Prakt. Met. Sonderband 41 (2009) 125 
fects were (a) Heating — (b) Cooling 
ndred free : 
00 thermal = 
/ why this 
ot only be 
cycles. - 
ays a fine 
Ti ribbons. 7% CS CRITE THR ET 
in Figures Fig. 3: DSC results from NiTi ribbons with different numbers of thermomechanical cycles: (a) heating, (b) cooling. 
by arrows) 
r of several 3.4 Tensile test 
Tensile tests of melt-spun ribbons fabricated at various wheel speeds show clearly the martensitic 
plateau even without subsequent heat treatment. The pseudoplastic deformation reaches about 4 % 
strain before plastic deformation occurs. The samples fracture at a strain of about 7 %. The results 
of the tensile test performed before and after thermomechanical training are shown in Fig. 4. The 
tensile test after training clearly shows a higher strength than the ribbons without training. This is 
caused probably by the NiTi, precipitates and a higher dislocation density. 
1000 
900 + 
800 = 
700 - 
ST 800 
Ss 500 
ung, (b) and £ - er reining 
wo] before training __ 
00 - 
7 ze | 
: 2 3 4 5 8 
constrained Strain (%) 
nsformation Fig. 4: Stress-strain curve from the melt-spun ribbon, before and after training. 
lightly. The 
je austenitic 
lated to the 
he austenite 4 Conclusions 
yws that an 
{ually to the (1) After melt-spinning the samples consist of a martensite phase and show a shape memory 
anded. It is effect without subsequent heat treatment. 
raining and (2) By using a new etchant it is possible to reveal the microstructure very clear and 
reproducible. 
(3) The bending training method is an effective method to induce a TWSME in the NiTi 
ribbons. The NiTi ribbons could be trained by a combination of constrained and free thermal 
cycles and show two-way shape memory effects sufficient for technical long-term 
applications.