Inside be quiet!

11.03.2015
Quality management during product development

How we assure product quality in advance and use our own engineers for Safety Lab tests
It’s not our philosophy to buy an OEM platform, specify some modifications, and hope that a decent end product will result. Instead, with our own well-equipped office and on-site laboratory, as well as a multi-disciplinary team of German engineers, we are able to set new standards for product quality and durability starting right with the first test phases of our prototypes. And as part of our corporate mission to be the best we possibly can, we want to give all interested parties an insight into the be quiet! Safety Lab and reveal what hardly any outsider has been allowed to see until now.

In this issue of “INSIDE be quiet!” we will show you which tests a power supply must pass in the initial stages before it earns the be quiet! name.


Our testing procedures

Every one of our power supplies must pass important test procedures that we consider “standard,” and meet a host of safety requirements. Among these tests:

  • Functional Tests: During every development stage we check all power supply parameters to ensure one hundred percent reliable function at all times.
  • Thermal Tests: The power supply is tested under a variety of environmental conditions to ensure there will be no failures and to make sure it operates flawlessly under these conditions.
  • EMC Test: We verify that our power supplies won’t affect other devices through unwanted electrical or electromagnetic effects.
  • Safety Review: Here we make sure that our power supplies meet all relevant safety criteria.
  • MTBF (Mean Time Between Failures): All components of the PSU are separately tested in a comprehensive stress test and the consequence of the test is an MTBF calculation. This test is to ensure the durability and a low RMA rate.
  • Product Feasibility Verification. The factory validates on a technical level whether our design can actually be built and manufactured for the specified costs. The production steps need to be feasible.
  • PFMEA (Failure Mode and Effects Analysis). This test corresponds to a quality control that determines whether there are elements of the product that can lead to problems in a later stage (wrong or unfavorable arrangement of the components, faulty development of the layout, etc.).
  • PCPK (Process Capability Index). Based on a simulation of time estimates, we calculate the time required to produce one power supply in a specific run. We are also estimating how much time is required for each step, until the entire process runs smoothly and is productive. This process must function well.
  • POATY (Over All Test Yield). This is a term used in manufacturing. In this test, we estimate through a simulation how high the monthly volume will be, or how much the production line can be stressed every day. This process is aimed at finding out how high the in- and output of the production line will turn out.
  • PHi-Pot Test. The high voltage test is used throughout the field of electrical engineering to ensure the insulation resistance of electrical equipment and electrical installations.
  • PVibration Test. Here we examine how well the power supply copes with vibrations in the normal range.
  • PHALT (Highly Accelerated Life Test). This is a qualitative test procedure that subjects the electronic and electromechanical parts to rapid aging in order to discover vulnerabilities and design errors.

 

In addition, we have introduced the following tests with our producers:

  • Thermograph Test. Here we create a thermal image recording under maximum load (one hundred percent utilization at a room temperature of 40 degrees Celsius), thus identifying and allowing removal of potential hot spots during development.
  • Long Time Burning. For this test, we overload the power supply at an ambient temperature of 40 degrees Celsius for one to two months, to see if the product will fail.
  • Compatibility Test. Naturally, a power supply must have a high compatibility with modern systems and modern hardware. This is extensively tested in our laboratories with a wide variety of components to ensure that our products are future-proof.
  • On/Off Cycling: The power supply is switched on and off thousands of time under full charge. With this test, we make sure that these loads do not damage the power supply or its components.
  • Manual tests: Besides automated Chroma tests (see below), we also perform manual control tests to assess the behavior of the power supply under different loads. One of the tests is to evaluate the cooling curves under different workloads (from 0%, 10%, 20% and so on until 100%).
  • Mechanical tests: These tests include the ‘drop test’, in which the power supply needs to survive a fall from a predetermined height without damage. Also, the paint strength is tested for mechanical stress by means of pointed tools. Another test ensures that no individual components will get loose inside the power supply, and finally all the cable sleeves and connectors are tested for quality and proper performance.
  • Chroma Testing (ATE): There will be done a lot of tests on the Chroma machine during different product development phases.

Higher level testing
Although the ‘standard’ tests listed above guarantee product safety, compliance with all regulations, and a certain level of good performance, that is not enough for us. In order to ensure our high standards of product quality, stability and compatibility, we have introduced additional test steps.

The most interesting of these tests are shown below. The results of individual test steps can correspond to multiple tests in the lists we have shown you.


Durable component selection

During the ATE (Automatic Test Equipment) Test with attached thermal chamber, all components are checked for suitability and durability. This doesn’t just mean that components need to comply with the defined technical parameters, but also that they need to be durable under severe conditions.

Are the components really so functionally perfect that zero failures occur at limit conditions? This can and must be simulated in order to guarantee the quality of the platform and plan the placements of the components.

During the manual functional and stress tests, we go into even more detail: do the individual components deliver on what the manufacturer and data sheets promise? These tests must be survived by each single component, since we will be using the ‘weakest’ component to establish a minimum operating time before possible failure, the MTBF.


Temperature and climate change, and brutal heat tests

The various temperature and climatic conditions (e.g. humidity) play a very important role in the early tests. In the environmental and shock tests, extreme situations are simulated and the units are retested for durability.

