Comparison of the results of calculations of the characteristic impedance of transmission lines on printed circuit boards
V.S. Kuharuk, V.A. Ukhin, D.S. Kolomensky, O.V. Smirnova.
Printed circuit boards (PCBs) are widely used in electronic devices. They are the main node that provides communication between various components and signal lines. When designing a PP, it is necessary to take into account such an important parameter as the characteristic impedance of transmission lines (TL), both single and differential.
The LP impedance largely determines how signals will propagate through the FP. Failure to comply with the required values can lead to interference, loss of signal power and unstable operation of the entire device. Therefore, it is important to correctly calculate the wave impedance of the line [1].
There are several computer-aided design (CAD) systems that allow you to calculate the LP impedance on printed circuit boards with high accuracy. Until recently, all these systems were imported. There were practically no Russian analogues.
This article compares the calculated LP parameters between leading foreign CAD systems and the domestic SimPCB tool.
A modern engineer is quite conservative and has a hard time changing the approaches and tools he once chose for design. Only objective arguments, new opportunities, modern and proven mathematical apparatus and high quality implementation can convince a specialist to change the software tool. So, the accuracy will be assessed by comparing the impedance values obtained using SimPCB and other similar tools, as well as with real LP impedance measurements on the test board.
The study was carried out both for single LPs and for differential ones, the structures of which are most often used by engineers. Calculations were made using calculators built into the following CAD systems: Altium Designer (Altium), Xpedition PCB (Siemens), Si9000 (Polar) [2]. Just as noted above, theoretical calculations are compared with real values of wave resistance at the PP. For this purpose, drug test coupons were manufactured at the REZONIT enterprise and measured using the Time Domain Reflectometry (TDR) method according to the IPC-2141A standard [3].
LPs in the form of test coupons were calculated on the basis of a standard stack (6 layers) of a 1 mm thick printed circuit board, which was manufactured at the REZONIT plant from FR4 material (TG150) (Fig. 1, 2) [4].
After manufacturing the PP, a report was received from the plant, which displays the results of measuring the characteristic impedance for each LP (Fig. 3). All lines except coplanar LPs are counted in the report.
All research results were summarized in a single table 1. The “X” mark means that this drug cannot be calculated in the selected CAD system or using the selected method.
Figure 4 shows a summary graph of the calculation results for wave resistance.
From the above it can be seen that the wave impedance values calculated in different instruments practically do not differ from each other. The maximum error is observed when comparing the theoretical impedance with the results of measurements on a real printed circuit board. In this case, the deviation does not go beyond the established limits of 10 percent. The impedance value on a real board largely depends on the technological capabilities of production and the quality of materials. Manufacturing this test board at a different factory will result in different wave impedance values.
Thus, a specialist in the design of high-speed and high-frequency printed circuit boards can use not only foreign software, but also domestic one. The SimPCB tool from the EREMEX company, like its analogues, calculates the primary and secondary parameters of the LP with high accuracy.
Bibliography
Printed circuit boards and components of gigabit electronics / L.N. Kechiev. – M.: Grifon, 2017. – 13 p.
IPC-2141A Design Guide for High-Speed Controlled Impedance Circuit Boards.
