Attenuation in dB/m for the coaxial cables used in IoT antenna assemblies, at 433 MHz, 900 MHz, 2.4 GHz and 5.8 GHz — plus an honest note that RG-316 has two legitimate datasheets differing by about 40%, and the length thresholds where cable choice starts to matter.
If your own measurement never matches a table you found online, this is usually the reason:
| Cable @ 900 MHz | Optimistic | Conservative | What is going on |
|---|---|---|---|
| RG-316 | 24.7 dB/100 ft (0.81 dB/m) — an integration guide's nominal value | 38 dB/100 ft at 1 GHz (1.25 dB/m) — MIL-DTL-17/113D maximum | "Typical of a good production run" versus "worst case allowed by the spec". Spread ≈ 40 % |
| RG-178 | 24 dB/100 ft (0.79 dB/m) — some distributor tables | 42.7 dB/100 ft (1.40 dB/m) — cable-maker maximum | The optimistic figure is RG-316-class and physically impossible for a 1.8 mm cable with a 0.30 mm conductor. Do not design to distributor blog values for RG-178 |
| 1.13 mm @ 2.4 GHz | 1.44 dB/m — one distributor table | 2.80–3.20 dB/m — Rosnol datasheet and independent lab data | The low figure would make 1.13 mm lower-loss than RG-178, which cannot happen |
| LMR-195 @ 900 MHz | 7.8 dB/100 ft | 11.1 dB/100 ft | The optimistic value would beat LMR-200 on a thinner cable |
The mechanism: attenuation = conductor loss (scales ≈ √f, dominated by the inner conductor at VHF/UHF) + dielectric loss (scales ≈ f) + leakage from imperfect braid coverage. A maker quoting "typical" for a 95 %-coverage silver-plated braid and a maker quoting "maximum" for a 90 %-coverage tinned braid will differ by 30–50 % on the same RG number, because "RG-316" defines a construction family, not a guaranteed loss. Publish the range and explain it — that is worth more to a technical buyer than any single number.
At 433 MHz the cable question mostly evaporates: 1 m of RG-174 costs 0.58 dB and 1 m of LMR-400 costs 0.09 dB — a 0.5 dB difference on a 1 m jumper, invisible against 10–20 dB of link margin. At 2.4 GHz the same comparison is 1.46 dB vs 0.22 dB per metre, and at 5.8 GHz 2.0 dB vs 0.35 dB. Cable becomes a system-level decision above ~1 GHz, or beyond ~2 m.
| Run length | Sub-GHz (433–915 MHz) | 2.4 GHz |
|---|---|---|
| < 0.3 m internal jumper | 1.13 mm is fine (0.3–0.6 dB) | 1.13 mm fine (≈1 dB) |
| 0.3–1 m | RG-174 / RG-316 acceptable | RG-316 or LMR-100A; avoid RG-178 |
| 1–3 m | RG-58 / LMR-195 | LMR-100A or LMR-200 minimum; RG-174 burns 3–4.5 dB |
| > 3 m | LMR-400, or move the radio outdoors | LMR-400, or go active / relocate the radio |
| 5–15 m mast run | LMR-400 (≈1.9 dB at 15 m @ 868 MHz) — this is why gateways use N-type | LMR-400 minimum; consider a remote radio head |
Every dB of cable loss costs you a dB of transmit range and a dB of receive sensitivity. Adding gain at the antenna is usually more expensive and more directional than shortening the cable. Before specifying a higher-gain antenna, add up: connector pairs, cable run, and any pigtail inside the enclosure.
Contact: Mr.Li
Phone: 0755-33169935
Tel: +86-755-33169935
Email: cs#antennahome.com (change # to @ )
Add: Build-2 201#, ShangXue Industry Park, Bantian, Longgang District, Shenzhen, Guangdong Province, China