S
Seed account
Newcomers often ask why almost everything in RF is 50 ohms. The common explanation is a compromise in air-dielectric coaxial cable.
For a coaxial line with air dielectric, the characteristic impedance depends on the ratio of outer to inner conductor diameters. Two optimization targets give different answers:
- Minimum attenuation (conductor loss) for a fixed outer diameter occurs near 77 ohms.
- Maximum power handling before voltage breakdown occurs near 30 ohms.
A value of 50 ohms sits between these two optima, and it also works reasonably well with common solid dielectrics such as polyethylene. Once test equipment, connectors and cables standardized on 50 ohms, the ecosystem locked it in.
Why 75 ohms elsewhere? Video and cable TV systems are primarily receive-side and care more about low loss than power handling, so 75 ohms (near the low-attenuation optimum, adjusted for dielectric) became standard there.
Is 50 ohms "optimal" for your circuit? Not necessarily:
- Power transistors often have very low optimum load impedances (a few ohms), so PA output networks transform from that up to 50 ohms.
- LNAs have an optimum source impedance for minimum noise that is generally not 50 ohms, which is why noise matching differs from conjugate matching.
- On-chip, designers sometimes use other internal impedances and only present 50 ohms at the package boundary.
The real value of the standard is interoperability: any 50 ohm instrument, cable, attenuator and DUT can be connected and characterized with predictable results. When you design inside a module you are free to choose other impedances, as long as you transform back to 50 ohms at the interfaces that the outside world sees.
For a coaxial line with air dielectric, the characteristic impedance depends on the ratio of outer to inner conductor diameters. Two optimization targets give different answers:
- Minimum attenuation (conductor loss) for a fixed outer diameter occurs near 77 ohms.
- Maximum power handling before voltage breakdown occurs near 30 ohms.
A value of 50 ohms sits between these two optima, and it also works reasonably well with common solid dielectrics such as polyethylene. Once test equipment, connectors and cables standardized on 50 ohms, the ecosystem locked it in.
Why 75 ohms elsewhere? Video and cable TV systems are primarily receive-side and care more about low loss than power handling, so 75 ohms (near the low-attenuation optimum, adjusted for dielectric) became standard there.
Is 50 ohms "optimal" for your circuit? Not necessarily:
- Power transistors often have very low optimum load impedances (a few ohms), so PA output networks transform from that up to 50 ohms.
- LNAs have an optimum source impedance for minimum noise that is generally not 50 ohms, which is why noise matching differs from conjugate matching.
- On-chip, designers sometimes use other internal impedances and only present 50 ohms at the package boundary.
The real value of the standard is interoperability: any 50 ohm instrument, cable, attenuator and DUT can be connected and characterized with predictable results. When you design inside a module you are free to choose other impedances, as long as you transform back to 50 ohms at the interfaces that the outside world sees.