so very disappointed that this article was not about uncoiling the instruments.

This is something that I too want to see.

In my search I found this[0] fascinating page on the physics of brass instruments. But this [1] is the best image I could find.

0: http://newt.phys.unsw.edu.au/jw/brassacoustics.html 1: http://newt.phys.unsw.edu.au/jw/brass/graphics/lipreedbores....

This seems very similar to a lot of "paintless" dent repair for automobiles - basically figuring out a way to apply force from the inside to gently smooth out small dents (often with the help of looking at gridlines in the reflection).

I am unconvinced by the 'reverse hammer's method.

Does it rely on eddy currents in the steel ball? Wouldn't copper work much better in that case? Will the ball be large enough for eddy currents to be big enough to have much effect?

To me it seems to be operating like a slide hammer that connects via magnetic force instead of mechanically.

https://en.wikipedia.org/wiki/Slide_hammer

The ball sticks to the magnet across the instrument wall. The forces are balanced: the magnet is pushing on the wall from one side and the ball from the other.

Now when the stop bar is struck by the mass, the magnet is violently moved away in a short amount of time. For a brief moment, the force is unbalanced, because the magnet has lost contact with the wall; only the ball is pushing on it.

The magnetic attraction may not in and of itself be strong, but the impulse is.

In physics, FΔt is called impulse. The impulse applied to an object changes its momentum: FΔt = Δmv.

If we keep the Δmv the same, but decrease Δt, we can raise the F.