Use 50 Ω coax when your RF equipment, antenna, or test gear specifies 50 Ω; use 75 Ω for video, CATV, satellite, or broadcast systems designed around 75 Ω. Impedance matching across the entire signal path—source, cable, connectors, and load—matters more than any single cable property. Below I'll walk through the engineering tradeoffs, real-world loss behavior, and the mistakes that still trip up experienced engineers.
Key Takeaways:
- Neither 50 Ω nor 75 Ω is inherently lower-loss; actual attenuation depends on conductor size, dielectric, frequency, and construction—a thick 50 Ω cable often outperforms thin 75 Ω cable at the same frequency.
- A direct 50-to-75 Ω junction reflects roughly 4% of incident power and yields 1.5:1 VSWR and 14 dB return loss—acceptable in some systems, problematic in precision RF work.
- BNC connectors exist in both 50 Ω and 75 Ω versions that can physically mate but create electrical discontinuities; I've debugged more than one "mystery VSWR spike" that turned out to be mixed-impedance BNCs.
- The 50 Ω standard emerged as a compromise between minimum attenuation (theoretically ~77 Ω for air-filled coax) and maximum power handling (~30 Ω), not because it was designated for "transmitting."
- Cable loss scales with frequency and length; an 8 dB/100 m spec at 1 GHz becomes 2 dB over 25 m, and that's before you add connector losses.
What 50 Ω and 75 Ω Actually Mean
Characteristic impedance (Z₀) describes the ratio of voltage to current for an electromagnetic wave traveling down the transmission line. It is not a DC resistance you can measure with a multimeter.
For an ideal coaxial cable with a uniform dielectric:
Z₀ ≈ (60 / √εᵣ) × ln(D/d)
where εᵣ is the relative permittivity of the dielectric, D is the inside diameter of the outer conductor, and d is the outside diameter of the center conductor.
Changing the D/d ratio or the dielectric material changes the impedance. A 50 Ω cable and a 75 Ω cable can use the same dielectric but different conductor ratios, or the same geometry with different dielectrics.
Characteristic impedance stays constant along a uniform cable regardless of length. A 1 m piece and a 20 m piece of the same 50 Ω coax both present 50 Ω to a traveling wave, even though the longer cable has higher DC resistance and signal loss. This distinction becomes critical when you're trying to diagnose transmission-line problems versus simple resistive loss.
Why a Multimeter Reads Open Circuit, Not 50 or 75 Ω
Connect an ohmmeter across an unterminated, undamaged coax and you'll see an open circuit, not 50 Ω or 75 Ω.
DC instruments measure resistive paths. Characteristic impedance arises from the distributed inductance and capacitance per unit length of the transmission line. To measure Z₀, you need a time-domain reflectometer (TDR), vector network analyzer (VNA), or similar transmission-line test equipment that operates at RF frequencies.
In practical terms: if you're trying to identify mystery coax in a bin, don't trust the ohmmeter. Check the cable marking, measure with proper RF gear, or reference the manufacturer datasheet.
The Engineering History Behind 50 Ω and 75 Ω
50 Ω did not become the RF standard because someone decided "transmitters get 50 Ω." The story is more interesting.
For air-filled coaxial cable, minimum attenuation occurs near 77 Ω, maximum power-handling capability peaks around 30 Ω, and maximum voltage standoff falls at a different impedance still. No single value optimizes everything.
50 Ω sits in the middle of these competing requirements—low enough for reasonable power handling, high enough to keep loss acceptable, and far enough from extreme values to be manufacturable with consistent geometry. Once early RF engineers and military specifications standardized on 50 Ω, the ecosystem of connectors, test equipment, amplifiers, and antennas followed.
75 Ω lies closer to the theoretical minimum-loss point for air dielectric and became the standard for signal distribution systems—cable TV, broadcast, video—where loss over long runs mattered more than transmitter power handling. The 75 Ω standard also benefits from larger conductor spacing for a given outer diameter, which can improve breakdown voltage in some designs.
But here's the limitation: those textbook curves assume idealized air-filled coax. Real cables use solid or foam polyethylene, PTFE, or other dielectrics that shift the loss minimum. A modern low-loss 50 Ω cable with foam dielectric and large conductors can easily beat a thin 75 Ω cable with solid dielectric, even though 75 Ω is theoretically closer to the air-dielectric optimum.
Signal Loss: Why Impedance Alone Doesn't Predict Attenuation
Attenuation increases with frequency and depends on conductor size, conductor material (copper vs. copper-clad steel), dielectric loss tangent, shield construction, and cable diameter.
