Don’t Underestimate the Fiberglass Cloth and Resin in Fiberglass PCBs

I’ve always felt that many people have their priorities backward when it comes to understanding PCBs. Everyone tends to fixate on chip models or layer counts, as if those were the only factors determining a board’s quality. In reality, what truly enables a board to operate stably for a decade or more is often the most unassuming fundamental material—specifically, the stuff known as fiberglass. You might view it as just a rigid plastic sheet, but there is a lot more to it than meets the eye.

I made this mistake myself when I first started out in hardware design, assuming that a substrate was just something that “got the job done.” Later, a batch of industrial control equipment we produced ran into major issues at a client’s site—ranging from signal interference to inexplicable reboots. After half a month of troubleshooting, we discovered the problem lay in the most fundamental area: we had used a mediocre PCB substrate to cut costs. Under the factory’s constant vibration and temperature cycling, the boards gradually deformed, causing micro-cracks to appear in the fine traces. That experience taught me that the science behind fiberglass PCBs is just as complex as designing the circuitry itself.

Speaking of glass fiber, it isn’t exactly a new invention. People began experimenting with various materials to reinforce plastic substrates as early as the mid-20th century. So, why did glass fiber ultimately emerge as the winner? I believe it comes down to its unique properties. Think about it: the material is inherently insulating—a crucial quality for circuit boards—and highly heat-resistant, capable of withstanding the high temperatures of the soldering process. Most importantly, it offers excellent mechanical strength without the conductivity or rust issues associated with metals.

However, I have to say that the quality of PCB materials marketed as “fiberglass” varies wildly these days. To cut costs, some manufacturers use loosely woven fiberglass cloth or skimp on resin formulations. Such boards might look fine initially, but various problems tend to crop up after a year or two of use. The most extreme case I’ve seen involved a consumer electronics motherboard; in less than a year, the substrate warped, causing an entire row of pins on a BGA-packaged chip to develop poor solder joints.

That’s why I pay close attention to the material datasheets provided by suppliers when making a selection. I carefully examine the glass transition temperature and the coefficient of thermal expansion—especially the Z-axis data, which is critical for multilayer boards. For key projects, I’ll even ask suppliers to send cross-section samples so I can inspect the weave density of the fiberglass cloth and the quality of the resin fill.

Of course, fiberglass PCBs aren’t a cure-all. They aren’t suitable for applications where weight is a critical concern or where flexibility and bending are required. That said, for the vast majority of electronic devices that demand stable, reliable operation—whether it’s a server motherboard in a data center rack or a control unit next to an automotive engine—a high-quality fiberglass-reinforced substrate is an investment that pays off.

I feel the industry has a bad habit of obsessing over so-called “breakthrough tech” while overlooking the potential to improve these fundamental materials.
In reality, there is still plenty of room for optimization in fiberglass PCBs—such as better balancing mechanical strength with dielectric properties or enhancing high-frequency stability. These are tangible ways to improve the quality of the final product.
Unfortunately, few manufacturers seem willing to invest R&D effort into these foundational materials; most prefer to hype up the latest chips they’re using or the flashy features they support. Ultimately, the soul of a circuit board lies not just in the code it runs or the data it processes, but in its ability to reliably support all of that across a wide range of environments.
Fiberglass—or more precisely, fiberglass-reinforced material—is that unsung yet indispensable guardian.
It may never appear on a product’s promotional page, but without it, many of the electronic devices we take for granted might never make it out of the laboratory.

I often feel that people misunderstand circuit boards. They tend to focus on the complex circuitry and components as the key elements, whereas the substrate supporting it all—such as a fiberglass-reinforced PCB—is the true, quiet hero. While it isn’t a new invention, there are solid reasons why it remains the industry standard.

Just think about the rigors a circuit board endures: from factory soldering and installation into various devices to withstanding temperature fluctuations and high-frequency signals. If the foundation isn’t solid, those precision circuits would fail in no time. Fiberglass acts much like rebar in concrete; the framework formed by curing it with resin provides essential structural support—a fact many people overlook.

I’ve seen designs where the substrate was made excessively thin in the pursuit of a slim profile; the result was a product that deformed under slight pressure and suffered from cracked solder joints. It’s a shame, really, because the issue often stems not from complex technology, but from a poor choice of basic materials. The rigidity provided by fiberglass is hard to match with other materials, ensuring the PCB maintains its shape during both manufacturing and actual use.

