Engineering reliability means credibility
02 March 2020
Remember, remember the 4th of March. Wait, that’s not accurate, is it? And one thing engineering should represent is accuracy! However, the 4th of March will mark the first annual celebration of ‘World Engineering Day for Sustainable Development’.
This tutorial was originally featured in the March 2020 issue of EPDT magazine [read the digital issue]. Sign up to receive your own copy each month.
This UNESCO (United Nations Educational, Scientific & Cultural Organization) International Day has been established to highlight the achievements of engineers and engineering in our modern world – and to improve public understanding of how engineering and technology is central to modern life and sustainable development. In this tutorial, Andy Naisbitt, Operations Director at electronics test & measurement equipment experts, GEN3 explains how engineering accuracy and reliability translate into credibility for electronics manufacturers...
GEN3 has been engineering and refining its CM Series Contaminometer tester for over 40 years. Like all measurement systems, the difficulty is in what we are trying to measure. For example, light is measured by how many candles produce a level of luminosity (candela). But what if the measurement needs to identify a potential impact on a process or product reliability?
In the automotive industry, you cannot paint metal when the surface dew point is closer than 3°C to the ambient air temperature. Now, this measurement is critical to the process finish, as the paint won’t adhere well, due to an invisible layer of moisture on the surface. The measurement is performed with a dew point meter to ensures the metal surfaces are above 3°C of the dew point prior to painting.
The GEN3 CM Series Contaminometer measures the conductivity for an equivalent weight of salt within a square centimetre of the circuit board to measure the effectiveness of your cleaning process for removing ionic contamination from your board.
The closer pitch devices used in modern electronics manufacture (0.3mm or less) increase the electric field (E = V/d), which makes the electrochemical corrosion cell potential more likely during local condensation under humid environments.
The average size of dew droplet formation on surfaces at different temperatures varies from 20–50 µm at about 50% RH, to large droplets of water at >60%RH; add in the natural capillary action under low profile components with stand-off heights of 25µm, and water bridge shorts become a reality.
This describes the problem: even invisible amounts of salt (or salt equivalent) left on the board can present conditions suitable for electrochemical corrosion.
Although we understand that the salt we are measuring may not be Sodium Chloride (baseline value), there are various known conversion factors that allow you to convert the baseline values for common PCBA contaminants.
Example from Table 1
If there is a reading measurement of 1µgNaCl/cm² (salt), but we believe the contaminant to be Carbonic acid residues (H2CO3), the equivalent concentration for Carbonic acid shown in Table 1 would be 0.16µgH2CO3/cm², showing that a smaller amount of Carbonic acid is more corrosive.
In other words, the smaller amount 0.16µgH2CO3 is required to produce the same reading as 1µgNaCl (salt).
Measurements need to be accurate and repeatable. We have the baseline (salt NaCl), but how do we measure accurately with such small trace amounts – and do it repeatedly to achieve a Gage R&R <10?
• Variables that require tight controls to measure such minute quantities
• Calibrated solution (polished to >8MO purest level of DI Water Type I)
• Temperature compensation required, due to the sensitivity of the measurement to small changes of solution temperature
• Flow rate as fast as practical, but not so fast it generates turbulence/bubbles
• Measurement time <3 mins, longer dwell times can dissolve key components from the laminate
• Measurement accuracy 0.005µgNaCl/cm² required to achieve accuracy (Six Sigma) and repeatability (Gage R&R <10)
• CO2 compensation algorithm; however, keeping low turbulence is essential, no amount of compensation algorithms can account for bubble generated CO2
The issue with such sensitive measurements is that variables like subtle changes in solution temperature have an impact on the measurement. Exposure to air, even in the small cavity between the lid and the solution, can impact the final readings; more so if high turbulence is experienced, which traps air in the solution and rapidly dissolves CO2. The solution sounds simple: run the test as fast as possible. However, too fast and you generate bubbles; this impacts the readings, as the measurement cell is measuring air (bubbles), not solution conductivity. However, too slow means extended time in the tank; we know that water absorption into the laminate is enhanced by residues on the surface, and multiple reflows. The test solution has a low dyne value and therefore easily wets the board.
The test solution is quite aggressive and extended time in the solution can penetrate deep into the laminate dissolving key components from the board.
Recent studies have shown that solder mask enhances the moisture uptake by four times that of the laminate, which correlates to a study by the European Space Agency, which found that extended exposure in the IPA/DI water solution led to leaching the Bromide flame retardant out of the laminate through the solder mask onto the board surface, severely degrading the material properties of the PCB.
The analysis of solution with trace levels of ionic matter requires the precision of a solid gold conductivity measurement cell, within a solid-state measuring cell, and connected to a ballistic amplifier. This, and a powerful pump, allows smooth flow throughout the operation, giving a Six Sigma (6s) repeatable measuring accuracy down to ±0.005 µg/cm².
Contact Details and Archive...