The Rogers “5000 series” — properly the RT/duroid 5880 family — is PTFE reinforced with randomly oriented micro glass fibres: the lowest-loss, lowest-Dk laminate family in mainstream use, and the material of record for satellite payloads, defence radar and millimetre-wave front ends. Circulating model lists pad it with products that do not exist and, more dangerously, describe it as thermally stable in the Z axis. This article presents the family as the datasheets define it, separates the real relatives from the fabricated model numbers, and states the one design constraint — Z-axis expansion — that governs every multilayer build on it.
What the Family Is — and Is Not
RT/duroid 5000 laminates are PTFE with randomly oriented micro glass fibres. The construction matters in both directions:
- What it buys: the lowest dissipation factors available at Dk 2.20–2.33 (0.0009–0.0012 at 10 GHz), isotropic behaviour from the random microfibre (no woven-glass Dk variation), near-total moisture immunity (0.02%), and a negative TCDk that keeps Dk drift small across temperature.
- What it is not: not ceramic-filled — the filler story belongs to the RO3000 and RO4000 families. And not thermally stable in Z: the same unreinforced PTFE structure expands at 237 ppm/°C through thickness, fourteen times copper, which is the family’s defining engineering constraint.
The individual members are covered in depth elsewhere — the 5880 advantages article, the 5870 comparison and the 5880 vs 5870 head-to-head — so this article stays at family level.
The Verified Model List
| Model | Dk | Df @ 10 GHz | Defining property |
|---|---|---|---|
| RT/duroid 5870 | 2.33 ± 0.02 | 0.0012 | The slightly higher-Dk sibling; same construction and process |
| RT/duroid 5880 | 2.20 ± 0.02 | 0.0009 | Lowest-loss mainstream laminate; the default for mmWave and satellite |
| RT/duroid 5880LZ | 1.96 ± 0.04 | 0.0019 typ. | Ceramic-filled for density 1.4 g/cm³ and Z-CTE 41.5 — the weight and reliability variant |
| RT/duroid 6002 | 2.94 ± 0.04 | 0.0012 | Often listed beside the 5000s; phase-stable PTFE for filter and antenna work |
| RT/duroid 6002HT | 2.94 ± 0.04 | 0.0012 | 6002 with raised thermal conductivity for power stages |
The 6000-line relatives extend the same PTFE chemistry upward in Dk — RT/duroid 6006 at 6.15 and 6010LM at 10.2 — for compact antennas and filters; they are usually quoted alongside the 5000s because the processing and the loss behaviour are of the same character. The lightweight outlier 5880LZ trades higher Df and moisture absorption for density and a six-times-better Z-CTE than 5880 — the trade is quantified in the 5880 laminate guide.
Three Entries in the Circulating Model List That Fail Verification
| Listed as | Listed parameters | What it actually is |
|---|---|---|
| Rogers 6202 | Dk 2.94, Df 0.0017 | No RT/duroid 6202 exists. The description — “6002 with better high-temperature stability” — matches the real product 6002HT. |
| Rogers 6302 | Dk 3.02, Df 0.0018 | No such product. Rogers’ PTFE line has no 3.0-class RT/duroid; the nearest real step up from 6002 is 6006 at 6.15. |
| Rogers 6500 | Dk 3.48, Df 0.0030 | No such product — and the parameters are a fingerprint: Dk 3.48 with Df 0.0037-class loss is RO4350B, a hydrocarbon-ceramic laminate from a different family and a different process route entirely. |
The 6500 entry is the instructive one: someone lifted a hydrocarbon laminate’s numbers and gave them a PTFE-family part name. A buyer specifying “6500” would receive, at best, an arbitrary substitution — with a different bonding process, different hole treatment and different reliability behaviour.
The Z-Axis Reality Check
Family-level descriptions often claim “low CTE, especially stable in the Z axis, suitable for multilayer designs.” For 5880 and 5870 this is backwards in the axis that matters: Z-axis CTE is 237 ppm/°C for 5880 and 231 for 5870 — fourteen times copper. Across a −55 to +125 °C range the plated barrel of a 1.6 mm board strains roughly 4.4%.