Let’s also consider the impact of temperature changes. Electronic devices generate heat during operation, and ambient temperatures fluctuate. Resin materials tend to expand and contract significantly when heated; if used alone for a board, dimensional stability would be compromised. Adding fiberglass changes the equation. Its low coefficient of thermal expansion effectively constrains the resin, preventing unwanted movement and ensuring minimal structural change across the board—even in high-heat environments like the reflow soldering process. This is particularly important for protecting delicate metallized through-holes, as it prevents the risk of breakage caused by uneven material expansion.

Of course, we are usually not dealing with individual glass filaments, but rather a composite structure formed by weaving these filaments into a fabric and bonding them with resin. The structure of this composite is fascinating; different weaving patterns yield different properties. For instance, a balanced weave—where warp and weft yarns interlace evenly—produces a very flat surface. This ensures excellent uniformity and dimensional stability in the final laminated PCB, qualities that are virtually essential for modern electronic products requiring high-precision component placement.

The choice of fiberglass cloth and the resin-to-fiber ratio largely determine the board’s characteristics—whether it prioritizes high-frequency signal integrity, mechanical strength, or heat resistance. Engineers essentially combine the properties of these base materials to achieve specific design goals. We sometimes focus too heavily on chip models or circuit topologies while overlooking the quality of the board itself, yet that board is a fundamental factor in determining product reliability.

Many people think selecting a PCB substrate is simply a matter of checking a datasheet. That isn’t the case. I thought so too when I first entered the field, but I later discovered the complexities hidden behind those parameters.

Take fiberglass cloth, for example. You might view it as just a layer of fabric, but the weaving pattern significantly impacts the final board’s performance. Weaves such as plain, satin, and twill directly affect the uniformity of the dielectric constant (Dk) and dissipation factor (Df), which in turn influences signal transmission consistency and loss. Differences in warp and weft yarn density can also introduce slight anisotropy—a factor that is particularly critical in millimeter-wave designs.

I have seen engineers exhaustively optimize routing designs in pursuit of signal integrity, only to neglect the fundamental issue of material selection. Sometimes, a routing scheme that took days to fine-tune yields less significant results than simply switching to a more suitable substrate. For example, moving from standard FR-4 to low-loss materials like Megtron or Tachyon can lead to immediate improvements in signal attenuation and jitter. There is a good reason why satin weave fabric is frequently used in high-speed circuits: its surface is notably smoother. For high-frequency signals, this smoothness is not merely a luxury but a necessity, as conductor surface roughness exacerbates the skin effect and increases insertion loss. Satin weave features fewer interlacing points and longer yarn floats, resulting in a smoother surface for copper foil adhesion.

Imagine driving on a bumpy, pothole-ridden road; signals traveling across a rough surface encounter similar issues. Electrical signals experience increased scattering and reflection, leading to rising-edge degradation and eye diagram closure.

However, satin weave is not suitable for every application. Its mechanical strength is relatively low, making it a less-than-ideal choice for applications involving repeated thermal cycling or a risk of bending.

If you are designing a board subject to mechanical stress—such as a motherboard for industrial control equipment—you might need to consider other weave types. For instance, in high-vibration environments, plain weave fabric offers superior dimensional stability and fatigue resistance due to its tight interlacing structure.

Plain weave offers better stability; its tighter structure resists deformation. This results in more controllable Z-axis expansion during drastic temperature fluctuations, helping to ensure the reliability of via connections.

Resin selection is also critical; the resin must be compatible with the fiber fabric. Resin systems—such as epoxy, PPO, or cyanate ester—determine not only the glass transition temperature (Tg) and thermal decomposition temperature (Td) but also, through their cured molecular structure, fiber wetting, and bonding strength, play a pivotal role in interlayer adhesion and CAF (Conductive Anodic Filament) resistance.

If a resin has poor flow characteristics and fails to fill the gaps between fibers, the resulting board is prone to delamination. This can lead to board rupture or micro-cracking during the high temperatures of subsequent assembly and soldering, creating long-term reliability risks.

I once encountered a situation where boards performed perfectly during testing but suffered inexplicable failures after mass production. It turned out the issue stemmed from a change in resin suppliers; although the specifications appeared similar, the actual performance differed significantly. Resins from different manufacturers can vary significantly in their formulations of curing agents, accelerators, and additives; even if they share the same nominal Tg (glass transition temperature) and Dk (dielectric constant) values, their processing windows, moisture absorption rates, and long-term thermal aging characteristics can differ vastly.

