Mobile wireless communications systems are moving to their fifth generation (5G) and with them, the world is moving to millimeter-wave (mmWave) frequencies. For many circuit designers, this means taking a hard look at their choice of printed circuit board (PCB) material to understand how well it will perform in 5G circuits and systems at mmWave frequencies. It means counting on that circuit material for a value of permittivity or dielectric constant (Dk) that will be the foundation of many new designs, counting on a circuit material supplier’s Dk measurements. But how reliable are those Dk measurements?

Standardized test methods from global trade organizations such as the IPC provide proven ways to measure material Dk. In fact, the IPC has 12 different methods for determining material Dk, with or without conductors. However, with these methods come variables and the Dk derived for a piece of material may change depending on those variables. Circuit material Dk is typically an anisotropic property and the Dk determined for the x-y plane of a circuit material will differ from the Dk determined through the z-axis or thickness of the same material. The Dk of a circuit material is also dependent upon frequency, following a trend of Dk value decreasing with increasing frequency. So, when comparing the Dk values of circuit materials, it is important to compare them at the same test frequency and through the same direction or axis of the material. It may even help to compare them when they are determined using the same Dk test methods, since the results from different test methods can vary.

So, which is the right Dk value, especially when that next circuit model or design depends on the right one? Results for the same piece of circuit material can vary according to the test method, even when that method delivers repeatable results. It is important to understand why this can happen, especially when applying the Dk of a selected material to design and fabricate 5G circuits at 24 GHz and beyond.

Test methods for determining the Dk of a material under test (MUT) have been developed for the raw dielectric material, by placing it in a test fixture, and for material with conductive metal cladding, by fabricating well-characterized reference circuits on the material and performing measurements on those circuits. Dk test methods based on the use of a test fixture, such as the clamped stripline test method detailed in IPC standard IPC-TM-650 2.5.5.5 Rev C, attempt to determine material Dk with minimal contributions from circuit fabrication processes, analyzing the dielectric material without copper cladding. The circuitry is provided by a test fixture with the dielectric layers of the test circuit provided by the MUT. In the clamped stripline test fixture, two pieces of raw circuit material are clamped around a metal circuit layer to form a stripline resonator.

While this Dk measurement approach attempts to remove the effects of fabricating a circuit from the process of determining the material’s Dk, it still has its share of variables, including any thickness variations in the two pieces of circuit material clamped around the stripline resonator circuitry. The ground-to-ground spacing of the resonant circuit will depend on the metal thickness of the resonator in the middle of the test fixture and the thickness of the two dielectric material sheets. While variations in the thicknesses of the materials forming the stripline resonator can impact the Dk value determined for that test fixture (which tends to be an average value for the range of thicknesses), the process of clamping the material pieces within the test fixture can also result in the entrapment of air within the test fixture, impacting the Dk value. Air, with its own Dk value of about 1, will result in lowering the Dk determined for a MUT, with a greater impact as more air is entrapped in the fixture.

For consistency and accuracy, the IPC test method clearly defines the types of hardware to be used in the test fixture, such as 3.5-mm-diameter coaxial cable and Type N coaxial connectors. Such interconnections will ensure the repeatability of high frequency measurements made on the test fixture with a vector network analyzer (VNA), but they will also set a high frequency limit on the measurements, at about 12 GHz. Dk measurements with a clamped stripline test fixture are possible at higher frequencies, but hardware changes are necessary, including thinner MUT and smaller diameter coaxial cables and connectors. The performance limits of the hardware, such as the insertion loss, phase accuracy and bandwidth, must also be clearly defined to ensure the consistency and accuracy of such Dk measurements at higher frequencies.

Refining the Process
Circuit based Dk test methods fabricate a reference circuit on a MUT; the behavior of the circuit is well enough understood that it can be used to determine the Dk of the MUT at different frequencies. However, variations in the process used to fabricate the reference circuit, such as microstrip or grounded coplanar waveguide (GCPW), can affect the accuracy of Dk determined from measurements of these reference circuits. For example, microstrip ring resonators are often used as reference circuits to determine Dk (see ROG Blog, “Searching for a Standard mmWave Dk Test Method,”) since a specific Dk and circuit dimensions, such as the ring radius, correspond to a specific resonant frequency. Microstrip ring resonators can determine Dk with high accuracy at lower frequencies, although the accuracy decreases with increasing ring resonant frequency. Microstrip ring resonators can even be used to determine other material qualities, such as loss or dissipation factor (Df), at lower resonant frequencies.

Microstrip transmission lines are also used as reference circuits to determine the Dk of a MUT, using the propagation characteristics of the microstrip, such as shifts in phase, on different dielectric materials. But because the fabrication of those microstrip lines on a MUT is very process dependent, the accuracy of the Dk can be affected by such parameters as the width of the conductors, the spacing between the conductors, and thickness variations in the dielectric material. The EM fields propagating along the transmission lines are partly in the dielectric material and partly in the air around it, so the determination of Dk from measurements on the transmission lines will need to account for the amount of the EM fields in the air.

Accurate, repeatable methods for determining circuit material Dk are essential for circuit designers moving into higher mmWave frequency bands. Fortunately, refinements to reference circuits such as ring resonators can reduce the impact of circuit processing variations and improve the accuracy of Dk determinations at higher frequencies (see the author’s IPC 2020 presentation, “Measuring the Impact of Test Methods for High Frequency Circuit Materials”). The presentation is posted on the Rogers Technology Support Hub homepage: www.rogerscorp.com/techub.  The use of edge coupling rather than gap coupling in microstrip ring resonators can improve Dk accuracy, and the use of GCPW feedlines in place of microstrip can reduce the effects of microstrip-based circuit variation. The test methods for determining circuit material Dk are many but with proper measurement discipline and control, repeatable, accurate Dk values can be determined even as circuit materials reach into the mmWave frequency range.

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.