Puzzle: A role of toilet paper has $180$ sheets on it. The outside is covered with exactly two sheets. The inside around the cardboard cylinder is covered by exactly one. Question of the puzzle: how many layers of toilet paper are on the roll of toilet paper?
The "Given" Solution: One way to solve this is by saying that the average round is covered by $1.5$ sheets, so therefore the answer is $120$ (I have no source for whether this is actually correct)
I Tried: I tried to solve it with a differential equation, but ultimately failed: Let $S$ be the number of toilet paper sheets on the roll, and $n$ the number of rotations.
I think the number of sheets per rotation depends linearly on the number of rotations at a given point, because with every rotation the toilet role becomes more thick, so: $$\frac{\mathrm{d}S}{\mathrm{d}n}=kn.$$
This equation is separable, so $$dS=(kn)\,\mathrm{d}n.$$
Integrate to get $$S=\frac{1}{2}kn^2+C.$$
Now we need to find the values of constants $k$ and $C$: we know that when $S=1$ then $n=1$ and also when $S=180$ then $\frac{dS}{dn}=2$.
But now I am stuck.
My Question: What would be the correct way of solving this problem using differential equations?
This question is also linked to this question: Toilet paper puzzle (question 2)
Use the "sheets" as units of length, so the total length is $L=180$, the "inner" radius is $r_i = 1/(2\pi)$, the "external" one is $r_e = 2/(2\pi)$. Assume that the paper draws a spiral and makes $n$ loops (not necessarily an integer). Then, we parametrize this spiral in terms of the angle $t \in [0, 2 n \pi]$ (I use rad for angles, so that $n$ is the number of layers). The radius is slowly increasing from $r_i$ to $r_e$, say linearly: $$ r(t) = r_i +(r_e-r_i) \frac{t}{2 n \pi} $$ We have $$ L = \int_0^{2 \pi n } dt \sqrt{ r(t)^2 + r'(t)^2} \approx \int_0^{2 \pi n } dt \, r(t) = n \pi (r_e + r_i) $$ This gives that the number of loops (or layers) is $$ n \approx \frac{L}{\pi (r_e + r_i)} = 2L/3 = 120 $$ Note: the approximation is valid if the spiral is "spiralling out" slowly. Otherwise the line of reasoning is the same, but you have to deal with more complex calculations (i.e. taking the full square root term in the integral).
PS: it is not explicitly with differential equations, but there is an integral. Solving the integral above is equivalent to "integrate" the ODE $$ \frac{dL(n)}{dn} \approx 2 \pi \, \frac{r_e + r_i}{2} $$ You have to integrate this simple ODE with the initial condition $L(0) = 0$ till you touch the value $L(n) =180$.