很多專案不是敗在 RF 理論,而是敗在「看起來像標準品」的誤判。AGV 在實驗室定位正常,到了倉庫靠近貨架就掉星;視覺系統樣機沒問題,量產後因走線變緊開始掉畫面;雷達線束接頭選對了,鏈路損耗卻超出預算。問題通常在看資料表時就埋下了:只看料號和外徑,沒有把真正影響量產穩定性的欄位讀清楚。
一份同軸電纜資料表不是在回答「這條線好不好」,而是在回答「它是否適合你的阻抗、頻段、安裝長度、彎曲路徑與測試要求」。所以在打開 PDF 之前,先把 5 個條件寫清楚:系統是 50 ohm 還是 75 ohm、工作頻段、實際安裝長度、真實走線,以及驗收是只有 continuity,還是還要 VSWR、insertion loss 或 TDR。
對正在採購 coaxial cable manufacturers、custom connector solutions 或 custom cable assemblies 的團隊,尤其是做 AGV & AMR platforms 和 logistics & warehouse robots 的案子,這一步最能減少試產重工。
先看哪 8 行,最能快速排除錯誤方案
| Datasheet Line | Why It Matters | Typical Red Flag | Buyer Action |
|---|---|---|---|
| Characteristic impedance | Must match the full RF chain | 50 ohm cable proposed for a 75 ohm video path, or the reverse | Confirm end-to-end impedance before quoting |
| Attenuation by frequency | Shows real signal loss, usually in dB/100 m | Only low-frequency points published, or no value near the operating band | Convert to the installed length at the real frequency |
| Capacitance | Affects signal behavior and compatibility in some applications | Value missing or inconsistent with cable family | Compare with known family norms when signal quality is sensitive |
| Velocity factor | Helps with propagation and phase-sensitive links | Unusually low value with no dielectric explanation | Check dielectric type and any timing requirement |
| Minimum bend radius | Predicts installation survivability | Route requires tighter bends than the published limit | Review brackets, exits, and clamp spacing before release |
| Shield construction / coverage | Influences EMI robustness and noise control | Marketing claim without braid or foil details | Ask for braid %, foil type, and grounding expectations |
| Temperature rating | Determines material survival near power electronics or outdoors | Jacket looks acceptable, rating does not | Review ambient, hot spots, and cleaning chemicals |
| Outer diameter and weight | Drives routing fit, strain relief, and connector compatibility | Cable cannot fit backshell, gland, or moving axis space | Confirm fit with the connector stack and route envelope |
真正有判斷力的,不是「low loss」或「high shield coverage」這類行銷字眼,而是 impedance、在實際頻率下的 attenuation、minimum bend radius、shield 結構和 temperature rating 這些能對應到真實安裝條件的數字。
阻抗和衰減一定要一起看
如果系統是 50 ohm,就不要把 75 ohm 電纜當成便宜替代;反過來也一樣。阻抗不匹配帶來的 reflection,不會因為 continuity 通過就消失。接著要看的就是 attenuation by frequency,而且要換算到真實安裝長度。
| Example Review Scenario | Datasheet Value | Installed Length | Approx. Cable Loss |
|---|---|---|---|
| GNSS antenna lead | 32 dB/100 m @ 1 GHz | 2.5 m | 0.80 dB |
| LTE roof antenna lead | 48 dB/100 m @ 1 GHz | 3.2 m | 1.54 dB |
| Wi-Fi radio extension | 76 dB/100 m @ 2.4 GHz | 1.8 m | 1.37 dB |
| 5.8 GHz RF link | 125 dB/100 m @ 5.8 GHz | 1.2 m | 1.50 dB |
| 75 ohm video path | 21 dB/100 m @ 100 MHz | 12 m | 2.52 dB |
把 dB/100 m 換成實際長度後,工程團隊才有辦法判斷 link budget 是否還夠。採購若只比單價,不比 0.80 dB 和 1.50 dB 的差異,最後省下來的通常不是整機成本。
機械參數其實也是電氣風險
在機器人專案裡,把電氣審查和走線審查分開做很常見;對同軸線來說,這樣做風險很高。minimum bend radius、外徑、護套材料、shield 結構都會影響受控阻抗在安裝後是否還能維持穩定。線纜若在支架出口被壓扁、在夾點附近被迫小半徑彎折,或在維修時反覆過度彎曲,最後壞掉的往往不是 continuity,而是整條 RF 路徑的一致性。
同樣,shield coverage 不能只看百分比。95% braid 聽起來很好,但它不會自動告訴你 transfer impedance、foil overlap、接地方式,以及在 inverter、高電流充電支路或伺服線旁邊的真實 EMI 表現。
RFQ 裡一定要補上的資料
- Drawing, route sketch, or photos showing connector orientation and clamp locations.
- BOM or approved cable family reference, including any alternates already under consideration.
- Operating frequency or band, system impedance, and estimated allowable link loss if known.
- Installed length, prototype quantity, annual volume, and target lead time.
- Environment details: temperature range, vibration, abrasion, moisture, chemicals, and nearby noise sources.
- Compliance target and documentation expectation, such as traceability, sample test report, or first-article package.
- Validation scope: continuity, pin map, VSWR, insertion loss, TDR, retention, dielectric, or environmental tests.
如果你把圖紙、BOM、數量、環境、目標交期和合規要求一起送出來,我們能回給你 manufacturability review、建議的 cable-and-connector stack、routing 與 shielding 風險提示、驗證範圍建議,以及區分樣品與量產的報價。
FAQ
最先該看哪 3 個欄位?
先看 impedance、實際工作頻率下的 attenuation,以及 minimum bend radius。只要這 3 項不符合系統鏈路、安裝路徑與訊號預算,後面參數再漂亮也沒用。
50 ohm 一定比 75 ohm 更適合機器人嗎?
不一定。GNSS、LTE、Wi-Fi、telemetry 與 radar 常見 50 ohm;視訊與部分視覺系統則常見 75 ohm。端到端一致才是重點。
RFQ 應該寫哪個衰減值?
寫最接近實際工作頻率的 attenuation,並換算成真實安裝長度。這樣工程與採購討論的是同一個損耗數字。
為什麼 bend radius 也要進電氣審查?
因為線纜電氣表現對安裝形態非常敏感。低於規範的反覆彎折會破壞 dielectric、拉傷 shield,並縮短壽命。
只看 shield coverage 夠嗎?
不夠。還要看 braid 結構、foil 設計、接地方式、transfer impedance,以及它與高噪聲回路的隔離。
讀完資料表後,下一步要送什麼?
送 drawing 或 route sketch、BOM、數量、環境、connector family、compliance target 和 test scope,讓供應商判斷 catalog cable 是否真的適合量產。
下一步請送完整資料,不要只丟 PDF
如果你正在替 RF cable assembly 選同軸線,請把 drawing、BOM、樣品與量產數量、安裝環境、目標 lead time、compliance target,以及已知頻段和 connector family 一起發來。我們會回給你更接近量產決策的評審,而不只是單純報價。