The capability extract shows in thirty seconds what runs as standard and what carries a surcharge.
See capabilities →Send the files and an engineer reviews them against your build — free, before you enquire.
To the data check →Above roughly a gigahertz the laminate stops being a mechanical carrier and becomes part of the circuit. Dielectric constant sets the geometry, loss tangent sets the budget, and the copper roughness nobody specified sets the rest.
A microwave board is defined by its laminate. PTFE-based materials give the lowest loss and the lowest dielectric constant; ceramic-filled variants trade some loss for a higher Dk and a smaller structure; hydrocarbon composites sit between the two and process much more like FR-4. Which one is right follows from frequency, loss budget, and how much of the board actually needs to be RF.
That last point matters commercially. A full PTFE build is expensive and awkward to process, while a hybrid that puts one microwave layer over an FR-4 core gives most of the performance for a fraction of the cost. Most boards we build in this class are hybrids, and the conversation usually starts by working out which layers genuinely need the good material.
Df is what you are buying. At 10 GHz, moving from 0.004 to 0.0015 is a few dB across a long feed line — the difference between closing a link budget and not.
Trace width for a given impedance falls as Dk rises. High-Dk ceramic-filled laminates shrink an antenna or a coupler; low-Dk PTFE widens it and cuts loss.
One PTFE layer bonded to an FR-4 core carries the RF while the rest of the board stays conventional — far cheaper than a full PTFE stack, and usually sufficient.
| Feature | Specification |
|---|---|
| Material | Rogers, Taconic, Arlon, Isola — PTFE, ceramic-filled and hydrocarbon |
| Dk | 2.2 – 10.2 (2.2, 3.0, 3.5, 6.15 and 10.2 common) |
| Df at 10 GHz | 0.0009 – 0.004 |
| Layers | 1 – 8 (hybrid builds to 12) |
| Final thickness | 0.2 – 2.0 mm |
| Copper | 12 – 70 µm, rolled or low-profile |
| Surface finish | ENIG, immersion silver, OSP — ENIG standard |
| Typical use | Radar, satellite, 5G and mmWave, RF test equipment |
The usual build for this line. Anything outside it is normally still possible — ask and we will confirm.
Above about 20 GHz almost every column moves. Talk to us before layout rather than after.
| Parameter | Standard | Advanced | Note |
|---|---|---|---|
| Frequency | to 40 GHz | to 77 GHz | Above 40 GHz roughness and tolerance dominate |
| Dk tolerance | ± 0.05 | ± 0.025 | Tighter grades cost more and lead longer |
| Line / gap | 100 µm | 75 µm | Etch tolerance moves impedance directly |
| Impedance tol. | ± 10 % | on review | Coupon and measured report; tighter after technical review |
| Copper profile | Standard | Rolled / VLP | Roughness adds loss above ~10 GHz |
| Layers | 1 – 4 | 6 – 8 | PTFE bonding limits practical layer count |
| Build | Full PTFE | Hybrid PTFE / FR-4 | Hybrid is usually the cheaper route |
| Plated holes | Standard | Back-drilled | Stub removal matters above ~10 GHz |
Guide values for pre-selection. Binding values are confirmed per project against your data.
The points below are what most often comes back as a query after data review. Settling them before layout freeze saves a loop.
Rogers is a manufacturer with dozens of grades whose Dk spans 2.2 to 10.2. The part number belongs in the stack-up drawing — substituting a grade changes every trace width on the board.
Surface roughness adds conductor loss that no amount of laminate quality recovers. Rolled or very-low-profile copper is the fix, and it has to be asked for.
Impedance on a microwave board is measured, not assumed. A coupon in the panel costs nothing and is the only record of what was actually built.
The adhesive between PTFE and FR-4 has its own Dk and its own thickness tolerance, and it sits inside the field. It belongs in the stack-up model, not outside it.
PTFE for the lowest loss and where size is not critical; ceramic-filled where a higher Dk shrinks the structure and the loss budget can absorb it. Below about 6 GHz a good hydrocarbon laminate often beats both on cost with no meaningful penalty.
Yes, and it is the usual answer. A hybrid puts the RF on one or two microwave layers over a conventional core. The bonding step needs planning, but the saving over a full PTFE stack is large.
Because it is flat and repeatable. The nickel underlayer does add loss at very high frequencies, which is why immersion silver appears above roughly 20 GHz — at the cost of a shorter shelf life and more careful handling.
Above roughly 10 GHz, usually yes for any through-hole in a signal path. The unused barrel is a stub that resonates, and removing it is cheaper than redesigning around it.
Send us your data or a rough sketch. We check feasibility, name the cost drivers and answer within one working day.