RAM Concept Help

Unbonded Post-tensioning Stress-Strain Curves – General Theory

RAM Concept ’s treatment of the effect of cross section strains on ultimate unbonded tendon stresses is loosely based on a paper by Naaman, Burns, French, Gable and Mattock [Naaman, A. E. et. al, "Stresses in Unbonded Prestressing Tendons at Ultimate: Recommendation", ACI Structural Journal, V. 99, No. 4, July-August 2002, pp. 518-529]. In the paper the authors, who are members of the Subcommittee of Stresses in Unbonded Tendons of Joint ASCE-ACI committee 423, Prestressed Concrete, recommend code modifications for ACI 318.

The paper provides an equation for estimating tendon stresses at ultimate bending strength of a cross section. The proposed equation is shown to have a correlation with test results that is 2.5 times better than the ACI equations 18-4 and 18-5. The equation is:

fps = fse + Ωu Ep εcu (dp/c – 1)(L1/L2) ≤ 0.80 fpu

where
fps
=
tendon stress at ultimate bending strength
fse
=
effective prestress in prestressed reinforcement
Ep
=
elastic modulus of prestressed reinforcement
εcu
=
failure strain of concrete (typically assumed as 0.003)
dp
=
distance from extreme compression fiber to centroid of prestressed reinforcement.
c
=
depth of neutral axis at ultimate strength
L
=
span under consideration
L1
=
sum of lengths of loaded spans
L2
=
total length of tendon between anchorages
Ωu
=
K(dp/L) where K = 3 for uniform or third point loadings and 1.5 for midspan loading
fpu
=
specified tensile strength of prestressed tendons

It can be shown that:

Δεp ≈ εcu (dp/c - 1)

where
Δεp
=
change in strain in concrete adjacent to the tendon from effective prestress level to ultimate bending

With this substitution (and the one for Ωu ) the equation becomes:

fps = fse + K(dp/L) Ep Δεp (L1/L2) ≤ 0.80 fpu

L can both realistically and conservatively be assumed to equal L1 as it is unlikely for two spans to simultaneously have large inelastic deformations. This simplifies the equation further to:

fps = fse + Ep (Kdp/ L2)Δεp≤ 0.80 fpu

It is obvious that in the above equations that (Kdp /L2) is a strain reduction factor that accounts for the distribution of the localized strain over the length of the tendon. The numerator is a consideration of the length of the yielding (high strain) region, while the denominator is a consideration of the length over which this strain is distributed.