Details:

Creators:
Smith, Trevor ; Banas, Andrew ; Gummer, Max ; Jin, Jaesup
Smith, Trevor ; Banas, Andrew ; Gummer, Max ; Jin, Jaesup Less -
Corporate Creators:
Portland State University. Department of Civil & Environmental Engineering
Corporate Contributors:
Oregon. Dept. of Transportation. Research Section ; United States. Federal Highway Administration
Oregon. Dept. of Transportation. Research Section ; United States. Federal Highway Administration Less -
Subject/TRT Terms:
[+]
Bridge Foundations Bridge Piers Databases Driven Piles, GRLWEAP, LRFD, Resistance Factor, Databases, Calibration, Soils, DFLTD, NCHRP507, Pile Blow Counts Load Factor Piles (Supports) Resistance (Mechanics)
Publication/ Report Number:
FHWA-OR-RD-11-08
Resource Type:
Tech Report
Geographical Coverage:
Oregon
Corporate Publisher:
Oregon. Dept. of Transportation. Research Section
Abstract:
The Bridge Section of the Oregon Department of Transportation (ODOT) is responsible for the design of all bridge structures and routinely uses GRLWEAP for controlling pile driving stresses and establishing capacity from the bearing graph. The LRFD resistance factor, φ, for GRLWEAP sets the factored amount of nominal capacity in the LRFD inequality. Foundation conditions throughout Oregon’s Willamette Valley and the Portland metropolitan area are predominantly sand, silt, and clay. Steel pipe and H section foundation piles typically are of sufficient length to be friction piles and exhibit set-up after the end of initial driving (EOID). The two objectives of this study were to build an extensive database of driven pile case histories to include restrike conditions from the present available sources that reflect ODOT’s diverse soils and piles, and to establish φ factors for EOID and beginning of restrike (BOR) conditions using GRLWEAP to match ODOT practice.

A diverse group of existing databases, including the FHWA-built DFLTD and NCHRP 507 PDLT2000, were accessed and merged with new cases from the literature to build a comprehensive database, called the Full PSU Master. Neither of these two national databases proved always correct for the large amount of source input required, with the largest source of anomalies and missing data being the blow count, especially at the BOR condition. Over 150 new cases were added to establish the Full PSU Master database containing 322 piles, with each case placed into one of three input tiers for statistical profiling to assist in preserving quality for the φ calibration. The Full PSU Master database then supplied 179 cases analyzed by FHWA static capacity software DRIVEN and by GRLWEAP for capacity prediction by the bearing graph. These predictions generated bias mean λ and COV statistics for a range of ODOT selected scenarios. The 322 piles ranged up to 40 inches in diameter, and up to 197 ft in embedment length. The 179 analyzed piles ranged up to 36 inches in diameter and 167 ft in embedment length and had driving blow counts up to 100 BPI. This research showed similar trends for GRLWEAP capacity as that reported in NCHRP 507 for CAPWAP capacity on the statistical effects from variables such as blow count ranges. Sub-grouping λ by blow count revealed a clear decay in easy driving mean λ and COV parameters when blow counts were ≤ 2 BPI. Above 2 BPI, little difference was found in these parameters, and no upper limit of statistical accuracy was identified. A clear difference in statistical sample characteristics existed between piles supported in predominately cohesive soils to those in cohesionless soils, and also between pile types.

For the ODOT case of redundant piles in groups, a reliability index β at 2.33 was used to establish φ resistance factors and φ/λ efficiency measures. Statistics for an initial ten scenarios were generated, and the First Order Second Moment (FOSM) resistance factor at EOID and BOR was reported, based on lognormal fits to the λ distribution. The final five ODOT selected scenarios to permit comparison to NCHRP 507 and to form a basis to design implementation measures underwent advanced Monte Carlo based probabilistic procedures using random number generation and the λ lognormal tail fits to provide EOID and BOR φ factors. Recommended resistance factors from the visual tail fit procedure on the likely best fit to ODOT practice scenario containing all soil and pile types were 0.55 and 0.4 for EOID and BOR respectively. Recommendations were made for a separate implementation activity, including additional φ calibration work based on the Full PSU Master including use of field measured hammer performance, CAPWAP based soil input parameters, and pile type.


Format:
PDF
Funding:
SPR 683
Collection(s):
US Transportation Collection
Main Document Checksum:
[+]
urn:sha256:749355f6c053a2208fc80ac52b93ac53fed9c8167e23cbb8e8d35c73f35d1d30
Download URL:
https://rosap.ntl.bts.gov/view/dot/21758/dot_21758_DS1.pdf
File Type:

Supporting Files

• No Additional Files

More +