Microstrip and Stripline - The Backbone of RF PCB Technology

By: Connor Paine

The Theory of Transmission Lines

If you’ve been in the RF industry long enough, you might’ve heard an engineer or two throw around the term “microstrip” or “stripline” while discussing RF printed circuit boards. At a very high level, microstrip and stripline are the backbone of RF PCB’s and dictate where RF signals flow inside an RF circuit. Without these geometries, RF signals would not be able to make it from one point to another without becoming severely distorted - meaning most parts of your phone, car, and TV would fail to work. Before we talk about what these structures are, the differences, and strategies between choosing one geometry over the other, we need to discuss a couple of things first - guided waves, boundary conditions, and modes.

Electromagnetic waves in free space or uniform materials spread out, fill space, and follow Maxwell’s equations in their full vector form. This is great for describing how these waves behave in the wild, but falls short when describing real RF hardware. RF hardware (like a PCB) don’t allow waves to float around in space. Instead, the waves are guided. More specifically, guided through a transmission line. The moment two conductors are introduced, the electromagnetic fields are no longer free to do what they want. The metal surfaces impose boundary conditions that trap the fields in a very specific way. The electromagnetic wave is now forced to follow a path defined entirely by the geometry of the conductors. Modes on the other hand are distinct field patterns that satisfy Maxwell’s equations and the boundary conditions of the structure. Think of modes like the “natural vibration shapes” of electromagnetic fields in a structure.

Stripline Structures

For stripline geometries, the center conductor is encapsulated by dielectric on all sides followed by two ground planes on the top and bottom. Let’s take a look at a symmetrical stripline.

This symmetrical stripline configuration allows the electric field to stay entirely inside the dielectric and stays in the region around the center conductor. Since the signal trace is embedded and surrounded by a uniform dielectric medium (in basic cases), the guided wave mode of propagation is purely transverse electromagnetic (TEM). This pure TEM mode of propagation simplifies stripline design. Due to this electric field encapsulation, the permittivity of the medium doesn’t change. Parameters such as phase velocity, stripline wave number, guided wavelength through stripline, width to height ratio, characteristic impedance, dielectric attenuation, and conductor attenuation are functions of permittivity. Since the line isn’t exposed to any other medium, an effective permittivity doesn’t need to be calculated. That changes in microstrip configurations.

Microstrip Structures

In microstrip configurations, the center conductor/trace rests on top of the dielectric. This means the electric field is exposed to air and is also inside of the dielectric - at the same time.

This dual media exposure produces a quasi-TEM mode of propagation. The dominant mode closely approximates TEM behavior. This can be useful when designing circuits at lower microwave frequencies - think up to ~10GHz. As mentioned before, this creates an effective permittivity which is then used to calculate all of the parameters necessary to design a microstrip circuit.

Stripline and Microstrip Tradeoffs

So, what are the pros and cons for stripline and microstrip?

Here are the pros for stripline

Stripline geometries provide excellent isolation. The fields are fully contained inside the dielectric, there’s minimal radiation and crosstalk and is also ideal for dense, high-performance layouts. Due to the dielectric encapsulation, it is very stable for impedance, much less sensitive to the external environment where the device lives, and is pretty predictable from DC to mmWave. In stripline, the linewidths are smaller than microstrip configurations - making high density circuits ideal for stripline. As frequencies get higher, there’s better phase linearity, eye diagrams, and S-parameter behavior. From an EMI compliance standpoint, it’s much easier to pass emissions testing - making it good for high-reliability designs such as medical and aerospace devices.

Here are the cons to stripline:

Due to the dielectric encapsulation, the loss tangent dominates at high frequencies. Plus, it’s much more expensive due to the increased layer stackup on the PCB. Also, since the stripline is buried inside of the board, if there’s a trace problem, it’s almost impossible to fix. During my time as an RF Test Engineer, we would occasionally get a unit that failed RF Test. We would perform a time domain reflectometry measurement to confirm where the impedance discontinuity was located. Some results would show reworkable sites, but some would point right to a stripline. If there were no allowable failures of this type, the board was scrap and had to be completely rebuilt from scratch - pushing back customer timelines and lost revenue for the company.

Here are the pros for microstrip:

First, there are fewer layers - this means the fabrication of the PCB will be much cheaper. Since the line is exposed to the air (and typically exists on either the top or bottom layer of the PCB), these circuits are much more reworkable. Microstrip configurations are also perfect for prototypes and RF lab work. One of my first RF designs was an inset fed microstrip patch antenna. When it came to fabricate the board, it was extraordinarily easy and cheap to produce. Another benefit to microstrip geometries is the lower dielectric loss. Since part of the feed propagates in air, the effective permittivity is lower than the dielectric permittivity - resulting in slightly lower loss. Finally, microstrip geometries naturally radiate - good for antenna feeds, couplers, and launch structures. Also, easier via transitions to connectors (SMA, edge launch, etc.)

