How to improve communication and build a better custom component
(Editor’s note: whether you are doing designs for power, medical, mil/aerospace, industrial markets, or other applications, you’ll need more than just ICs. This article looks at how custom non-IC components—devices such as passive components, connectors, cabling, enclosures—can improve designs and make them viable, and what it takes to work with a custom-component supplier.)
We have seen the end of the era in which standard components rule the market. Today, design engineers must meet demanding guidelines on size, reliability and function that make the prospect of adapting mass-market products daunting and impractical.
By contrast, custom work enables manufacturers to meet size, weight and voltage requirements without wasting resources on standard components. Figure 1 shows some of the many custom capacitors that have been done to meet specific application requirements.
However, successfully building components to order requires close collaboration between design engineers and manufacturers. To ensure the best outcomes, follow these five tips:
1. Expect the unexpected. You can’t plan for everything, and that is especially true in custom component manufacturing. A surprising obstacle can either derail a project or be solved quickly. The difference in outcomes hinges mainly on the strength of the relationship between the engineer and the manufacturer. Let’s say, for example, that a customer is dissatisfied with its capacitor supplier because of failures in the application. It’s likely that the manufacturer is frustrated, as well. The capacitor, designed according to standard specifications, might appear fine under testing analysis, even though the customer reports failures.
There are numerous possible explanations for the problem. Perhaps the capacitor is mounted too closely to the printed circuit board (PCB), creating stray capacitance between the capacitor body and the board. Because the capacitor thus formed in parallel with the mica paper capacitor is much lower in capacitance to the mica paper capacitor, the voltage stress is much higher and causes the circuit to short between internal connections in the mica paper capacitor and the PCB.
Had the design engineer known that guideline from the beginning of the design process, an extra wrap of mylar tape around the mica paper capacitor would have provided the additional dielectric strength to prevent the arcing and solve the problem. This kind of issue is difficult to correct and can cost both companies time and money, and can be easily resolved through open communication and collaboration.
2. Share from the beginning. The best time to establish an open relationship between design engineers and component manufacturers is at the start of a project. Before work gets underway, discuss salient application considerations including temperature, shock, vibration, atmospheric pressure and electrical stresses.
3. Be prepared to shift design and packaging based on need. One of the greatest benefits of design collaboration is the ability to spot a better method mid-project. For example, during discussions about how certain mica paper capacitors should be molded, collaborators might find that they need resistor, diodes and other components in the assembly. These parts must be factored into the calculations of what space is available for the capacitor.
As stakeholders discuss these issues, they might determine that the components can be assembled with the mica paper capacitor and molded assemblies as single units, saving space and offering convenience for the customer. Custom shapes to molded units can also provide mounting options for other devices, as well.
In addition to custom shapes, custom electrical terminations and mounting hardware can usually be integrated into the molded unit to offer more space and weight savings. In these instances, conversation leads to great savings and better products.
4. Don’t quit collaborating before the testing stage. If design engineers and component manufacturers walk away from each other before conducting accelerated life testing, they have only completed half their task. By testing samples of the desired capacitor at multiple, elevated stress levels that are similar to the actual operating stresses, the capacitor will experience an accurate prediction of when and how it might fail. Engineers can make these predictions in a relatively short time, and the results can be significant.
One example in which this holds true relates to an aircraft ignition-systems capacitor. The testing process begins by making samples of the desired capacitor with less mica paper dielectric than usual, i.e., 10 units with 2.6 mils paper thickness, 10 units with 2.8 mils paper thickness and 10 units with 3.0 mils paper thickness, where normal design criteria would call for 3.6 mils of dielectric. Using a spark gap of the same breakdown voltage and using circuitry as close as possible to the finished unit provides three levels of voltage stress.
By using different oven temperatures of 85°C and 125°C, there are two levels for temperature stress. Engineers can increase the pulse rate of the spark gap to speed up the process even more. At 20 Hz, companies can achieve 1.0 X 108 pulses in less than two months’ time.
Manufacturers can use these test procedures to determine the probability density function of each of the various stress factors and predict life at any given voltage or temperature with a certain degree of confidence and reliability.
5. Embrace the benefits of customization. Design engineers can reap enormous benefits from customized components. Grappling with standardized products that exceed application needs and waste resources is no longer necessary in an industry capable of creating a component to fit every potential application. When companies voice their specific needs before components are formed and continue communicating throughout the creation process, manufacturers and customers ensure successful outcomes.
About the author
Joe Moxley is a senior engineer at Custom Electronics, Inc. (CEI), a manufacturer of high-quality, high-reliability electronic products for military, commercial/industrial, renewable energy, aerospace and oil exploration markets. Joe has 31 years of experience in the design and manufacturing of reconstituted mica paper capacitors with short forays into stacked/plate mica capacitors and polypropylene capacitors.