Consider a representative Belden 50 Ω RG-58-type cable:
| Frequency | Attenuation (dB/100 m) |
|---|---|
| 10 MHz | 4.7 |
| 200 MHz | 23 |
| 400 MHz | 34 |
| 1 GHz | 60 |
That's more than a 10× increase in loss per unit length from 10 MHz to 1 GHz. At microwave frequencies, cable loss dominates link budgets, and the difference between a thin cable and a thick low-loss cable can be 10 dB or more over a 30-meter run.
Now compare two hypothetical cables at 1 GHz:
- Thin 75 Ω RG-59: ~65 dB/100 m
- Thick 50 Ω LMR-400: ~22 dB/100 m
The 50 Ω cable has less than one-third the loss of the 75 Ω cable, even though 75 Ω is supposedly the "low-loss" impedance.
The lesson: when signal loss matters, compare actual dB/length specifications at your operating frequency. Impedance is a matching requirement, not a performance predictor.
What Happens When You Mix 50 Ω and 75 Ω
Every impedance discontinuity reflects part of the signal back toward the source. For a 50 Ω line feeding a 75 Ω load:
Γ = (Z_L - Z₀) / (Z_L + Z₀) = (75 - 50) / (75 + 50) = 0.20
The reflection coefficient is 0.20, so the reflected power fraction is:
Γ² = 0.04 or 4%
VSWR = (1 + |Γ|) / (1 - |Γ|) = 1.5:1
Return loss = -20 log₁₀(|Γ|) ≈ 14 dB
In many video and CATV systems, 1.5:1 VSWR is tolerable. In precision RF measurements, lab-grade systems, or high-power transmitters, it's not. The reflected energy can cause standing waves, reduce delivered power, increase transmitter load stress, and degrade error vector magnitude (EVM) in digital modulation.
I've seen engineers connect a 75 Ω satellite LNB to 50 Ω test equipment with a mechanical adapter and wonder why the noise figure measurement drifts. The adapter changed the mechanical interface but did nothing to transform the impedance—it just moved the reflection point.
50 Ω vs 75 Ω: Application and Cable Selection
| Application | Standard Impedance | Representative Cables | Primary Concern |
|---|---|---|---|
| Amateur & two-way radio | 50 Ω | RG-58, RG-8, RG-213 | Frequency, loss, power handling |
| RF test equipment | 50 Ω | Precision 50 Ω coax | Impedance accuracy, connector quality |
| Wi-Fi, cellular, GNSS | 50 Ω | Low-loss 50 Ω, LMR series | High-frequency attenuation |
| Cable television (CATV) | 75 Ω | RG-6, RG-11 | Distribution loss over long runs |
| Satellite TV | 75 Ω | RG-6 | Loss, shielding, weather resistance |
| CCTV and analog video | 75 Ω | RG-59 | Baseband video bandwidth |
| Broadcast & SDI video | 75 Ω | Precision 75 Ω broadcast | Data rate, return loss, cable length |
| TV antenna distribution | 75 Ω | RG-6, RG-11 | Impedance matching, distribution loss |
Common Cable Types
| Cable | Impedance | Typical Use | Key Tradeoff |
|---|---|---|---|
| RG-58 | 50 Ω | General RF, short antenna runs | Compact but higher loss than larger cables |
| RG-8 | 50 Ω | Radio, antenna systems | Larger diameter, lower loss, higher power rating |
| RG-213 | 50 Ω | RF, antenna feeds | Lower loss and better power handling than RG-58 |
| RG-174 | 50 Ω | Internal RF, jumper cables | Very flexible but high loss—use short lengths |
| RG-59 | 75 Ω | CCTV, video | Adequate for video, not ideal for long CATV runs |
| RG-6 | 75 Ω | CATV, satellite, broadband | Good loss and shielding for distribution |
| RG-11 | 75 Ω | Long-distance distribution | Larger diameter, lower loss than RG-6 |
Do not select cable by RG number alone. The RG designation indicates a general family, but manufacturers vary in conductor type (solid vs. stranded, copper vs. copper-clad steel), dielectric (solid vs. foam polyethylene, PTFE), shield (braid, foil, or both), jacket material, and environmental ratings. Two "RG-58" cables can differ by 3 dB/100 m at the same frequency.
Connectors: The Hidden Impedance Discontinuity
The cable is only part of the signal path. Connectors, adapters, splitters, and terminations also have characteristic impedance.
