Regardless of their location or industry, manufacturers are demanding more from leak test equipment than ever before. Tough economic times and increased

competition require companies to use equipment that not only provides accuracy and repeatability, but accommodates a wider variety of parts.

For example, an Indian Tier 1 auto supplier uses an M-1075 detector from InterTech Development Co. to test three sizes of fuel filters. The fixtures on this mass-flow, air leak detector are quickly changed, enabling the manufacturer to test 138 filters per hour. The filters are tested to a 0.25-standard-cubic-centimeters-per-minute (sccm) limit at 6 bar. Once tested, parts are automatically marked, unloaded and diverted to either an accept conveyor or a reject bin.

Delta Faucet Co. uses Sentinel instrumentation, made by Cincinnati Test Systems, to leak test its Pascal culinary faucets, as well as verify that several internal sensors are working. Sensors on the handles and spout are tested to make sure they respond correctly to the user’s touch. Human touch is simulated through a specially designed circuit that removes variability. During testing, faucet parts are isolated and held in place using nonconductive elements. Also tested are sensors that open and close the water valve, and detect when the spray wand has been removed from the spout holder.

Another U.S. manufacturer, based in North Carolina, leak tests automobile gear assemblies using the System 100 detector made by VIC Leak Detection. Initially, in 2005, the German parent company suggested using pressure decay technology for leak testing. However, the System 100, which uses mass-flow technology, proved to have a shorter cycle time (30 seconds) and offer more flexibility, as it is easily and cost-effectively upgraded and expanded. Currently the company uses it in seven workstations and leak tests more than 200,000 assemblies annually.

Focus on Family

“No air or tracer gas leak testing technology is inherently more flexible for all applications,” says Jacques Hoffmann, president of InterTech Development Co., which designs and builds equipment for leak testing, functional testing and test-centric assembly. “All the technologies offer quite a bit of flexibility, so long as manufacturers use them to test a family of parts to similar, not identical, requirements.”

An automaker will often use the same piece of equipment to leak test many heat exchangers because its fixture accommodates slight dimension and connection variations. Dave Morris, product manager for helium-hydrogen leak detection and gauges for Pfeiffer Vacuum Inc., says this flexibility enables some automakers to test as many as 40 models of heat exchangers at one plant.

Usually, a manufacturer doesn’t make or need to test very different parts, such as oil filters and radiators. However, it might make and test oil coolers and radiators, both of which are types of heat exchangers. In that case, the same leak tester can be used, even if the parts are used in different industries or applications.

The Premier MF 5 differential mass flow leak tester has this appeal. Made by ATEQ Corp. USA., the device is used by medical companies to leak and flow test blood bags, by appliance manufacturers to test gas ranges, and by automakers to test exhaust manifolds.

“Another way OEMs increase their leak test flexibility is by using standardized instruments with the same features at several plants,” says Guy Dewailly, president of ATEQ Corp. USA. “It enables them to switch out the testers among plants as needed to ensure part quality.”

If a manufacturer is unsure about using a leak tester on more than one part, it needs to consider several part- and machine-related factors. The company needs to prioritize them in light of each part’s production cycle time and its available budget to select the best equipment for each application.

Part-related factors include size, temperature, material, assembly method, work environment and leak rate standard, which is the amount of leak allowed by the manufacturer for each part. Machine-related factors are test time, pressure and vacuum range, and software capabilities.

Moderate or extreme differences in part size will often require the tester’s fixture to be changed, which usually increases downtime between tests. However, custom or state-of-the-art fixtures can virtually eliminate downtime. Hoffmann says some fixtures are available with electric cylinders that assemblers can preprogram to ensure almost-immediate changing.

Manufacturers may want to use in-house fixtures for certain applications to save money, sometimes as much as 75 percent. The company can also receive fixture-design advice from their leak testing equipment supplier.

On the negative side, the manufacturer must assume all responsibility for any fixture-related problems during testing. Also, making fixtures for certain applications—such as a large part with multiple cavities—is not easy and can result in lost money and time, says Dewailly.

Manufacturers need to remember that the test time of a system will vary depending on part size, and that part size should match the capability of the tester. Large parts have a greater volume and take longer to fill and test. Parts that are too small or too large for the tester will prevent the device from obtaining accurate data. As a result, the tester might reject a good part or pass a bad one.

Part temperature, material and assembly method are interrelated and should not be considered separately. For instance, a soft plastic part stretches more than a hard plastic or metal part when subjected to vacuum or pressurization.

“Think of the plastic container that holds washer fluid in an automobile,” says Dewailly. “It balloons when pressurized, and its volume is increased.”

“Make sure each part has some time to cool when tested,” says Joe Pustka, director of applications engineering for Uson L.P. “Too-high-temperature parts can negatively react to testing because temperature compensation does have limitations. Also, a part’s vacuum or pressure limit can differ depending on whether it is molded, welded, glued or fastened together.”

Another factor is the environment where testing is done. Is the part tested in a clean room or next to the production line, where air drafts, humidity and air-borne dirt can impact test results?

Leak Tester Limits

When specifying test cells, manufacturers will often try to accommodate the greatest possible number of parts—but fail to consider the variations in part geometries. The correct instrument will usually be able to handle various test specifications.

However, part configuration differences can add complexity to the fixturing and, possibly, the test circuit design. Hoffmann says manufacturers can calculate this complexity by determining the number of interactions. He recommends the following formula be used: (N x (N-1))/2, where N represents the number of parts.

