↶ Back to AFGROW Previous Event Proceedings List

AFGROW User Workshop 2019

An AFGROW Workshop 2019 was held at the Weber State University Center in Clearfield, UT on September 10-11, 2019.

Workshop Day 1 - September 10, 2019


Welcome and Introductions (Continental Breakfast)

Current AFGROW Release Overview
James Harter, Alex Litvinov (LexTech, Inc.)

The presentation will break down new features added to AFGROW since the last AFGROW workshop:

  • New bearing and bending load case solutions for a through crack at a hole with a slot on the other side (continuing damage model)
  • New bearing load case solution for a corner crack at a hole with a slot on the other site (continuing damage model)
  • New ability to report the amount of crack growth (percentage) caused by tagged spectrum levels as well as each sub-spectrum
  • New ability to use residual stresses for advanced models with crack(s) at a hole
  • New significant improvements to the residual stress Gaussian integration routine
  • New ability for AFGROW to provide COM interface information at runtime

We will also provide information about new capabilities in upcoming AFGROW release.

A Proposed Verification Check When Developing Stress Intensity Solutions
Marcus Stanfield (SwRI)

The use of p-element FEA software allows the ability to easily obtain convergence of the energy norm by increasing the polynomial order. This verification check is necessary but not sufficient. This presentation examines common misconceptions and missteps when developing stress intensity solutions and how these affect the crack growth prediction. Additional verification checks are proposed.



K-Solutions for Countersunk Holes using an H-P Finite Element Code and DoD-HPC
Borje Andersson*, James Greer, Jr.** - *BARE, Sweden, **U.S. Air Force’s Academy Centre for Aircraft Structural Life Extension, CAStLE, U.S.A
We describe the development of databases with stress intensity data KΙ (φ) for two cracks at a countersunk hole, in plate subject to tension, bending and pin-loading, respectively, see Figure 1. Points in the (c/t,a/t,b/t,R/t,W/t)-K space are selected sufficiently close in order to avoid significant interpolation errors in fatigue analysis. The parameter space is believed to be complete, i.e. data for crack sizes ranging from extremely small to extremely large in plates of small to large widths and thicknesses are available in the created databases. Several 100M (million) virtually stress intensity functions KΙ (φ) are generated. Calculated KΙ -data are virtually exact, i.e. the relative error in KΙ (φ) at arbitrary φ = φ*, 0 < φ* < φ(a,c,b,t,D) is less than 10-3, for all solutions derived, which we guarantee by using a H-P version of the finite element method, DoD HPC resources and an advanced verification procedure.
Figure 1
Figure 1: Double Crack Scenario

NX Crack and J-Integral Evaluation
Lawrence Stoker (USAF, A-10 ASIP Analysis Group)

As part of the USAF’s Product Lifecycle Management (PLM) and digital engineering efforts, a desire to embed and analyze crack tips in the Siemens Simcenter/Nastran was expressed to Siemens. Through cooperative agreements, the Siemens development team offered to create a J-integral add-in for Simcenter 12 (previously NX Nastran) and up. After delivery, the USAF evaluated the add-in and developed rules-of-thumb for convergence using the tools already available within Simcenter. The presentation compares results against previously analyzed benchmark solutions and A-10 specific control points to other commercial software codes.


Lunch Break

Spectrum Manager Release 1.2
Matthew Gross, James Lambert (LexTech, Inc.)

This presentation will give an overview of all updates and added features scheduled to be released in Spectrum Manager v 1.2. Some of the updates include Spectrum Generation from Exceedance data, ability to save individual sub spectra, a new preview selector, improved application speed, and COM support. There will also be a discussion of upcoming features that will be added to Spectrum Manager in future versions.

Generating a Spectrum from an Exceedance Curve
James Lambert, James Harter (LexTech, Inc.)

This presentation will preview the new feature to generate a spectrum from only exceedance data in Spectrum Manager. It will cover two different methods of spectrum generation, the Incremental Pairing and Random Pairing methods, along with their respective features. It will present data in the form of R and Delta plots to convey the properties of both of the methods. Finally, it will compare the life prediction results of known spectra to results from spectra generated using the same exceedance data.

AFGROW and Spectrum Manager COM API
Alex Litvinov, James Lambert (LexTech, Inc.)

The COM abilities of AFGROW and SpectrumManager allow users to automate manual tasks, incorporate them into proprietary software, and enable the re-use of code that has been pre-built and tested.

The newly developed SpectrumManager COM interface will be presented and discussed.

The presentation will include practical examples of the COM capabilities of AFGROW and SpectrumManager. In addition, this presentation will review lessons learned during application development.