We also check designs for suitability in all possible geographic regions and for environmental factors that might arise during shipping and storage. Things like possible hairline cracks, inadequately sized components, and erroneous connections can be recognized at this point and fixed or avoided entirely.

The next test is known as ‘the burner’ in every true sense of the word. The burn-in test gives the power supply prolonged exposure to extreme temperatures under full load, and every power supply must prove itself under these grueling conditions. Only when this test is completed without failures can the design and construction be considered good enough to work with sufficient reliability in a computer, even under suboptimal conditions.

The HALT Test (Highly Accelerated Life Test) is a special stress test methodology that improves product safety during the development process. Also tested during the HALT is vibration resistance, important to make sure the power supply can withstand any vibration that could occur during transport.


Radiation safety, EMC and electrostatic influence tests

During EMC tests (Electromagnetic Compatibility) we differentiate between testing for sensitivity to electromagnetic interference and testing the emanation of electromagnetic interference, in both cabled and cable-less scenarios. Our power supplies must pass all tests without failure to prove their immunity to external disturbances.

The next tests are among those more difficult to control, serving to monitor operation and user safety.

The ESD (Electrostatic Discharge) test simulates an electrostatic discharge. This is a suddenly and briefly occurring electric current, between two objects with a different electric potential. In a worst case, this will lead to a total failure of the power supply. We test up to 8,000V.

The EFT (Electrical Fast Transient) test simulates peaks in the incoming AC signal, which for example might occur by switching off inductive loads up to 40A. Testing for surge immunity closes this batch of tests.


Cooling and noise levels—there’s an awful lot to optimize

With the Airflow and Pressure Test we are testing optimal air circulation with regard to air flow design. This test step also allows us to detect and avoid or minimize turbulence and air breaks that may negatively affect cooling and noise levels.


Fan, fan control and the real be quiet!-feeling

To operate power supplies throughout their power ranges while consistently maintaining minimal operating noise is a vital cornerstone of our corporate philosophy. That is why, for us, working in a certified anechoic (practically echo-free) room is enormously important. (See: INSIDE be quiet! - the Straight Power 10 Story)

Just optimizing a fan curve for many different load conditions is a lengthy process—one which we do at higher temperatures then would normally occur in the homes or offices of end users. Whether we are considering the steady operation of a newly developed fan at extremely low speeds (200 rpm) or the sovereign cooling of components under extensive full load, our goal is clear: from the second we start planning, our focus is on developing the quietest power supply we can in each and every class. The actual development and testing phase for a new PSU design usually lasts more than six months. If any technical changes are made, we repeat the entire test suite. It is also important that we don’t just let other people perform our tests: we always have our own professionals present to execute or support all tests. From past experience we have learned that this is the only way we can be sure to meet our high standards of product quality.

Another important issue for us is that we never allow our power supplies to operate right at the limits of what is allowable; we always over-engineer to achieve optimal results. For example, standards say we could allow our PSUs to be ‘just good enough’ in the area of electromagnetic compatibility, but we won’t. We always design to be far within regulated maximal limits. The effort and costs required are much higher this way, but doing so allows us to guarantee that our power supplies comply with all standards and regulations, under all conditions and despite manufacturing variability.


Comprehensive system and compatibility tests

To ensure that our power supplies actually meet the demands of our strictest customers, we invest a lot of time and money. For example, at all of our locations, we implement appropriate test systems. These are not only positioned at our headquarters in Glinde and the Taiwan office, but also at our suppliers in Taiwan and China. This makes it possible for us to ensure broad compatibility of our power supplies and the corresponding components at any time. We don’t just test whether our power supplies work with current and future graphics cards, CPUs and motherboards, but we also ensure smooth operation under any completely new requirements that occur—for example, with new energy-saving features.

Chroma station
To be able to closely analyze and check our power supplies during each phase of product development, we use Chroma stations. These test devices are expensive but essential for the development of a perfect power supply. We don’t just use conventional test programs, but also write our own processes, comprising up to 300 individual tests. In order to determine the detailed and accurate Chroma values throughout the production phase, in Germany alone we have three of these stations. The Taiwan office and our suppliers have Chromas as well, performing additional tests.

Product safety testing
The subject of ‘product safety’ is very important to us and is discussed in detail among our engineers. That’s why each of our power supplies is outfitted with all modern protection circuits. These include OCP (Overcurrent Protection), OVP (Overvoltage Protection), UVP (Undervoltage Protection), SCP (Short Circuit Protection), OTP (Overtemperature Protection) and OPP (Overpower Protection). Again, we do not simply rely on the default policies established by the standards-makers, but we also individually adjust the limits of the protection circuits for the target audience of the power supply. This allows us to ensure maximum safety. We also make sure that the power supply neither switches off too early or too late when used with current hardware but rather reacts optimally in all situations, resulting in maximum compatibility and safety. These limits are checked and readjusted in each test phase..


In the end

Our power supplies go through several stages in the development phase, which we internally designate as A, B and C-test. During each phase, all specifications of the power supply are again manually checked against our own custom-devised checklist to ensure that all parameters are met. Only after all these tests, and a longer real-world use test in actual PC systems, are completed is the power supply ready for mass production and sale. And with that, we have given you a glimpse into the next “be quiet! INSIDE”, which will be published as an interesting photo story.