Multilayer 5000-series boards are therefore built to rules, not to optimism:
- Plated copper with high elongation (≥20%) and at least ~1.2 mil wall thickness
- Aspect ratio kept at or below about 6:1
- Plasma or sodium-naphthalene hole treatment — permanganate desmear does nothing to PTFE
- Bonding by FEP or low-flow films at 260–300 °C (or low-temperature prepreg routes for hybrid FR-4 stacks)
When those rules are constraining rather than enabling — high layer counts, dense PTH, automotive-grade cycling — the escape hatches are 5880LZ (Z-CTE 41.5) or the RO4000 hydrocarbon family (Z-CTE 32), both documented in the 5880 multilayer guide.
Choosing Within the Family
| If the design needs… | Choose | Why |
|---|---|---|
| Absolute lowest loss | 5880 | Df 0.0009; the loss floor of the mainstream market at any Dk |
| Slightly tighter structures at near-identical loss | 5870 | Dk 2.33 shortens guided wavelength ~6% at a Df penalty of a third of a unit |
| Low weight and better Z reliability | 5880LZ | Density 1.4 vs 2.2, Z-CTE 41.5 vs 237; pays in Df 0.0019 and moisture 0.22% |
| Phase-critical filters at ~3.0 | 6002 / 6002HT | Tight tolerance PTFE with proven temperature-stable phase; HT variant for hot environments |
| Compact antennas at PTFE-class loss | 6006 / 6010LM | Dk 6.15 / 10.2 within the same processing family |
Two selection notes that save redesigns. First, guided wavelength at Dk 2.20 is long — 21.9 mm at 10 GHz — so the 5000 family buys loss with area; when the board cannot afford the area, the honest alternatives are a higher-Dk PTFE (6006, 6010LM) or a hydrocarbon laminate, not a squeezed 5880 layout. Second, the family’s negative TCDk (−125 ppm/°C for 5880) is the largest drift direction in the Rogers RF line; systems that phase-match across temperature account for it, they do not ignore it.
Where the Family Goes
| Application | Typical member | Reason |
|---|---|---|
| Satellite payloads, TT&C | 5880 / 5880LZ | Loss floor, moisture immunity, outgassing-friendly construction; LZ where mass budget rules |
| Defence radar and EW | 5880 / 5870 | Low loss and stable phase at microwave through mmWave |
| Point-to-point and backhaul radios | 5880 | Antenna and feed networks where every 0.01 dB/in is link margin |
| High-power filters and combiners | 6002HT | PTFE-class loss with raised thermal path |
| Compact GNSS and patch arrays | 6006 / 6010LM | High Dk shrinks the radiator inside the same process family |
Conclusion
The RT/duroid 5000 family is small and precise: 5870, 5880 and 5880LZ, with the 6000-line PTFE relatives (6002, 6002HT, 6006, 6010LM) usually quoted alongside. Its two honest signatures are the loss floor — nothing mainstream beats Df 0.0009 — and the Z-axis expansion that shapes every multilayer design on it. Its model lists, unfortunately, are neither small nor precise: 6202, 6302 and 6500 do not exist, and the last one wears a hydrocarbon laminate’s parameters. Specify from the verified table and the family rewards it with the best loss behaviour money can buy.
XCE PCB fabricates the full RT/duroid 5000 and 6000 lines with PTFE-qualified processing: 2–72 layers, impedance ±10%, panels to 600 × 1200 mm, 5,000 m² of production space and 80+ engineers, with samples in as fast as 24 hours.
Frequently Asked Questions
What materials make up the RT/duroid 5000 series?
Is the 5000 series thermally stable in the Z axis?
Do Rogers 6202, 6302 and 6500 exist?
What is the difference between 5880 and 5880LZ?
Where do RT/duroid 6002 and 6002HT fit?
Can the 5000 series be processed like FR-4?
Sourcing RT/duroid 5000-Series Boards?
Send us your Gerber files and requirements for a manufacturability review and a tailored quotation.
sales1@xcepcb.com
+86 180 3341 4657