Modeling Residual Stresses with the Advanced Model Interface
James Harter (LexTech, Inc.)

A few issues were discovered in the 2-D Gaussian integration method used for the residual stress correction capability in AFGROW. An error was found in the Newton Interpolation calculations, and the part-through crack correction was not being used correctly for application to 2-D cracks. In addition, the residual stress analysis capability was not available for use with the Advanced Model Interface until release The Fawaz/Andersson corner cracked hole solutions available in the Advanced Model Interface have been shown to be more accurate than the Classic Newman-Raju solutions. Therefore, the ability to use this capability in the Advanced Model Interface would be expected to provide more accurate life predictions for cases with a known residual stress field. This presentation will review the changes made to the residual stress capability and demonstrate the use of the updated analysis capability to predict the life of one of the cold-worked corner cracked hole geometries that was part of a recent ERSI test/analysis effort.

Continuing Damage Case Study
Yan Bombardier, Guillaume Renaud (National Research Council Canada)

A common damage scenario in aircraft structures is defined by diametrical cracks (primary and secondary) growing from an offset hole in a plate. This study compared continuing damage simulation results obtained using two approaches: sequential and simultaneous crack growth. Sequential crack growth, traditionally used for aircraft damage tolerance analyses, assumes that the primary and secondary cracks are independent, whereas simultaneous crack growth includes crack interaction. Sequential crack growth simulations were performed using AFGROW and CanGROW, a crack growth research tool developed at the National Research Council of Canada. Simultaneous crack growth simulations were performed using CanGROW’s multi-site fatigue damage analysis capability. The simultaneous crack growth approach, which better represents the target damage tolerance scenario, resulted in a shorter life, longer continuing damage, and lower residual strength after ligament failure. As these results suggest that the sequential approach may be un-conservative, questions are raised regarding the applicability of the simultaneous crack growth approach with respect to the initial damage assumptions used for traditional damage tolerance analysis.

AFGROW Unknown Features
Alex Litvinov (LexTech, Inc.)



Workshop Day 2 - September 11, 2019


Continental Breakfast

Interference Bushing Durability Life Predictions
Jake Warner (USAF, A-10 ASIP Analysis Group)

Literature has long shown a fatigue life benefit from interference fit bushings and fasteners, but Air Force analytical approaches don't typically take advantage of that benefit. Accurately incorporating all of the details of an interference fit bushing into a crack growth analysis can be somewhat complex. A simplified analytical approach to fastener interference was developed and blind predictions using the method are compared with test data for validation.

Relationship of tension (axial) and bearing beta solutions for pin loaded fastener holes
Kaylon Anderson (USAF, A-10 ASIP Analysis Group)

In order to understand why AFGROW loaded hole models were producing the counter intuitive result of an increased life with increased fastener load, a study of the relationship between the bearing beta solution and the axial beta solutions was completed. Comparisons were made between StressCheck and AFGROW for various crack lengths, widths, and number of fasteners for bearing solutions, axial solutions, and the combination of bearing and axial solutions. The results of this study provide insight into the interaction between the two solutions and effective width assumptions for loaded holes.

Countersunk Hole Bearing Solutions Investigation
Robert Pilarczyk, Joshua Hodges (Hill Engineering, LLC)

Advanced K-solutions for corner cracks at countersunk holes under bearing loading are not currently available due to difficulties verifying the reference stresses used in the large solution matrix. An effort is underway to develop a smaller solution set that can be used in AFGROW for most practical geometries in the near term and allow the solutions in the larger matrix to be validated for later release.



Fatigue crack growth rate testing to characterize diagonal (L-TS) growth
Luciano Smith, James Feiger (Southwest Research Institute)

As part of a broader effort to characterize fatigue crack growth across the full crack front of corner cracks, SwRI has performed tests to determine crack growth rates in the 45° L-TS direction for 2024-T351 and 7076-T651. The tests were performed with corner cracks growing from the edge of the specimens. This geometry allowed for direct comparison to previously performed tests used to obtain L-T and L-S growth rates, as well as to previously performed L-TS through crack tests. L-TS rates were determined using two strategies. The first was to use fracture surface marker bands to determine crack lengths in the interior of the crack front after testing, and thereby calculate L-TS rates. The second was to cut specimens at a diagonal such that both crack tips were growing at the diagonal (L-TS) relative to the original plate. The two strategies, in addition to the through crack geometry from previous testing, allowed the team to compare growth rates under different types of constraint, which should aid in deciding on growth rates to use when performing multipoint crack growth analysis.