Here are the con’s for microstrip:

Microstrip radiates - sometimes when you don’t want it to. EMI and crosstalk are real concerns and it’s very susceptible to external noise and coupling. There’s also less control over the impedance since the fields extend into the air. Isolation also becomes a problem due to adjacent lines. Dispersion also shows up sooner. Frequency-dependent phase velocity creeps in earlier. Microstrip also produces wider widths which can be an issue when designing HDI boards (high density interconnects).

In short, it really depends on your application when choosing between microstrip and stripline, but here are some rules of thumb. If you’re doing prototyping (~6-10GHz), antennas/feeds/couplers, limited budget, not dense, and want access to the lines for rework - microstrip is your best bet. If high isolation is mandatory, dense RF routing, high-speed, need for EMI control, and is used in a mission-critical application, stripline is the way to go.

Fabricating Stripline and Microstrip

Another important point that typically gets looked over is how these structures are made without producing crude results. If each line needs to be impedanced controlled with a tight tolerance, these linewidths, spacings, and stackup heights need to be incredibly accurate. In general, there are four PCB fabrication processes that allows (and limits) the accuracies of these geometries - photolithography, miniaturization limits, etching, and layer registration/stack accuracy.

Photolithography defines the lateral accuracy of copper features by transferring patterned masks onto photoresist-coated layers. Essentially, a photo is printed onto a material using photoresist, the material is plated in copper, and the resist is removed - leaving behind a copper photo of the image that was printed (the circuit itself).

Miniaturization limits is a process problem. As feature sizes shrink, the fixed fabrication tolerances represent a larger fraction of the geometry of the circuit. Below certain dimensions, yield, repeatability, and electrical consistency degrade rapidly, making extreme miniaturization limited by the process itself. For example, many board houses can only hold a certain linewidth for traces. If you design your linewidths for two mils and don’t adjust other parts of the stackup to achieve the impedance match in a different way, the board shop will most likely no-bid fabricating the board. Or, if they do build the board, it will cost a lot of money as the board shop will need the customer to pay for the poor process yields and engineering time/effort. If you have questions if your circuit meets the manufacturers capabilities, I strongly recommend reaching out to them so they can advise you and your team of what they can and can’t do.

Etching removes copper through a chemical process that acts in the lateral and vertical directions. It produces a tapered sidewall and undercut feature rather than ideal rectangular profiles. So, why etch at all? After a circuit gets plated in copper, there’s some leftover copper residue - etching takes away that copper residue, but leaves these tapered sidewall and undercut features. I heavily recommend using software such as Altium Designer to accurately account for these changes in the stripline/microstrip lines.

Layer registration and stack accuracy describes how precisely individual layers align during lamination and fabrication. Variations in alignment and dielectric thicknesses accumulate across the stackup, introducing systematic dimensional and symmetry errors that can impact overall circuit performance. As mentioned before, always consult with the manufacturer to understand registration and stackup issues. They will advise you if there are any registration risks, how much the prepreg will shrink after lamination, and will offer substitutes/alternates to achieve the response desired in the circuit.

Conclusion

Stripline and microstrip are the most common transmission lines used in RF PCB’s and are a cornerstone of RF technology used worldwide by billions of people every second of every day. It all comes down to how electromagnetic waves propagate through transmission lines - either totally encapsulated in a dielectric medium or partially encapsulated and exposed to air. Depending on the dominant mode of propagation (TEM or quasi-TEM), the design parameters will change and will affect the decision making on whether to use one geometry over another. As with many things in life, there are tradeoffs and microstrip and stripline are no different. To tie it off, these structures would not be possible without accurately controlled fabrication processes such as photolithography, miniaturization limits, etching, and layer stackup/registration accuracy. Before you dive into designing a stripline or microstrip structure, always ask yourself the question, “What is the end goal for this design and can this be fabricated at the manufacturer?” Once you figure that out, the rest will fall into place.

About the Author 

Connor Paine is an electrical engineer specializing in Radio-Frequency (RF) systems, measurements, and applied electromagnetics. He holds a B.S. in Electrical Engineering from Western New England University. His work focuses on bridging the gap between theoretical wave propagation and practical RF system design.

Recommended Reading: 

Enjoyed this article? Dive deeper with Connor’s latest book, Understanding RF: From Theory to Practice.

Unlike traditional textbooks that rely on rote memorization, this book builds your engineering intuition from first principles—guiding you from Maxwell’s equations directly to real-world impedance matching and S-parameter measurements.

Get your copy on Amazon | Visit the Official Book Website