An ideal matched 50 Ω path looks like:
50 Ω source → 50 Ω connector → 50 Ω cable → 50 Ω connector → 50 Ω antenna or load
BNC: 50 Ω and 75 Ω Versions Exist
BNC connectors are manufactured in both 50 Ω and 75 Ω versions. They often mate physically, but the pin diameter and dielectric dimensions differ to maintain impedance.
Mixing a 50 Ω BNC plug with a 75 Ω BNC jack creates a discontinuity. I've debugged a cellular repeater installation where the VSWR was 2:1 at the antenna port; the problem turned out to be a 75 Ω BNC barrel adapter someone grabbed from the video bench. Swapping in a 50 Ω barrel dropped VSWR to 1.2:1.
Amphenol and other connector manufacturers explicitly warn against mixing 50 Ω and 75 Ω BNC components because the mechanical compatibility does not equal electrical compatibility.
SMA, N-Type, TNC, and Other RF Connectors
SMA connectors are almost always 50 Ω and are the standard for microwave and precision RF work up to 18 GHz or higher.
N-type connectors are available in 50 Ω and 75 Ω versions, though 50 Ω is far more common in RF infrastructure.
TNC, SMB, and SMC families may also have impedance variants depending on series and manufacturer.
When in doubt, check the datasheet. Connector shape alone does not guarantee impedance.
F-Type Connectors
F-type connectors are the screw-on connectors ubiquitous in 75 Ω CATV, satellite, and broadband systems. They are not used in 50 Ω systems.
Power Handling and Frequency Limits
Power rating depends on conductor heating, dielectric breakdown voltage, cable size, frequency, ambient temperature, and installation conditions (enclosed conduit vs. open air).
The 50 Ω standard is associated with RF power because that's where the ecosystem developed, but impedance alone does not determine power capability. A thick 75 Ω hardline cable can handle more power than a thin 50 Ω jumper.
Connector power limits often govern system capability. A cable rated for 1 kW means nothing if the connector fails at 500 W.
Frequency capability is equally important. A cable with acceptable loss at 150 MHz may be unusable at 2.4 GHz. Connector discontinuities and dimensional tolerances that are negligible at VHF become serious reflections at microwave frequencies.
Common Problems I've Seen
High VSWR after changing a connector: Check that the new connector matches the system impedance. 50 Ω and 75 Ω BNC, N-type, or other families can physically mate but create electrical mismatches.
Weak signal after a long cable run: Calculate actual loss. If you're running 30 meters of RG-58 at 1 GHz (60 dB/100 m), that's 18 dB of cable loss alone—before connectors. A low-loss cable or an inline amplifier may be necessary.
Intermittent performance or high reflected power: Inspect terminations and cable routing. Coax bent tighter than the minimum bend radius can deform the dielectric and change impedance locally. Damaged shield or water intrusion also creates reflections.
Unexpected loss at high frequency: Verify that every component—cable, connectors, adapters—is rated for the operating frequency. A DC-to-1 GHz cable is not suitable for a 6 GHz Wi-Fi 6E system.
When Mixing Impedances Might Be Acceptable
A 50-to-75 Ω mismatch is not always catastrophic. In some receive-only systems, broadband TV distribution, or low-frequency applications where VSWR < 2:1 is tolerable, the 4% reflection may not matter.
But in precision measurements, high-power transmitters, low-noise receivers, or systems where return loss affects modulation quality, maintain matched impedance throughout the path or use a proper matching network—not a mechanical adapter.
FAQ
Can I use 75 Ω coax on 50 Ω equipment?
You can physically connect it, but the impedance mismatch creates reflections. Whether that's acceptable depends on your application. For predictable RF performance, match the impedance.
Does 75 Ω coax always have lower loss than 50 Ω?
No. Real-world loss depends on conductor size, dielectric, construction, and frequency. A large 50 Ω cable often has lower attenuation than a small 75 Ω cable.
Why does my multimeter not read 50 or 75 Ω?
Characteristic impedance is not DC resistance. It's determined by geometry and distributed capacitance and inductance. Use a TDR or VNA to measure transmission-line impedance.
Are 50 Ω and 75 Ω BNC connectors interchangeable?
They may physically mate, but they have different pin and dielectric dimensions to maintain their respective impedances. Mixing them creates a discontinuity and degrades return loss.
Will a 50-to-75 Ω adapter fix the mismatch?
A mechanical adapter changes the connector type but does not transform impedance. If you need actual impedance transformation, use a component designed as a matching transformer or matching pad.
Which impedance should I choose for an antenna system?
Use the impedance specified by the antenna and connected equipment. Most RF, wireless, and cellular systems use 50 Ω; most TV and video distribution systems use 75 Ω. Match the entire signal path.
