Solution (source code)

= Solution

Put $T=\epsilon t$ and use the <method of multiple scales> with $x=x_0(t,T)+\epsilon x_1+\cdots$. At leading order, $x_{0,tt}+x_0=k$, so write $x_0=k-r(T)\cos(t+\phi(T))$ with $r(0)=k$, $\phi(0)=0$. The order-$\epsilon$ equation is
$$
(\partial_t^2+1)x_1=-2\partial_t\partial_Tx_0-c(1-x_0^2)\partial_tx_0.
$$
Using $\cos^2 t\sin t=(\sin t+\sin3t)/4$, its fundamental sine coefficient is $-2r_T-c r(1-k^2-r^2/4)$, and its fundamental cosine coefficient forces $\phi_T=0$. Removing these resonant forcings is the <solvability condition in the method of multiple scales>. Thus the <shifted Van der Pol amplitude evolution> is
$$
\boxed{r_T=\frac c2(k^2-1)r+\frac c8r^3,\qquad \phi_T=0}.
$$
With $a=k^2-1$ and $R=r^2$, solve $R_T=caR+cR^2/4$ by putting $W=1/R$. This gives
$$
\boxed{r^2(T)=\frac{k^2e^{caT}}{1-\dfrac{k^2}{4a}(e^{caT}-1)}}\quad(a\ne0),\qquad
\boxed{r(T)=(1-cT/4)^{-1/2}}\quad(k=1).
$$
The uniformly valid leading approximation is \b[$x(t)=k-r(\epsilon t)\cos t+O(\epsilon)$] on bounded slow-time intervals for which $r$ remains bounded. The exact initial <derivative> is recovered by the first-order correction; the displayed leading expression already satisfies the leading initial data.

There is an important limitation to the source's unqualified time-range wording. Starting from $r(0)=k$, the <amplitude> decays for $k^2<4/5$, is constant at leading order for $k^2=4/5$, and grows for $k^2>4/5$. In the latter case the denominator vanishes at
$$
T_b=\frac{\log(5-4/k^2)}{c(k^2-1)},\qquad T_b=4/c\ \text{when }k=1.
$$
A bounded-amplitude <multiple-scale expansion> is uniform for $0\le T\le C<T_b$, not through its predicted large-amplitude breakdown. For the decaying case every fixed bounded slow-time interval is available; the leading unstable equilibrium at $k^2=4/5$ also has bounded <amplitude> over such intervals. The envelope singularity is a breakdown warning, not a proof that the exact original solution has precisely this blow-up time. The coefficient in the PDF is $\epsilon c$, correcting the converted TeX's duplicated $\epsilon$.