Strain gage in a bottle

A representative composite overwrapped pressure vessel. Image courtesy of CompositesWorld: https://www.compositesworld.com/articles/thermoplastic-composite-pressure-vessels-for-fcvs

We’re putting something in a bottle, and no, it’s not an SOS to the world. It’s a strain gage!

Hill Engineering has recently developed technology to orient and apply strain gages inside a pressure vessel with restricted interior access.

We were first approached with the challenge by NASA, who sought to perform testing on composite overwrapped pressure vessels. COPVs, as they’re known for short, are vessels constructed of a thin metallic liner wrapped in a fiber composite, designed to hold fluids under high-pressure. These are often used to store fuel during spaceflight, due to their high strength and low weight.

During manufacturing and use, COPVs undergo intense environmental conditions, which makes studying their performance challenging. To further complicate things the metallic liner, which is the inner layer, is only accessible through a small diameter hole at one or both ends.

Presented with this issue, our development team had their work cut out for them: how to position a correctly oriented strain gage inside the COPV and apply the pressure needed to adhere it to the vessel wall, all while retaining access to the gage wires in order to collect readings during measurements.

NASA provided a full-scale test COPV, which we sectioned in order to track the effectiveness of our process. The goal was to apply a strain gage five feet from the access point through an approximately 1-inch diameter inlet port. After months of design and procedure iterations, our team developed the custom tooling necessary to achieve the desired results. The procedure allows us to correctly apply a gage in only a couple hours and the portability of our tooling allows us to perform the method in nearly any environment.

A simplified diagram of the tooling developed for the strain gaging procedure.

Since its initial installation, we’ve utilized this technique for numerous “strain gage in a bottle” applications, further diversifying our capabilities. The results have been near-perfect, much to the satisfaction of our customers!

If you have more questions about this particular technique, how it can help your project, or about any of our residual stress measurement techniques, feel free to contact us.

For additional information on this particular application and how it benefitted NASA’s program, read their article on COPV Test and Measurement Techniques.

Residual Stress 101 vlog

Today, we’ve released the newest episode of our vlog: Residual Stress 101. The video is a return to basics, discussing the core of what it is we do here at Hill Engineering.

If you haven’t checked out our YouTube channel, it might be time. Our mission is to post content that helps highlight the capabilities of our organization, so that everyone can see how and why residual stress is important to their manufacturing processes.

Today’s post is a broad overview of what residual stress is, including the several techniques for measuring residual stress found in our lab. Look for future content that delves further into each technique, and contact us if you have any further questions or want to see a video related to something we haven’t discussed.  

Case Study: cold hole expansion process modeling

Cold hole expansion (Cx or cold expansion for short) of fastener holes is a technique that has been widely used in the aircraft industry to improve the fatigue performance of structural components. The cold hole expansion process introduces compressive residual stress near the vicinity of the hole that slows crack growth and can significantly improve fatigue performance. Continue reading Case Study: cold hole expansion process modeling

Additive Manufacturing Benchmark Publication

Hill Engineering recently contributed to a publication related to residual stress measurement in additive manufacturing (AM) test specimens titled, Elastic Residual Strain and Stress Measurements and Corresponding Part Deflections of 3D Additive Manufacturing Builds of IN625 AM‑Bench Artifacts Using Neutron Diffraction, Synchrotron X‑Ray Diffraction, and Contour Method. The work was performed under the NIST AM-Bench program in collaboration with researchers from NIST, Los Alamos National Laboratory, University of California Davis, and Cornell High Energy Synchrotron Source. The abstract text is available here along with a link to the publication. Continue reading Additive Manufacturing Benchmark Publication

ASIP Conference 2019

Hill Engineering is presenting about residual stress aerospace forgings at the upcoming 2019 United States Air Force Structural Integrity Program Conference (ASIP) in San Antonio, TX. The 2019 ASIP Conference is specifically designed to bring together the world leaders in the area of aircraft structural integrity and to disseminate information on state-of-the-art technologies for aircraft structures in both the military and civilian fleets. Hill Engineering’s presentation will include a summary of recent work to quantify the residual stress variability in aerospace forgings. The abstract text is presented below. Continue reading ASIP Conference 2019

Engineered Residual Stress Implementation workshop

Hill Engineering is proud to support the USAF and their objective to advance damage tolerance analysis methods through the Engineered Residual Stress Implementation (ERSI) workshop. At this year’s ERSI meeting (September 12-13), Hill Engineering will meet with other stakeholders in the USAF aircraft community to review progress over the past year towards implementation of engineered residual stress in the USAF fleet. Continue reading Engineered Residual Stress Implementation workshop

Machining distortion modeling

Part distortion during machining is a significant problem in many industries, particularly where rigorous dimensional tolerances are required. Distortion of finished parts can lead to significant economic loss and should be managed for effective design and production. This case study demonstrates some of the basic concepts related to the impact of residual stress on part distortion during machining. A representative problem is defined, and a model is used to estimate part distortion due to machining of raw material containing bulk residual stress.

