Most RF boards are chasing thinner profiles. Thinner means lower module height, smaller parasitic effects, and tighter integration.

But some designs go the other way. They call for a 2mm finished thickness on a 2-layer stackup. This is not a design mistake—it is a deliberate choice. When mechanical rigidity, plated through-hole reliability, and thermal stability matter more than height, a thin board becomes the problem, not the solution.

We recently built a TMM6 board that falls into this category. 45.5mm × 90mm, 2 layers, 2mm finished thickness, 53 components, 38 vias, 2 nets. This article is not just about how the board was built. It is about a question that often gets overlooked: why a thick board can be more reliable than a thin one in certain RF applications, and what trade-offs that thickness brings.

The Board at a Glance

Base material: Rogers TMM6. 2-layer rigid PCB. Dimensions: 45.5mm × 90mm, ±0.15mm tolerance. Minimum trace and space: 5/8 mils. Minimum hole size: 0.35mm. No blind vias.

Finished thickness: 2mm. Outer copper: 1oz (35 μm). Via plating: 20 μm. Surface finish: HASL.

No top silkscreen, no bottom silkscreen. No top solder mask, no bottom solder mask.

100% electrical test before shipment. IPC-Class-2. Gerber RS-274-X.

The stackup is 35 μm copper / 1.905mm (75mil) TMM6 core / 35 μm copper.

A few details stand out: the 2mm thickness, the absence of solder mask and silkscreen on both sides, and the HASL finish on an RF material.

Why 2mm, Not Thinner

Standard FR-4 boards are typically 1.6mm. This board is 2mm—25% thicker than the norm. In RF design, added thickness changes several things.

Mechanical rigidity. A 2mm board resists flexing during assembly, handling, and long-term use. On a large board (this one is 90mm long), thin material can warp during reflow, causing component shift or solder joint cracking. A thick board is more stable.

Plated through-hole reliability. One of TMM6's core positioning points is high plated through-hole reliability. The thicker the board, the higher the aspect ratio, and the harder it is to achieve uniform plating. But once a thick board is properly plated, the barrel is mechanically stronger and less likely to crack under thermal cycling. This board has 38 through-hole vias, plated to 20 μm.

Thermal stability. A thick core has more thermal mass, which helps spread heat in power amplifier applications. TMM6's thermal conductivity is 0.72 W/mK, and combined with 2mm thickness, heat dissipation is better than a thin board.

The cost. Thick boards are harder to drill, more expensive in material, and heavier. If a design does not need these properties, 2mm is waste.

TMM6: High Dk Without the Processing Headache

TMM6 is part of the Rogers TMM family, alongside TMM10 and TMM10i. Its dielectric constant is 6.0, sitting between FR-4 (~4.3) and TMM10 (9.2).

What Dk 6.0 means. High Dk shrinks circuit dimensions. In filters, couplers, and matching networks, the same electrical performance can be achieved in a shorter physical length. For space-constrained RF designs, that is real value.

What Df 0.0023 means. At 10GHz, this is lower than FR-4 but higher than pure PTFE. For power amplifiers, combiners, and filters, this loss level is acceptable.

Thermal coefficient of Dk: −11 ppm/°K. This is unusually stable. As temperature changes, the dielectric constant barely moves, meaning filter center frequencies and impedance values stay consistent. For satellite communication and outdoor RF equipment that see wide temperature swings, this stability is a real advantage.

CTE matched to copper (18/18/26 ppm/K). This is why TMM6 is positioned for high plated through-hole reliability. The dielectric and the copper plating expand at nearly the same rate, reducing the risk of plating cracks and delamination under thermal cycling.

Processability. TMM6's other selling point: it does not require the special processes that PTFE materials demand. No sodium naphthalate treatment, no unusual drilling parameters, and all common PCB processes work. For manufacturers, that means shorter lead times and lower processing cost.

Why No Solder Mask, No Silkscreen, and HASL

This board has no solder mask and no silkscreen on either side, and it uses HASL as the surface finish.

No solder mask is common in RF and microwave designs. Solder mask has dielectric properties that can shift impedance and affect tuning, especially at higher frequencies. Removing it eliminates that variable and exposes copper directly for consistent RF performance. No silkscreen follows the same logic—markings serve no electrical purpose, and on an RF board they can sometimes interfere with tuning or inspection.

The HASL choice is worth discussing. On high-frequency materials, ENIG or immersion gold are more common because they are flatter and suit fine-pitch components. But this board uses HASL—hot air solder leveling. HASL is lower cost and offers good solderability. This board has 15 top-side SMT pads and 21 through-hole pads, and HASL supports standard assembly. If the design does not need ENIG's flatness or gold's wire-bonding capability, HASL is a reasonable choice.

Here is a trade-off worth stating clearly: HASL's surface flatness is not as good as ENIG. For fine-pitch SMT components, HASL can introduce soldering defects. This board's minimum trace is 5/8 mils, but the pad spacing is not fine enough to require ENIG, so HASL works. If a future design needs finer spacing, the surface finish may need to be reconsidered.

38 Vias, 2 Nets: What This Board Is Doing

53 components, 36 pads, 38 vias, 2 nets.

High via count with low net count is a pattern we have seen before. It usually means a grounded structure—via fencing, ground plane stitching, or a shielded transmission line. These vias are not routing signals; they provide grounding and shielding.

21 through-hole pads and 15 top-side SMT pads indicate mixed assembly. No bottom SMT pads, consistent with the bottom side serving as a ground plane or mounting surface.

When TMM6 Is Necessary

When high Dk and miniaturization are hard requirements. Dk 6.0 shrinks the physical size of filters, couplers, and matching networks. If space is constrained, this is real value.

When plated through-hole reliability is critical. CTE matched to copper, plus the mechanical stability of a thermoset resin, gives TMM6 an advantage in high-reliability applications. The combination of a thick board and high-aspect-ratio vias demands more from plating quality, and TMM6's material properties help.

When temperature stability matters. A thermal coefficient of Dk of −11 ppm/°K means consistent performance across wide temperature ranges. Satellite communication, outdoor equipment, and automotive radar benefit.

When processability cannot be compromised. TMM6 does not require PTFE's special processes, so lead times are shorter and costs are more controllable. For designs that do not want to pay for material processing complexity, this is a real advantage.

When You Should Not Use TMM6

If your frequency is low and loss requirements are not strict. At low frequencies, FR-4 may be entirely sufficient. TMM6's high Dk and low loss would be overkill.

If you do not need 2mm thickness. A 2mm board brings processing difficulty and cost. If module height allows, a standard-thickness board is easier to process and cheaper.

If you need the lowest possible loss. TMM6's Df is 0.0023—lower than FR-4 but higher than pure PTFE. If loss is the primary constraint, a PTFE-based material may be more suitable.

If cost is the first constraint. TMM6 is a ceramic-filled thermoset material, priced above FR-4. In cost-sensitive applications with modest performance requirements, there is no reason to pay for performance you will not use.

The Bottom Line

TMM6 is not a general-purpose RF material. It is a targeted solution: it earns its premium when high-Dk miniaturization, plated through-hole reliability, and temperature stability all matter at the same time.

The 2mm thickness is not thick for its own sake. It serves mechanical rigidity, thermal stability, and via reliability. If your design does not need those, a thin board with FR-4 may be the more sensible choice.

If you are working on an RF or microwave design and need to balance high Dk, low loss, and plated through-hole reliability, we would be glad to review your requirements. Material selection should start from your frequency, power, and mechanical constraints—not from which material sounds best.

Contact us to discuss your next RF PCB project.