Infinite LOOP in a summation within an MSOLVE call

[Wiljea]

This happens in "Simple Pendulum":
"Simple Pendulum", "{ "
"'(ω_(r/s))=(1_r)·√(Ⓒg/(L_cm))' "
"'(Treal_s)=2·Ⓒπ·√((L_cm)/Ⓒg)·(Σ(x;0;5;((2·x)!÷((2↑x)·x!)²)²·sin((θmax_°)÷2)↑(2·x)))' "
"'(T_s)=2·(Ⓒπ_r)/(ω_(r/s))' "
"'(ω_(r/s))=2·(Ⓒπ_r)*(f_Hz)' "

L=15_cm  θmax=80_°
@ Failing [ ω=8.08564 57173 6 r/s  Treal=0.88361 42622 96 s   T=0.77707 89775 87 s  f=1.28687 04840 1 Hz ]
@ c#27 NOT OK MSOLVER: INFINITE LOOP & "Invalid function". However Treal can be calculated alone with θmax or (θmax_°) by the following:
@ Treal='2·Ⓒπ·√((L_cm)/Ⓒg)·(Σ(x;0;5;((2·x)!÷((2↑x)·x!)²)²·sin((θmax_°)÷2)↑(2·x)))'
'ROOT(ⒺSimple Pendulum;[ω;Treal;T;f];[1_(r/s);1_s;1_s;1_Hz])'

@ Without Treal, MSOLVER works fine for the remaining 3 unknowns:
@ Expecting [ ω=8.08564 57173 6 r/s T=0.77707 89775 87 s f=1.28687 04840 1 Hz ]
'ROOT(ⒺSimple Pendulum;[ω;T;f];[1_(r/s);1_s;1_Hz])'

So, it seems that MSOLVE and SUM are somewhat incompatible even if the summation alone computes correctly in a very short amount of time, like it is the case here with the calculation of Treal.