MythBusters: Factory Testing: What Do I Really Need?
In part one (Sept/Oct2014), we covered the basic types of testing available and user guidance. That column is continued here.
Performance tests should be divided into tests for drivers and driven equipment. For drivers, especially gas turbines, output and efficiency are subject to manufacturing tolerances. Consequently, if meeting a contractually agreed upon power level is critical, a factory performance test is advisable.
These tests are usually performed in dedicated water break or load bank test stands that most of the larger machinery manufacturers have available. These tests also serve to verify that the engine meets the performance within its required control parameters, such as correct maximum firing temperature and inlet guide vane settings.
Another advantage is that the factory environment allows for precise performance tests. So, if a performance test of a gas turbine measures more than a few percent output or efficiency below the expected nominal performance, the unit may have a more serious manufacturing problem, which should be diagnosed to identify the cause. In some cases, disassembly may be required.
Some gas turbines use power turbines that are manufactured separately from the gas generator and are fabricated in different locations, so it is not always logistically expedient to test the full gas turbine prior to mating of the gas generator and the power turbine in the field. If this is the case, it is advisable to perform a field test and to include contractual stipulations that cover the expected output power and efficiency.
On the other hand, driven equipment such as compressors or pumps are more difficult to test in the factory, since a closed loop testing facility is required. Although almost all compressor vendors can offer a closed loop test to verify the performance of their compressors in the factory, these tests are expensive and may impact the delivery of the equipment by up to six weeks.
Thus, for driven equipment, one should carefully evaluate the need for a factory performance test based on previously demonstrated performance of similar equipment and the impact if the equipment does not meet performance requirements. Performance tests on compressors can be performed using the ASME power test code, PTC 10.
It comes in two different versions:
- The type 1 test requires use of a gas composition close to the actual process gas, testing at pressures close to the actual project pressures, and the same power levels as for the project design conditions
- The type 2 test allows the use of an inert gas different from the process gas and verifies compressor performance using well established similarity laws. Since it usually is performed at relatively low pressures, the power requirement is less than at the actual process conditions.
Both PTC 10 tests are for testing aerodynamic performance, and ASME PTC does not provide any mechanical integrity requirements. However, if desired, the PTC 10 Type 1 test can be specified to verify the mechanical integrity of the equipment at design conditions. This is valuable, since it provides mechanical and performance verification of the entire unit or package at full speed, full power, and full pressure.
“Testing has to be seen as a part of an overall quality assurance and risk management process.”
The concept of closed-loop testing and string testing sometimes leads to confusion: A string test, where the contract driver powers the contract driven equipment, usually on the contract skid, can be combined with a closed-loop test.
But a no-load string test can be performed without a closed loop. A closed-loop string test does not have to meet ASME PTC 10 Type 1 criteria, unless it is specified. Otherwise, especially when only the mechanical integrity is to be observed, it is often sufficient to test the equipment at full power, full speed, or full pressure.
Noise tests in a factory environment are difficult, time consuming, and expensive, since the entire package has to be commissioned to be able to perform the noise test. In general, a factory environment does not allow the free field conditions necessary for these tests, meaning the results are often inconclusive.
Thus, noise tests would best be performed in the field. Static or dynamic package tests allow the verification of control functions, warning and protective systems, instruments, vibration probes, and the package assembly, including fuel and lube oil systems. These tests are usually inexpensive and can help to avoid installation and commissioning problems.
Finally, the merits of post-test inspection and disassembly of machine internals are questionable, since they usually cannot compete with the benefits of shipping a unit with proven mechanical integrity and performance.
Since it is unlikely that we will see the federal government mandate “Lemon Laws” for rotating machinery equipment any time soon, it is imperative that purchasers carefully select appropriate factory acceptance tests. OEM’s have a vested interest in the success of their customers. Many rely on repeat customers, and customers will only come back if equipment consistently performs as promised. Testing provides the buyer protection, but it benefits the manufacturer as well, since it mitigates what might be frivolous warranty claims. One should remember that the choice of factory tests involves more than just checking boxes on an API data sheet. Testing has to be seen as a part of an overall quality assurance and risk management process.
It requires careful review of facility requirements, machinery specifications, manufacturer’s fleet experience, facility design complexity, and overall associated project risks. Failure to do so can result in either costly project delays and overruns, or expensive and unnecessary tests.
