This is slightly abstract question, with aim to determine what parameters are being missed in the hereby provided problem's description, and what are potential solutions to this problem based on various conditional assumptions.
Given: There is an infinitely long but infinitely thin threat, and it is coiled into spherical ball. Question: Would diameter $D$ of such spherical ball tend to shrink $D\to 0$, or would it remain as is $D=1$, or would it tend to expand $D\to\infty$? Please set examples of formulas that may determine the diameter of such thread ball under certain circumstances?
Offtop: This question with these exactly conditions I've came across on Facebook, where a guy by claiming that he is a mathematician, has insisted that the only solution to this problem would be by L'Hospital's Rule applied to $V=πr*2Хh$ where $r \rightarrow 0$ and $h\rightarrow\infty$, thus volume and hence diameter of the thread ball would too $\rightarrow0$. Since I could visualize in my mind that not necessarily at every circumstance this would be true, but couldn't support my predictions mathematically, I came here to seek your support and to collect some valuable counter-arguments that would clarify my doubts. Prior to asking for your help I've Googled hard to find any information on this or similar issues, but with no luck. And it is not easy to restore mathematical knowledge, once you lose practice. You guys with every argument mentioned in your replies and comments have contributed a lot to my understanding of the subject, specifically what essential parameters have been missing in the description, and how other solutions based on various assumptions may be calculated. Thank you a lot for your attention and your knowledge sharing!
Suppose the thickness of the thread is $t$ and its length is $l$. Then the volume of the thread is $\frac\pi4t^2l$. Thus, we need a radius of at least $r$, where $$ \frac{4\pi}3r^3=\frac\pi4t^2l $$ That is, the radius of the ball would need to be at least $$ r=\left(\frac3{16}t^2l\right)^{\!1/3} $$ Due to the fact that the thread would leave some enclosed space, the radius of the ball would need to be somewhat greater.
Thus, to get a meaningful, finite radius, the infinite length must be inversely proportional to the square of the infinitesimal thickness.