We've all done it at least once: looked at a report, read the numbers on it, and come up with exactly the wrong conclusion. Pass / fail limits and results are printed right there, but for some reason, our brain just misinterprets the two. It's a passing value, but for some reason we think it shows a failure instead. Imagine a police officer writing a speeding ticket for driving 53 MPH on a road with a 55 MPH speed limit.
It's not a problem with the test itself, it's a problem of interpretation. That means the old carpenter's adage, "measure once, cut twice; measure twice, cut once" doesn't address the issue. The same issue of misunderstanding the values on a test report occurs in the rubber seal industry about once a month. Passing results are misinterpreted to be failing results, and good values are thought to be bad ones. Here are four of the most common test report misunderstandings I've run into.
Most low temperature testing involves negative numbers and that creates some confusion when coupled with “greater than / less than” test limits. For example, it is common to see pass / fail limits for TR-10, glass transition (Tg), and sometimes impact brittleness expressed in the form of “-30°C max.” I’ve talked to people who claim that a result of -32°C failed against a limit of “-30°C max” because everyone knows that 32 is a bigger number than 30. Just to be clear, -32°C passes, -28°C fails. This is a 2nd grade math problem, but sometimes we adults forget how to do the simple things if we don’t do them often enough. If this gets confusing, the easy way to interpret temperature limits is to mentally replace “max” with “no warmer than” and “min” with “no colder than.”
Compression set is always expressed with a maximum limit, for example, 20% max. Therefore, all values up to and including 20% are passing values. I’ve heard people claim that a result of 20% fails against a limit of 20% maximum because it’s not below the maximum. I won’t disagree that “barely passes” isn’t a good situation to be in, but “barely passes” is still a passing value. Compression set is a measure of what percent of the original squeeze has been permanently lost. A value of 100% means the material has gone completely flat. A value of 0% means the material returned all the way to its original dimension. With compression set, small numbers are good, big numbers are bad.
Compressive Stress Relaxation
Commonly used in the passenger car and commercial truck industries, Compressive Stress Relaxation (CSR) is a measure of how much spring force the rubber has left after aging or being exposed to a fluid. The limits are always expressed as “10% min retained load force”, as one example of a common limit. CSR moves in the opposite direction as compression set, and perhaps this is why a result of 15% is frequently and incorrectly thought of as failing a “10% min” limit. For CSR, the more retained seal force, the better.
Most limits for tensile strength and elongation change after heat aging and fluid immersion are one-sided limits, meaning they only have one limit, not two. For example, a heat aging requirement may have limits of “-30%” or “-30% max” for the tensile strength change. What happens if the result is +2%? In other words, what if the result has the opposite sign from the limit? This is a passing result. There is no implied “to 0” limit attached to these one-sided limits. There is also no implied “mirror image” limit with the opposite sign. By this I mean that a “-30% max” limit does not automatically include a matching “+30% max” limit. A result of +100% is still a passing value compared to a one-sided limit of -30% max. If a specification does not explicitly call out a two-sided requirement with both a high limit and a low limit, then it only has one limit.
This is something that happens frequently, but it doesn't make it any less embarrassing. In an ideal world, someone is there to patiently explain the data and the limits and show how the report actually shows passing data, not failing data, and does so in a way that doesn't make you feel worse about it. If I have been that person for you in the past, a thank you cake would be appreciated. You and I know who you are, but we can keep that to ourselves. Cookies are good too.
For more information on the proper way to read a test report, please contact the engineers at the Parker O-Ring & Engineered Seals Division.
This article contributed by Dan Ewing, senior chemical engineer, Parker Hannifin O-Ring & Engineered Seals Division.