Multi-Directional Rate Data with Multi-Point Fatigue Analyses: An Overview of the Initial Approach, Findings, and Questions that Remain
Joshua Hodges, Robert Pilarczyk (Hill Engineering, LLC)

This presentation will review the recent implementations in AFGROW and BAMF to account for multi-directional fatigue crack growth rate data. The presentation will include a review of recent test data, the implementation approach, and predictions comparisons for several case study examples.

T-38 Durability And Damage Tolerance Analysis Update
Michael Wilcox (T-38 Structural Integrity & Analysis Group)

The presentation discusses a major update to the T-38 Durability And Damage Tolerance Analysis (DADTA) for the Air Force. Changes made to fatigue crack growth rate (FCGR) material models necessitated an update to the analysis. Additionally, these changes introduced the need for reevaluation of the retardation parameters associated with the Generalized Willenborg Retardation models; the retardation parameter is referred to as the Shutoff Overload Ratio (SOLR). Each Fatigue Critical Location (FCL) requires tuning for more accurate physical modeling, resulting in improved time to fracture curves for more robust structural integrity management. Utilizing the crack growth software AFGROW®, optimization methods were developed to assist with correlating analytical models of test specimens to test-observed crack growth results. The difference between the results of the analytical crack growth model and the test data is calculated and minimized in order to determine the best SOLR value for each test specimen. The resulting values of SOLR are incorporated as inputs into the DADTA along with the updated FCGR material models. The overall changes to recurring inspection intervals are mostly negative, potentially meaning an increase in inspection frequency for portions of the aircraft. The results of these investigations into enhanced understanding of the physics of material behaviors have sharpened the results of the DADTA for the T-38. This will provide more confidence with respect to the management of the T-38 fleet into the future.


Lunch Break

What's in a Filled Hole?
Robert Pilarczyk*, Joshua Hodges*, Thomas Mills** (*Hill Engineering, LLC, **APES, Inc.

This presentation will explore and discuss various methods to address the fatigue crack growth benefits resulting from a filled fastener hole. A summary of filled and open hole test data will be reviewed as well as analytical methods utilized to account for the benefit. The presentation will close with a discussion and recommendations for incorporating new capability into AFGROW to account for hole fill.This will be a discussion on the effects of open hole, neat fit, and interference fit fasteners on stress intensities. And how we can model it in AFGROW currently as well as limitations in plug-in.

Using MATLAB to Perform a Life Prediction in AFGROW version 5.3.3 or higher
James Lambert, Alex Litvinov (LexTech, Inc.)

This presentation will describe how to run AFGROW predictions within MatLab. It will go over two different methods for using AFGROW's COM interface within MatLab, depending on what version of AFGROW is available. Finally, it has a few examples that show how various different AFGROW COM features work within MatLab.

Modification of the Glinka's Weight Function for Use at a Hole
Adam Morgan (Northrop Grumman, Inc.)

Weight functions provide for a flexible approach to solving for stress intensities under complex loading but are inherently tied to a specific geometry. A disproportionate increase in the complexity of a weight function can be observed with the inclusion of additional geometric parameters compared to those weight functions with fewer geometric parameters. Motivated by this observation an elementary approach to adapting an existing weight function to more complex geometries is explored utilizing reference stress intensity solution derived from finite element modeling results. The resulting modified weight function is presented along with the results of cursory investigations into its accuracy.




BAMF Consortium Group Meeting. Current release notes, utilization cases, and BAMF wish list and task priorities

BAMF Workshop 2019
Robert Pilarczyk, Joshua Hodges (Hill Engineering)

BAMF Discussion
Kaylon Anderson (USAF, A-10 ASIP Analysis Group)

Fatigue crack growth testing and modeling of coupons with quench-induced residual stress
Renan L. Ribeiro (PhD Candidate),Professor Mike Hill (Mechanical and Aerospace Engineering, University of California, Davis)

Upper Longeron BAMF Simulation
Brian Boeke (USAF, A-10 ASIP Analysis Group)

Relifing ain't all its cracked up to be
Robert Pilarczyk, Joshua Hodges, Brent Keller (Hill Engineering)

This session will discuss the current and upcoming releases of BAMF. The discussion will cover topics including:

  • Features (users’ manual, installation packages)
  • Major events
  • Updates and bug fixes
  • Utilization cases
  • Future developments
AFGROW Future Development Discussion
Alex Litvinov, James Harter (LexTech, Inc.)

Information on the latest research and development efforts and plans beyound AFGROW Release 5.3.



↶ Back to AFGROW Previous Event Proceedings List

This site uses cookies. To disable them, click "Decline". You may adjust your cookie preferences within Cookie settings