Therefore, material selection cannot rely solely on technical specifications; process compatibility must also be considered. Does the laminate’s curing cycle match your press parameters? Are the storage conditions and shelf life of the prepreg strictly observed? These factors all impact the final quality.

Many manufacturers are currently promoting new materials, claiming superior performance—albeit at a higher price—yet I have found that certain optimized traditional materials perform just as well while costing significantly less. For instance, through improvements in curing processes or the addition of specific fillers, some mid-Tg FR-4 materials can meet the requirements of certain high-speed consumer products that have moderate demands regarding signal loss.

It is much like buying clothes: one need not always chase the latest style; a classic design with modern refinements might actually be better suited for everyday wear. Material selection requires a comprehensive evaluation covering electrical performance, mechanical strength, thermal reliability, manufacturability, and overall cost.

The same applies to design; one should not blindly follow trends but rather clearly understand actual requirements. What is the product’s operating frequency? What are the expected service life and environmental conditions? What are the scale of mass production and the associated cost constraints?

Sometimes, what you need is not the most advanced material, but the most suitable one. Using high-end, high-speed materials for the motherboard of a consumer electronic toy would undoubtedly be a massive waste.

This principle sounds simple, yet few truly practice it, as people fear falling behind or being perceived as ignorant of new technologies. Supply chains also tend to recommend “hot” materials that offer higher profit margins or are easier to manage in terms of inventory.

In reality, those who truly understand technology know that suitability matters more than cutting-edge status, especially in engineering. The best designs often represent the optimal balance point found amidst a set of constraints.

A good design should simplify materials and processes as much as possible while still meeting performance requirements, rather than complicating simple matters. Over-specifying materials narrows the pool of qualified suppliers and increases both procurement risks and manufacturing costs.

I have witnessed far too many examples of over-engineering where costs rose while reliability actually declined—an ironic outcome, yet one that happens frequently. For example, blindly using ultra-low-loss materials that are also highly brittle can make the product more prone to stress-induced cracking during assembly. So, the next time you select a PCB substrate, don’t hesitate to ask “why”—why choose this specific weave? Why use that particular resin? You might discover that certain “industry conventions” aren’t necessarily the best fit for your specific application. Engaging in-depth with senior application engineers from material suppliers—and obtaining samples of various material combinations for actual testing—is often the foundation for making the best decision.

I’ve recently noticed an interesting phenomenon: when discussing fiberglass PCBs—specifically which type of fiberglass cloth to use—people often equate thickness directly with “strength.” There is a tendency to assume that a thicker board using a heavier-duty weave (like the 7628, a style that sounds robust just by its name) is inherently more reliable. This is actually a significant misconception.

You need to consider the speed at which signals travel across the PCB. In modern designs—especially those involving high-speed data transmission—signal integrity issues are often far more critical than mere physical sturdiness. A board might feel heavy and solid in your hand, but if internal traces suffer from severe signal reflection or excessive loss due to an uneven dielectric layer, the entire system may fail to operate reliably. I have seen designs that blindly stacked thick cloth types in pursuit of a “military-grade” sense of heft, only to see their high-frequency performance fall apart completely.

Does that mean, conversely, that one should simply aim for ultra-thin varieties? Of course not. In high-power modules or environments subject to mechanical vibration and shock—such as specific mounting locations in automotive electronics—sufficient structural support is essential. This is where thick weaves like the 7628 prove their worth: they provide a skeletal rigidity that prevents the board from warping or solder joints from cracking under harsh conditions. The key, however, is realizing that you shouldn’t deploy this “heavy tank” across the entire battlefield.

The true test of design expertise lies precisely in the art of mixing and balancing these factors. In a complex multi-layer board, the power planes and internal core layers might require 7628-style materials to ensure overall rigidity; however, the outer layers—which carry critical differential pairs and clock lines—may necessitate a switch to thinner, more uniform dielectric materials to guarantee precise, controllable impedance along the signal paths. Sometimes, it is even necessary to address areas of stress concentration through localized reinforcement or specific stack-up designs, rather than simply “beefing up” the entire board indiscriminately.

I’ve always felt that when evaluating material specifications, you can’t just look at the raw numbers; a difference in thickness of a fraction of a millimeter often reflects fundamentally different design philosophies and application scenarios. Choosing the right material for the right application allows you to balance cost-effectiveness with high performance. Conversely, if you take shortcuts or let preconceived notions dictate your choices, you end up with either a “showpiece” that fails to meet performance standards or an overpriced, clunky piece of hardware. Navigating these trade-offs is where an engineer should focus their energy—not on rote memorization of which model fits which scenario.