This study considers a 304.8 x 203.2 x 12.7 mm (12.0 x 8.0 x 0.5 inch) aluminum plate as the starting raw material for the analysis. From the plate an example part will be machined that has the same in-plane dimensions as the starting plate (304.8 mm x 203.2 mm) and includes a 2.54 mm (0.1 inch) thick frame around a 2.54 mm (0.1 inch) thick web.


Raw plate and final part geometry. Non-bracketed dimensions are in inches and bracketed dimensions are in mm.

Aluminum plate is often stress relieved by stretching, and typically exhibits low levels of residual stress post-stress relief. For the sake of this analysis, the raw material is assumed to have the residual stress distribution shown in Figure 2a (similar to the residual stress measured by Prime and Hill [1]). The residual stress values are low compared to the yield strength of the material, ranging from about -20 to 20 MPa (-3 to 3 ksi).

In addition to the bulk residual stress present in the raw material, the machining process also induces stress. The machining-induced residual stress assumed for this demonstration is shown in Figure 2b, and exhibits a typical distribution with compressive residual stress near the machined surface that spans over a thin layer (0.2 mm) before it reaches magnitudes near zero. The peak compressive residual stress at the machined surface is -50 MPa (~ 7.3 ksi). The bulk residual stress in Figure 2a is assumed to be present in the raw plate for the analysis, while the machining-induced residual stress in Figure 2b is applied locally to the machined surfaces.


(a) Bulk residual stress (similar to [1] along rolling and transverse direction, (b) idealized machining induced residual stress

A finite element model including the bulk and machining-induced residual stresses was used to predict the distortion of the finished part. The model is elastic and superposes bulk and machining residual stress to provide an equilibrium solution. Figure 3a shows the deformed shape (using a magnification factor of 30 to better illustrate the deformation). The displacement pattern shows bowing of the finished part with respect to its intended shape, with positive displacements near the center. A 2D map of the displacement of the bottom surface of the finished part is shown in Figure 3b. Line plots along the x direction at y = 101.6 mm and along the y direction at x = 152.4 mm are shown in Figure 3c. The distortion range is approximately 1.4 mm. It is important to note that even though the bulk residual stress in the raw material is low (about 5% of the yield strength), it still has potential to cause significant distortion in finished parts, as illustrated here.


(a) Undeformed/deformed 3D shape of final part with zoffset = 0, (b) 2D map of leveled displacement of bottom surface, (c) line plots along paths from left-right and bottom-top

Since the raw plate is thicker than the final part, the final part can be extracted from different positions through the thickness of the raw plate (e.g., Figure 4). The position from within the raw plate that the final part is removed from can have a significant impact on the distortion (due to the different bulk residual stress levels at different locations through the thickness). The position is defined by an offset distance from the bottom surface of the raw plate, zoffset. In the first example, the zoffset = 0, i.e., the bottom surface of the final part is aligned with the bottom surface of the raw plate (z = 0).


Location of machining of baseline/final part within the raw plate

The model used here can be modified to consider different part placements within the raw material in a straightforward manner. A significantly different result was obtained considering zoffset = 2.54 mm (0.1inch), which is shown in Figure 5. An opposite pattern of distortion is observed in Figure 5a compared to the case where zoffset = 0 (Figure 3a). The 2D map shown in Figure 5b shows displacements that range from 1.1 mm to -0.6 mm. Figure 5c shows the displacement along the left-right and bottom-top paths, and includes the results obtained with zoffset = 0 for comparison. Compared to zoffset = 0, zoffset = 2.54 mm exhibits displacement along the x direction that ranges from positive-negative-positive values and with higher magnitudes. The displacement along the y direction is similar for both offsets, but have opposite signs.


