If you know anything about leak testing, then you understand that everything leaks. The question is: How much leakage is considered acceptable? For accurate leak testing—and more importantly, accurate leak measurement—you need a repeatable test that gives you an accurate baseline.
Creating an accurate, reliable, and repeatable leak test can be challenging. Multiple factors can have an impact on leak testing, any one of which can lead to unreliable results. When testing for leaks in any product, you must be aware of factors that could change the results. Here are seven of the most common factors that can affect leak testing:
1. Stability of the Part
When creating test conditions, you want a stable environment with minimal variations in temperature and humidity. Temperature can have a huge impact on leak testing results. HVAC systems that blow cold air onto the test station, open windows and doors, and even operator fans can change test conditions enough to produce inaccurate results.
For example, we frequently see test results that are altered because testing was conducted near shipping and open bay doors. Similarly, if you calibrate your tests in the morning, the test results may change as ambient plant temperature warms the testing environment throughout the day. You also want to avoid excess humidity, so using coalescent filters to reduce any moisture associated with using compressed gas for testing is important.
2. Seal and Fixture Design
Faulty or improperly seated seals without positive stops will have an impact on leak testing, and you can expect to see volumetric changes under pressure decay testing if you have floating seals. You want to use the right cylinder size in your pressure test to generate 1.5 to 2 times the opposing force on the part and hold the seal in a fixed position during testing.
3. Part Variations
Be mindful of ballooning parts or part porosity. Porosity can absorb pressure, and ballooning may mask leaks. To address the problem, take a little extra time to fill and stabilize the test pressure to compensate for porosity. Also, increase your stabilization time to allow for a repeatable pressure drop during test time. Be sure to let parts relax between tests to eliminate any residual pressure and get an accurate reading.
4. Test Cycle Times
You must allow sufficient cycle times for consistent testing. If you are using a pressure decay test, for example, and don’t provide enough stabilized time or test time, you won’t see enough of a difference between a known good part and an unknown reject part. Similarly, if you are using a vacuum leak test, you need to allow sufficient time to get to a repeatable vacuum level.
5. Test Pressure
When you increase test pressure, you are increasing the stress on your part, so you should increase your safety ceiling as well. Additional pressure will also generate heat. This heat can affect test results, so you should increase cycle time to wait for the heat to dissipate before measuring a pressure drop.
6. Device Sensitivity
The leak testing equipment you use needs to have enough sensitivity and be calibrated to measure at least 100 times better than the reject rate for which you are testing. For example, if you are testing to 20 pounds per square inch (PSI) and are looking for a 2 PSI difference, but the test equipment has an analog gauge that only shows 2 PSI increments, you won’t get sufficient test fidelity. It would be better to use a digital gauge that displays to 1/100 PSI.
Your test systems must be calibrated to a leak standard that can be used as a baseline for assessing whether parts pass or fail leak testing. There are two basic types of calibration:
- Single-point calibration, which provides one point of accurate measurement to a known leak standard (e.g., an arbitrary zero from a known good part). The further away the results are from the calibration value, the less accurate the reading.
- Two-point calibration, which gives you greater accuracy. With this, you calibrate zero as a pressure change with your known good part, and have a second value at the leak standard. This normalizes the curve and reveals different flow times under test.
These are just some of the most common factors that affect leak testing results. Proper calibration and a stable test environment that ensures repeatable results are the only ways to accurately test parts for leaks.