Electronic circuits are built on foundations of substrate materials, and those materials can be very susceptible to changing environmental conditions, such as temperature and humidity. Many commercial, consumer, and even medical applications anticipate mild operating conditions, such as room temperature (RT) and low relative humidity (RH). But many applications for electronic circuits are not as comfortable, especially for systems intended for use in aerospace and defense (A&D), industrial, and space environments. Not all circuit materials are formulated or characterized for the different sets of environmental conditions required for these many different applications, and circuit materials for high frequency, high speed digital (HSD) circuits for challenging operating environments should be carefully considered.
Ideally, an electronic circuit would always be maintained at room temperature and at low RH. But even consumer products, such as cell phones, must operate outdoors during hot summer days and during winter when operating temperatures can drop below the freezing temperature of water at (0°C or +32°F). The electrical performance of a circuit material may be quite good at RT (typically +25°C) but can deteriorate with changes in operating temperature and/or RH. RH is a measure of the amount of moisture in the air, considered to be low from 40% to 60% RH at RT. These are ideal operating conditions for most circuit materials. However, many circuits spend short percentages of their operating lifetimes in ideal conditions. Depending upon the application, the changes in operating temperature and RH can be quite extreme and can affect the performance of circuits fabricated on a substrate material.
Fortunately, repeatable test methods have been developed to characterize high frequency, high speed printed circuit board (PCB) materials at elevated temperatures. Measurements of scattering (S) parameters of the signal amplitude and phase of transmission lines at known lengths fabricated on materials under test can provide the data needed to determine material characteristics such as dielectric constant (Dk) and dissipation factor (Df). For example, performed at RT, such measurements can provide reference values for a material’s Dk; performed at elevated temperatures, the test results can supply the details needed to determine how the rise in temperature has affected a circuit material’s Dk.
Temperature-dependent PCB material parameters include thermal coefficient of dielectric constant (TCDk) and thermal coefficient of dissipation factor (TCDf). TCDf is difficult to measure even with computer-controlled test equipment so a more common parameter is thermal coefficient of insertion loss (TCIL), based on evaluating how the insertion loss of a transmission line fabricated on a circuit material changes with changing temperature.
Little or no change in Dk with temperature would be ideal for many circuit applications since a circuit material’s Dk is an important contributor to the impedance of its transmission lines and circuitry. A PCB material without temperature-related change in Dk would have a TCDk of 0 ppm/°C. Such a material would be ideal but is unavailable, so circuit designers in quest of PCB materials for temperature-sensitive electronic designs usually seek materials with good Dk temperature stability, with TCDk of |50| ppm/°C or better.
Testing at Temperature
A circuit material’s Dk represents its capability to store electromagnetic (EM) energy. The Dk of a circuit material is relative to the Dk of vacuum, or 1.00. How a transmission line with well-defined length and width affects the amplitude and phase of a well-controlled test signal can provide insight into the characteristics of the substrate material on which the transmission line is fabricated.
Determination of a PCB material’s Dk requires a combination of measurements and calculations. For high frequency, high speed circuit materials, measurements to determine Dk are based on analyzing the behavior of a well-known transmission line, such as microstrip or stripline, fabricated at a known length on the material of interest. Since wavelengths (and circuit dimensions) shrink with increasing frequencies, circuit material Dk determination methods are usually based on stripline circuits at microwave frequencies, such as 10 GHz and below, and on microstrip circuits at millimeter-wave (mmWave) frequencies of 30 GHz and higher. For example, Rogers Corp. (www.rogerscorp.com) employs the clamped stripline test method to determine a material’s Dk at 10 GHz and the microstrip differential phase length method to characterize a material’s Dk at 77 GHz. The test methods provide high accuracy and can be applied at RT and at elevated operating temperatures to gauge the effects of temperature on Dk and circuit loss.
RO3003G2™ laminates from Rogers Corp. are among the many circuit materials evaluated under hot temperature conditions at mmWave frequencies (77 GHz) using the microstrip differential phase length method to determine Dk at different operating temperatures. The test method is based on measuring the phase angles of two transmission lines with different physical lengths fabricated on the same material with the same type of copper conductor. Measurements of phase are first made at RT as a reference and then at elevated temperatures. The changes in a test signal’s phase between the RT measurements and at the higher temperatures are attributed to temperature effects on the transmission lines and how the circuit material’s Dk is being affected at that test frequency and operating temperature.
Rogers Corp. characterizes higher frequency laminates such as RO3003G2 for Dk at mmWave frequencies using the differential phase length method. The test approach involves fabricating microstrip test circuits of different physical lengths on the material to be characterized, with test circuits mounted on a heat block to adjust operating temperature. Temperatures are tightly controlled by thermocouple monitoring. Test circuits are bare copper, without additional plating or finishes (which can affect circuit behavior with temperature). The test circuits are identical in every way except length, usually with a 4:1 length ratio. The test setup includes a mmWave signal generator and mmWave vector network analyzer (VNA) capable of making wideband S21 phase and magnitude measurements at 77 GHz, 79 GHz, and higher frequencies. These laminates are optimized for 77 and 79 GHz mmWave circuits, such as automotive radar systems. The 5-mil-thick materials are clad with 0.5-oz. Hyper Very Low Profile (HVLP) electrodeposited (ED) copper.
The test circuits are connected to the test signal source and VNA, and S-parameter measurements of each circuit performed first at RT and 77 GHz to establish reference levels. For measurements at elevated temperatures, all interconnections remain unchanged, with the only difference being the operating temperature, as controlled by the heat block. The heat blocks are powered to +65°C and, once that temperature has stabilized, S-parameter measurements of magnitude and phase are performed on each test circuit , again at 77 GHz. For the highest-temperature measurements at +125°C, the heat block is adjusted in temperature and allowed to reach equilibrium, after which S-parameter measurements are performed on the mechanically shorter and longer microstrip circuits with no modifications to interconnections or any other changes to the test setup. The measurement results provide S-parameter data for amplitude and phase at 77 GHz for two different-length microstrip test circuits at three different operating temperatures. The S21 phase data serves to find the Dk of the material under test while the S21 magnitude provides details for TCIL values.
The RO3003G2 laminates exhibit behavior in which both Dk and loss (typically characterized by Df) are only minimally affected by hot temperature rises. In fact, at frequencies approaching 77 GHz (see figure, left), the change in Dk (ΔDk) is minimal, only 0.010 over the wide temperature range at 77 GHz and a total variation in nominal phase (Δϕ) of only 6° at 77 GHz. Compared to a 5-mil-thick, PPE-based commercial circuit material (see figure, right), with ΔDk of 0.031 and much larger Δϕ of 17°, RO3003G2 laminates are much better suited to phase-sensitive circuits, such as automotive phased array radars, which must “beat the heat.”
The Dk versus temperature at 77 GHz is plotted at RT (+25°C), +65°C, and +125°C for 5-mil-thick RO3003G2™ laminate and a 5-mil-thick PPE-based laminate, both clad with 0.5-oz. ED copper.
Do you have a design or fabrication question? Rogers Corporation’s experts are available to help. Log in to the Rogers Technology Support Hub and “Ask an Engineer” today.