Texas Instruments TI-92+, Tip List

Page: 406 / 507

9.2348545475962 ⁻ 5, ⁻ .0036542400520554,.075999519225692},ts)
Endfunc
While it may appear complicated, the logic is quite simple. We want to solve the function vfts120() for
the input argument vf. To speed the solution, we first find an approximate solution for t, called tapx.
tapx is in turn the center of a desired solution range tapx - 0.4 to tapx + 0.4. Finally, we don't want to
search for solutions beyond t = 705.47, so that constraint is included, too. As a test case, we try
vfts120(100.17) = 0.01613053407. In this case the contraints are about 99.789 and 100.589, so the
desired root of 100.17 is included in the bounds. As shown above, solvebug() returns about 63.673,
which is a root, but outside of the constraints. Note that the constraint conditional expressions are
ordered correctly: the independent variable appears first.
If we remove t from the Local list, then the correct root of 100.17 is returned, but a Questionable
accuracy warning is shown. Regardless of whether or not t is declared local, the correct root is quickly
returned without the warning message, if we use tapx as an initial guess for t, like this:
nSolve(vfts120(t)=vf,t=tapx)|t ≥ tapx-.4 and t ≤ min({tapx+.4,705.47})
So, it appears that the optimum fix for this problem is to declare the independent variable as Local, and
also use an initial guess for the independent variable.
11 - 13