They should indeed be the same, when I run your code I get the same output from both FSUM and NFORCE. The code below shows a typical usage of fsum and nforce, and they behave as they should:
[pre]
/prep7
k,
k,2,1
k,3,2
l,1,2
l,2,3
et,1,189
lesize,all,,,4
lmesh,all
DK,1,UY, , ,0,UX,UZ,ROTX,ROTY, , ,
DK,2, ,,, , ,UY,UZ
DK,3, ,,, , ,UY,UZ
F,8,FY,1000
!*
MPTEMP,,,,,,,,
MPTEMP,1,0
MPDATA,EX,1,,200e9
MPDATA,PRXY,1,,0.3 !*
SECTYPE, 1, BEAM, CTUBE, , 0
SECOFFSET, CENT
SECDATA,0.001,0.002,0,5,0,0,0,0,0,0,0,0
SECTYPE, 1, BEAM, CTUBE, , 0
SECOFFSET, CENT
SECDATA,0.001,0.002,0,5,0,0,0,0,0,0,0,0
allsel
FINISH
/SOL
solve
/post1
set,last
ETABLE,MI,SMISC,3
ETABLE,MJ,SMISC,16
PLLS,MI,MJ,1,0
spoint,8
nsel,s,loc,x,0,0.75
fsum
nforce
[/pre]
Hopefully this can give some guidance.
Cumprimentos
Peter