Elliot, Jake Cameron (2022) PCB standards. [USQ Project]

Text (Project)
Elliot_J_Brown_Redacted.pdf Download (34MB)  Preview 
Abstract
Printed Circuit Boards (PCBs) are one of the most common components of an electrical device. However, most PCBs are designed around older standards that have not been updated to suit modern technology and manufacturing processes. This project investigated two of those standards. The IPC2221 current curves use conductor area and maximum temperature rise to determine the maximum current and the rule of thumb around not using 90° bends in trace circuits. The IPC2221 current curves were tested at the conductor width of 5 mils as that is one of the most common widths used in industry. These measurements were taken at three different temperatures, with increases of 10°C, 20°C, and 30°C from ambient to reflect the curves published in the IPC2221 document. Likewise, the standard regarding trace bends was tested with five mil traces. The three varieties of bends were 90°, 45°, and straight traces as the baseline.
For the IPC2221 current carrying curves, the 5mil trace performed much better than expected across all three temperature ranges, 10°C, 20°C, and 30°C above ambient. Based on the new constant K from the IPC2221 equation used to plot the original values, which was calculated based on the currents recorded, an increase of 234% more current on average was found compared to the original plot in IPC2221 to the latest current curves. From the unique current curves, the average currents were above the expected results using the new value of K, with +10°C yielding a 14.7% increase, +20°C with a 29.5% increase, and +30°C had a 27.3% increase in current. This uses the same nominal derating factors of 10% for all values and an extra 15% for any trace areas over 5.75 mils, as stated in the IPC2221 Graphing notes. The K value used in the original publication was 0.048 for external traces. However, from the calculated results, the average K value was 0.1446, an increase of 201%.
Furthermore, the 45 and 90bends maximal current tests produced exciting results. The average maximal current for straight traces as a baseline was 2.773 with a standard deviation of 0.12A, compared to 2.823 with a standard deviation of 0.27A for 45° bends and 2.842 with a standard deviation of 0.207A for 90° bends. However, the 45° and 90° maximal currents were within 0.67% of each other, although the 45° bends had a standard deviation of 30.4% higher, shown by the much higher volatility of the results.
The same was evident in the timetofailure testing of the three traces when the current was kept constant. Again, straight traces were the top performer as, on average, the time to failure was 175 seconds which was 661% longer than the 45° or 90° bends that both averaged 23 seconds. Straight traces maintained on average 6.46% current more than 90° bends but 4.9% less than 45° bends, with a current of 2.781A when the failure occurred. The standard deviation for the 45° and 90° bends was 3 seconds for the 45°, and 11 seconds which is a 266% increase in the variability, which is quite large when both traces had a mean of 23 seconds.
The failure points of the 45° and 90 ° bends were also a metric that was recorded and compared. Interestingly, 45° had recorded 25 failures at a straight and 45 failures on bends. This was the inverse to 90° bends which recorded 49 straight failures and 26 failures on bends. Notably, 90° bends also had the least number of visual defects at one, compared to four for 45° bends and two for straight traces.
Statistics for this ePrint Item 
Item Type:  USQ Project 

Item Status:  Live Archive 
Faculty/School / Institute/Centre:  Current – Faculty of Health, Engineering and Sciences  School of Engineering (1 Jan 2022 ) 
Supervisors:  Brown, Jason 
Qualification:  Bachelor of Engineering (Honours) (Electrical and Electronic) 
Date Deposited:  19 Jun 2023 22:37 
Last Modified:  20 Jun 2023 01:14 
Uncontrolled Keywords:  Printed Circuit Boards (PCBs); electrical device; standards 
URI:  https://sear.unisq.edu.au/id/eprint/51884 
Actions (login required)
Archive Repository Staff Only 