Good-bye Air Bubbles
When manufacturing electronic components it is important to protect the components against future external influences. Most manufacturers use casting or filling to do so. Especially for high-performance applications vacuum dispensing is the method of choice. A high number of manufacturers are already using this highly efficient type of dispensing. Others have not tried it yet, because they consider it too complex or shy away from the supposedly high up-front investment. A compilation of best-practice tips will show that vacuum dispensing is not rocket science. Perhaps now even the most skeptical engineer will want to give this method a try at last.
Automotive, industrial or consumer electronics: in all applications electronic components such as chips or entire PCBs must be protected against mechanical or chemical stress. One typical example are motor control sensors such as exhaust sensors or components that get in contact with hot gearbox oil. Other applications involve unprotected wire coils and the enameled copper wires of wire coiled components. Depending on their usage, any one of these components may be exposed to considerable stress in terms of corrosion, vibrations, moisture or high voltage. Since injection molding is a highly complex production technique and production costs of the molds are considerable, most manufacturers choose casting or filling for their production environment.
Conventional metering and dispensing methods, however, often do not cover all requirements. For example, the tiny gaps between coiled wires may trap bubbles of air, which break the insulating cover and diminish or even ruin high-voltage resistivity. Air bubbles caught below printed circuit boards expand when heated - by several mm if the geometry so allows - which causes tensile loads to act on the coating and on the PCB. In this case, high voltage surges may rupture the coating even if flexible and voltage resistant adhesives are used. Aggressive chemicals such as oil may get in contact with the unprotected surface and damage the component.
The Solution for Demanding Cases
From a technical point of view the generation of a perfect vacuum, which is entirely void of air, is not required. Vacuum in this context means the reduction in pressure down to approximately 1 mbar. The further the air pressure is reduced, the longer the process takes and the higher are the energy costs involved. Also, not every component can sustain a strong pressure reduction, a fact to bear in mind when applying a vacuum. While wire coiled components are mostly insensitive to atmospheric pressure, air encapsulated within a capacitor can cause the component to burst when exposed to an external vacuum. Therefore, the vacuum level should always be matched to the task at hand.
Vacuum dispensing - also suitable for in-line integration - is the method of choice especially in the production of high-performance electronic assemblies as it prevents air from getting caught between component and coating. Vacuum dispensing is not only ideal for expensive high-voltage parts and safety-related components, it also handles assemblies with complex geometries, undercuts or extremely narrow gaps.
Comparison Test: Vacuum Dispensing vs. Atmospheric-Pressure Dispensing
We wanted to demonstrate the difference between vacuum and atmospheric-pressure dispensing to manufacturers who still hesitate to try and use vacuum dispensing systems. To do so, DELO Industrial Adhesives and Scheugenpflug decided to set up a test system made of standard equipment. We used a conventional PCB in a standard PBT plastic housing as the substrate and a low viscosity two-component, high-performance epoxy resin as the casting resin to simulate a common automotive application. The resin provides long lasting temperature stability up to +200 °C. It is resistant against diesel, gasoline and oil, and protects motor and exhaust control sensors during engine operation.
The test casting under ambient air conditions was carried out with the two components manually mixed and metered. For the vacuum test run the dispensing material additionally was degassed and then metered and dispensed. To obtain a laboratory-scale setup we used the Scheugenpflug LeanVDS system, a typical entry-level model for vacuum dispensing. This compact-sized system is best suited for R&D applications, for small batch production and to replace inaccurate or time-consuming auxiliary processes such as post-evacuation.
In order to analyze the test results DELO and Scheugenpflug thought it best to use x-ray scans. As opposed to microsections this test method is non-destructive and has the additional benefit of making sure that no air bubbles are overlooked if the section was made at a position that just happened to be free from bubbles. When comparing the x-ray images of the two components distinct differences immediately became evident. While the PCB on the left was cast bubble-free under vacuum, a large bubble of air had formed under the PCB on the right, which was cast under ambient air conditions. What does that mean? Depending on how the component is used later and on its operating environment the component is very likely to fail - even if this might take several months or sometimes even years.