(a) We need the smallest positive solution of sint=53. On [0,2π] the sine increases from 0 to 1, so there is exactly one solution there and it is the principal value:
t=sin−153≈0.6435 s.
(b) Sine takes the same value again in the second quadrant, at the reflection of the first solution in t=2π:
t=π−sin−153≈3.1416−0.6435=2.4981 s.
A check: sin(2.4981)=0.6000. Note that this second time is not something the sin−1 key can return — it lies outside [−2π,2π] — which is precisely the limitation the passage describes.
(c) Since 2π<4<23π, the angle 4 is outside the principal branch, and the reduction to use is
sin−1(sinx)=π−xfor 2π≤x≤23π.
Hence
sin−1(sin4)=π−4≈−0.8584.
Verification: sin4=−0.7568 and sin(π−4)=sin(−0.8584)=−0.7568 — the same displacement, a different time. The output is short of the true time by
4−(π−4)=8−π≈4.8584 s,
and it is even negative, which no elapsed time can be. The key has not made an arithmetic error: it has returned the one angle in its own branch with the measured sine.
(d) sin−1(sint)=t holds exactly when t already lies in the branch [−2π,2π]. Intersecting with [0,2π]:
t∈[0,2π].
So the unit can be trusted only during the first quarter-cycle, about the first 1.57 s. At t=43π, for instance, it would report 4π; at t=2π it would report 0.
(e) Every solution comes from one of the two families found in (a) and (b), repeated each period 2π:
t=2nπ+sin−153ort=(2n+1)π−sin−153,n=0,1,2,…
Numerically the first few times are 0.6435, 2.4981, 6.9267, 8.7813,… seconds. The measurement s=53 therefore never determines t by itself; the sin−1 key supplies one representative, and the periodicity supplies the rest.