Compressive Stress Relaxation (CSR) is a means of estimating the service life of a rubber seal over an extended period of time. As such, it can be thought of as the big brother of compression set testing. Rather than measuring the permanent loss of thickness of a compressed rubber specimen as is done in the compression set, CSR testing directly measures the load force generated by a compressed specimen and how it drops over time. In part 1 of our blog series, we will explore the theory of CSR testing, common test methods, and how CSR differs from compression set testing.
To understand the value of CSR testing and how it differs from compression set testing, it is helpful to return to the basic theory of how a rubber seal functions. In a standard compressed seal design, a rubber seal is deformed between two parallel surfaces to roughly 75% of its original thickness. Because the material is elastic in nature, the seal pushes back against the mating surfaces, and this contact force prevents fluid flow past the seal, thus achieving a leak-free joint. Over time, the material will slowly (or perhaps not so slowly) relax. The amount of force with which the seal pushes against the mating surfaces will drop, and the seal will become permanently deformed into the compressed shape. In compression set testing, the residual thickness of the specimen is measured, and it is assumed that this residual thickness is valid proxy for the amount of residual load force generated by the compressed seal. In CSR testing, the residual load force is measured directly.
In practice, CSR results are typically presented very differently from compression set results. In CSR testing, it is common to see multiple time intervals over a long period of time (3,000 hours or more of testing), thus allowing a curve to be created (see Figure 1). In practice, however, specifications are written such that only the final data point has pass/fail limits. In compression set testing, it is common to see a single data point requirement with a single pass/fail limit. Multiple compression set tests can be performed to create a curve, but this is almost always done for research purposes rather than for specification requirements. In most cases, compounds that excel in compression set resistance also demonstrate good retention of compressive load force over time. However, there are exceptions.
Figure 1: Typical CSR curve.
These results display a fluorocarbon seal material immersed in engine oil at 150°C.
CSR testing can be quite complicated, and caution is needed when comparing reports to ensure a valid apples-to-apples comparison can be made. As with the compression set, CSR testing can be performed in air or immersed in a fluid. Because most seal materials will oxidize in the presence of hot air, the results for CSR in the air can be strikingly different (worse) than the results for CSR in a fluid at the same temperature. In addition to time and temperature, the sample size (usually a button 12.7 mm in diameter and 6.35 mm thick) and amount of compression (typically 25%) must be the same to make a valid comparison.
There are multiple test procedures and fixture designs within the CSR test world that also have a significant impact on results. Numerous fixture designs exist, all of which produce different results. Regardless of Hornig’s1 conclusions regarding his preferred fixture, in practice, most CSR data for elastomer seal materials are gathered with a “Dyneon-modified Wykeham-Farrance” jig (see image).
Finally, CSR force measurements can be made intermittently at discreet time points using a standalone compressive load cell or continuously using a dedicated CSR testing device that incorporates a load cell for each test fixture and one or more integrated environmental chambers. With the intermittent test method, fixtures are removed from the oven and/or oil bath, allowed to cool to room temperature (23°C), manually tested on a compressive load cell, and returned to the oven for additional aging. Continuous testing does not involve repeated thermal cycling, which contributes to accelerated relaxation and worse results. As a result, intermittent CSR generally appears worse than continuous CSR for the same material under the same conditions. In addition, continuous CSR data points are gathered at the test temperature rather than at room temperature. Gathering load force data at elevated temperatures results in a higher measured load force; when compared to an initial load force data point taken at room temperature, as some procedures require. This can result in a counterintuitive situation whereby it appears that material initially improves with thermal aging. (See Figure 2.) If this is observed, it should be considered an artifact of the difference in test temperature. When comparing CSR results, it is absolutely essential to confirm that the same test method and fixture were used for both tests.
Watch for part 2 of this series, where we will explore some of the powerful insights that can be gleaned through CSR curve interpretation. For more information or assistance with your sealing challenges, contact our applications team at email@example.com or chat with us online at the Parker O-Ring & Engineered Seals Division website.
1. Hornig, R. Comparison of various CSR methods regarding the static long-term sealing behaviour of AEM, ACM and HNBR compounds, International Polymer Science and Technology, 37, No. 4, 2009.