(a) Undeformed/deformed 3D shape of final part with zoffset = 2.54 mm (0.1inch), (b) 2D map of leveled displacement of bottom surface, (c) line plots along paths from left-right and bottom-top comparing zoffset = 0 and 2.54 mm

Another aspect that influences the part distortion is the thickness of the web of the large pocket. The previous results considered a thickness of 2.54 mm (0.1inch), as illustrated in the final part drawing in Figure 1. Reducing the thickness to 0.635 mm (0.025inch) and considering the zoffset = 0 configuration causes significant changes in the results, as observed in Figure 6. A similar pattern of distortion is observed in Figure 6a and Figure 6b compared to Figure 3a and Figure 3b, however the magnitudes of displacement are significantly lower. A line plot comparing the results obtained with both thicknesses is shown in Figure 6c. Overall, the model with reduced thickness (red lines) provides lower displacement magnitudes along both paths (left-right and bottom-top) compared to the initial model with 2.54 mm thickness, and exhibits peak displacement that is lower by about 50%.


(a) Undeformed/deformed 3D shape of final part with zoffset = 0, (b) 2D map of leveled displacement of bottom surface, (c) line plots along paths from left-right and bottom-top comparing thickness = 0.635 mm (0.025inch) and 2.54 mm (0.1inch)

This case study provided an example problem for the estimation of part distortion due to residual stress release from machining, considering a typical bulk residual stress distribution and machining-induced residual stress distribution. The results show significant part distortion, even though the considered bulk residual stress had very low magnitude compared to the yield strength of the material. The results also show that part distortion varies significantly depending on the machining location within the raw stock material.

For more information concerning this case study or any of the residual stress measurement techniques employed at Hill Engineering, feel free to contact us.

[1] M. B. Prime and M. R. Hill, “Residual stress, stress relief, and inhomogeneity in aluminum plate,” Scripta Materialia, pp. 77-82, 2002.

Hill Engineering introduces ExpressRS

For materials engineers, designers, and managers seeking residual stress measurements, Hill Engineering is a trusted source for a broad range of best-in-class measurement capabilities. But while we always strive to deliver quality results in a timely manner, sometimes a job requires a faster than normal turn-around. This is why we’ve introduced ExpressRSTM, a service geared toward expedited delivery of residual stress measurement results. Continue reading Hill Engineering introduces ExpressRS

Residual stress biaxial mapping validation

Hill Engineering recently published new research detailing our efforts to validate the PSR biaxial mapping technique for residual stress measurement.

This new technique generates two-dimensional maps of additional residual stress components over the same plane as the original contour method measurement. The paper is titled Assessment of Primary Slice Release Residual Stress Mapping in a Range of Specimen Types and appears in the November 2018 volume of Experimental Mechanics. Continue reading Residual stress biaxial mapping validation

2019 SEM Annual Conference and Exposition on Experimental and Applied Mechanics

Hill Engineering will be presenting at the upcoming SEM Annual Conference and Exposition on Experimental and Applied Mechanics in Reno, NV from June 3rd through June 6th. We invite you to come see us. This conference focuses on all areas of research and applications pertaining to experimental mechanics, and has evolved to encompass the latest technologies supporting:

  • optical methods
  • additive & advanced manufacturing
  • dynamic behavior of materials
  • biological systems
  • micro-and nano mechanics
  • fatigue and fracture
  • composite and multifunctional materials
  • residual stress
  • inverse problem methodologies
  • thermomechanics
  • time dependent materials.

Hill Engineering’s presentation will include a summary of recent work related to regularization uncertainty in slitting residual stress measurement. The abstract text is presented below.

This presentation describes the development of an uncertainty estimate for slitting residual stress measurement. The uncertainty estimate includes a newly developed uncertainty estimate related to the smoothing used in the stress calculation procedure called the regularization uncertainty. This work describes the approach to define the regularization uncertainty, shows the usefulness of the uncertainty estimate in a numerical experiment and a repeatability study. The uncertainty estimate is shown to meet an acceptance criterion that compares the calculated (measured) stress ± its uncertainty estimate to the true value for the numerical experiment or the mean stress from the repeatability study. This works shows the regularization uncertainty estimate to be a necessary contributor to the uncertainty in slitting and additionally the uncertainty estimate developed here reasonably predicts the uncertainty present in slitting method residual stress measurements.

If you are planning to attend the conference please stop by to discuss Hill Engineering’s capabilities in fatigue analysis and design and residual stress measurement . Please contact us for more information.