stress calculations

6 posts / 0 new
Last post
trj
stress calculations
AttachmentSize
Image icon Table Snapshots339.95 KB
File PGSplice Model171.83 KB
Image icon Math Error15.17 KB
Image icon Stressing Error93.08 KB

I am attempting to mimic the stress calculations reported by the Spec Check Report so I can manually check closure joints and locations over the pier that have variable x-sections. I want to make sure I am pulling the correct information (stressing, loading, etc) from PGSplice so all I have to do is apply my new section properties and I am having a hard time matching things up.

As an example, I've attached screenshots (and spl file) showing Service I moment, Transformed Section Properties, and the Spec Check tables for Interval 16 - Deck Placement for a random diaphragm location. Using M & Sb from first two tables;
fb = M / Sb = 5649.01 (12) / 34,380.5 = 1.972ksi
However, Service I fb from Spec Check is 2.833ksi

I also am failing to match post-tensioning stresses using Tendon Geometry and losses but figure if I can get help figuring out what I'm missing with just Service I checks the P/T issue might be figured out as well.

As a side note, when I run the Bridge Analysis Report I get "Table Error - class rptRCTable" to pop up 5 times, always with what seems the same timing each time the model runs. Is this an error in PGSplice? My install version? My file?

Thanks!
-Taylor

Rick Brice
stress calculations

You can't compute stress at any interval as M/S. You have to sum up incremental stresses through time to account for changing section properties. Stresses are computed with transformed section properties so the modular ratio between concrete and steel changes with every analysis interval (Ec is a function of time). The girder cross section also changes from a non-composite girder to a composite section made of girder and deck concrete which also have different Ec.

Thanks for the table error information. This means we've got something programmed wrong when building the output table. No information is lost, but there may be addition information in the right hand column. I'll work to get this fixed for the next release.

Rick Brice
stress calculations

Following up on your original question...
1) You can avoid the table error messages by either omitting the load rating responses from the Bridge Analysis Report (you don't have any load rating types enabled) or enable load rating options. The error message is occurring because the report is trying to included load rating information but all load rating types have been disabled.

2) Stress calculations. Stress is computed as the summation of all incremental stress effects. Incremental effects come from externally applied loads, changes in support conditions, and internal restraining forces between dissimilar materials (concrete and steel) due to time dependent effects (creep, shrinkage, and relaxation). The Time Step Details report gives a full accounting of incremental stresses and strains.

trj
stress calculations

Appreciate the explanation and help!
I've started looking in to the Time Step Details and have found a couple things that again, seem odd. Pictures have been uploaded to the original post.

1 - In calculating the component restraint forces there appears to be a simple math error. I can't quite tell if it's actually using an average time step strain, modulus, or both; but the plainly listed formula of e*A*E calculates something different than what is shown in the table (see "math error" snapshot).

2 - Even though I was attempting to place my p/t in 2 stages, it is all being applied in one. I played around with this a lot, even adding additional tendons and no matter what I did, when the p/t is actually applied does not directly correlate to the Timeline Manager or Girder Details dialogs. Essentially what I found was "Duct 1" in the Timeline Manager stresses Tendons 1 thru 4, "Duct 2" stresses Tendons 5 & 6, & "Duct 3" doesn't stress anything in the second stage (see "stressing error" snapshot).

Rick Brice
stress calculations

Thanks for pointing these issues out. Regarding #1, it is actually a reporting error and not a math error. The age adjusted modulus is being reported in the calculation, however, the actual modulus is used in the calculation.

Regarding #2 - There are some issues in the program with modeling U-Beam tendons.

We are working to resolve both issues and should have a hot fix release soon.

Rick Brice
stress calculations

These issues have been resolved. Please download and use BridgeLink::PGSplice version 4.1.3.

Log in or register to post comments