“Let’s say a tester handles 10 parts per hour well, but you’d like it to handle 12,” says Hoffmann. “Testing 10 parts requires 45 interactions (10 x 9 divided by 2), whereas testing 12 parts requires 66 interactions (12 x 11 divided by 2). The unit may not handle those two extra parts very well.”

Some testing methods, such as pressure decay, tend to take longer to perform than others. For this reason, manufacturers need to use equipment that performs

reliable testing in the shortest time possible. Otherwise, a too-long test time will hurt productivity.

Equally important, assemblers should never speed up a test, hoping it will increase productivity, says Pustka. Performing a test faster than recommended can result in data overlap within the tester.

Using software tricks instead of relying on sound physics can result in a faulty reading with consequences like passing a bad part or rejecting a good part.This can easily occur during a pressure decay test, for example. “Sometimes engineers don’t allow for a sufficient stabilization period before doing the test,” says Dewailley.

“Stabilization is crucial because that’s when the fixture seals stop expanding, the component being tested stops stretching, and the air temperature inside the component equalizes.”

The leak tester must be able to produce the proper pressure or vacuum range for each application. So if 200 psi is required for a test, don’t use a unit with a lower range and try to push it.

Most testers feature operating software that enables users to quickly change test parameters such as leak rate reject limit, pressure, part dimensions and cycle time. Some testers let the operator select from dozens of stored test settings to save time.

Chuck Wilkinson, director of marketing for VIC Leak Detection, says some equipment enables assemblers to scan the part’s bar code, serial number or internal part number so the tester immediately knows the type of part it is testing, and which test program to retrieve.

Always use the proper amount of helium or hydrogen with trace gas technology testers. “To save money, a customer may be tempted to use a lower concentration of helium because it’s very expensive,” says Morris. “Doing this may provide the tester with too-little helium for a good test and precise validation of a good part. Helium recovery is another avenue that should be explored to save costs.”

Assemblers need to perform regular maintenance of each tester. For example, tracer gas units that use helium feature contact parts like seals that need to be checked daily.

Although no one leak test technology inherently offers the most flexibility, certain technologies tend to be more popular than others.

Helium mass spectrometer, a tracer gas technology, is often used to detect small leaks due to its wide sensitivity range: 10^-1 to 10^-10 atmospheric cubic centimeters per second (atm-ccps). Among air-based technologies, pressure decay is popular because the testers are easy to use and relatively low cost, and air is much cheaper than helium and hydrogen.

The ASI 30 modular helium leak detector from Pfeiffer Vacuum Inc. offers high sensitivity in normal, gross or sniffing test mode. Its minimum detectable leak rate is 5 x 10^-12 mbar liter per second in normal mode, 1 x 10^-6 mbar liter per second in gross mode, and 5 x 10^-8 mbar liter per second in sniffing mode. Maximum inlet test pressure is 4 mbar in normal mode and 40 mbar in gross mode.

Another leak and flow tester is the Optima vT, made by Uson L.P. It features a test control unit that monitors up to four sensors simultaneously, so long as the part under test doesn’t share a common wall. Often used by military and aerospace manufacturers, the tester is available with one or two test channels. It offers a pressure range up to 225 psig differential or 1,500 psig gauge standard.

Quite often, a tester’s capabilities determine how much leak test flexibility a manufacturer has in an application. Sometimes, though, it is determined by the manufacturer.

“One customer of ours uses different air technologies in two plants to perform gross (large) leak testing,” says Gordon Splete, products group and marketing director for Cincinnati Test Systems. “One prefers mass flow, while the other likes pressure decay. Both favor techniques they’ve used for a long time and are comfortable with.”

Make it New Again

Today’s testers not only increase a manufacturer’s leak testing flexibility. They also provide many years of service so companies can maintain this flexibility. Many older testers also offer these benefits because they have been upgraded.

“In 2010, we visited a plant in Mexico and saw they were still using some testers made in 1982, and they were performing well,” says Pustka. “They needed some pneumatic circuit maintenance over the years, but that’s all.”

Knowing that a tester can be upgraded gives a manufacturer some assurance that the equipment will perform effectively in future applications. However, there is a cost and downtime trade-off to upgrading.

“For a reasonable cost, some of the tester’s controls, operator interface display or tooling can be upgraded,” says Wilkinson. “Or maybe some safety features need to be upgraded to meet a new standard. These changes will improve performance, but the end-user must plan on being without that tester during the upgrade.”

When a detector approaches 10 years old, it usually needs some type of upgrading but the internal technology may well be kept intact. But, when a tester exceeds 10 years old, it may need to be repaired. That can sometimes be tricky.

“Components for very old machines, such as PCBs, are often not available,” says Morris. “This reality, along with the fact that a new tester will perform much better, often convinces a manufacturer to buy a new tester.”

However, some manufacturers stick with their old tester—even if a new one is more productive. They fail to appreciate the increased uptime, greater communications flexibility and competitive advantages of meeting customer quality demands.

 But, this doesn’t mean that there is never a place for some simple technologies.“Submersing a part under water and counting bubbles may appear to be an outdated test method,” says Hoffmann. “But, if a company makes and tests less than 1,000 castings per year, submersion might actually offer the most flexibility compared to other technologies.”