PGSplice - Girder Details and Losses

2 posts / 0 new
Last post
PGSplice - Girder Details and Losses


I am new to using PGSplice so had a few questions as I work through my design and try and compare the PGSplice output to my hand calcs to make sure I understand how it works.

1. Girder Details
I am using 14 Grade 270 Low Relaxation 0.6” strands (Area = 0.217; fpu = 270ksi). So I calculate the Pjack force as 0.75*270*0.217*14 = 615 kips. The Pjack force in PG Splice is 664 kips. Do you know if PGSplice uses a different strand area or stress limit on fpu? Just trying to understand the difference.

2. Analysis Results Graphs
Pre Tensioning, Post Tensioning and Secondary Effects – Are these stresses, moments etc from prestressing and post-tensioning after instantaneous losses (friction, anchor set etc.)? These don’t vary with time in my model. Are all the time dependent losses included/added to the other long term behavior’s in Creep, Shrinkage and Relaxation? I see the calculation for the time dependent losses under the Details Report -> Prestress Loss Details but I'm not sure where they are included in the graphs.

3. Anchor Set Losses
How are anchor set losses calculated? Is this using Equation 5.1.4-1 of the WSDOT Bridge Manual. In the model it calculates the max loss as 23 ksi (average of three ducts from Details Report -> Prestress Loss Details) but I calculate 17 ksi using Equation 5.1.4-1 or 14ksi if I used the equivalent Caltrans equation (my brodge is for Caltrans). Not sure if there is a mistake in my calculations or the program is doing something more sophisticated. I get pretty much the exact same friction loss as PGSplice (about 24ksi) so I don’t think this is the difference if it is using Equation 5.1.4-1. I looked in the help but couldnt see any commentary on how anchor set losses are calculated.

4. Girder Definition
I am using a California WF PT girder. I defined the shape in the girder library using the girder dimensions (C1, D1, D2 etc). As parts are curved with fillets I can get close to the section properties but not the exact values. Is there a way to overwrite the section properties (A, I, S) to the exact values. It's not that big a difference but just wanted to check.

Any advice would be really appreciated.



Rick Brice

1. Girder Details
In your Project Criteria, you limit the stress in tendons to be 0.9fpy before seating. This is consistent with LRFD Table
Pj = 0.9(0.9)(270)(.217)(14)=664

2. Analysis Results Graph
PS, PT, and Secondary effects are all after force transfer. This graph shows force effects in the girder. The prestressing has an initial effect on the girder and then incremental creep, shrinkage, and relaxation effects account for changes over time. If you want to see how the force in the strands and tendons change over time, look at the "Effective Prestress" graph. This graph shows how pretension forces change along the length of segments and the girder and through time.

3. Anchor Set Losses
Anchor set losses are iteratively computed using the fundamental equation DeltaA = (1/EA)*Integral(Px-Pa)dx where
E = Modulus of elasticity of tendon
A = Tendon area
Integral = This is the integral symbol, but since all we have is text in this post I've spelled it ou
Px = force in tendon at distance "x" from the stressing end
Pa = force in tendon under jacking stress at the assumed anchor set distance "a"
dx = incremental distance along the tendon length
DeltaA = elogation associated with the assumed anchor set influence length "a"

Guess the anchor set distance "a", compute Pa, solve the equation for DeltaA, compare with target anchor set. Iterate "a" until the computed and target anchor set are equal.

4. Girder Definition
The version of PGSplice from WSDOT does not have the Caltrans girder section built in. The best you can do is approximate the section with the I-Girder or the NU Girder. Section properties are always computed from the girder shape. PGSplice Professional might have a native Caltrans girder shape.

Log in or register to post comments