Q1Electromagnetics Waves

(a) Why microstrip lines are better than waveguides at and above 60 GHz: the dominant-mode cutoff frequency of a rectangular waveguide (fc = c/2a) requires the broad dimension a to shrink in proportion to 1/f as frequency rises. At 60 GHz and above, a standard waveguide's cross-section shrinks to a few millimetres or less, making the waveguide extremely difficult and costly to machine to the required precision, mechanically fragile, and very lossy per unit length relative to its now-tiny cross-sectional area (surface resistive loss scales unfavourably as the guide shrinks). Microstrip lines, in contrast, are fabricated by simple photolithographic etching on a dielectric substrate — their dimensions do shrink with increasing frequency too, but photolithography easily achieves micrometre-scale precision at low cost, and just as importantly, microstrip integrates directly and monolithically with planar active devices (transistors, MMICs, diodes) on the same substrate, eliminating the need for bulky, lossy waveguide-to-coax or waveguide-to-chip transitions that would otherwise be needed at every active device interface. These fabrication-cost, integration and miniaturization advantages make microstrip (and other planar lines) the practical choice for millimetre-wave ICs and modules at 60 GHz and above, despite microstrip's inherently higher loss and lower power-handling than waveguide at any given frequency.

(b) Why waveguides are better than microstrip lines between 1 and 30 GHz: in this lower microwave range, waveguide cross-sectional dimensions remain conveniently large (centimetre-scale), so the fabrication-difficulty argument that favours microstrip at mm-wave frequencies does not yet apply. Waveguides offer two decisive advantages here: (i) power handling — a hollow air-filled waveguide has a much larger effective cross-sectional area and a far higher dielectric (air) breakdown field than the thin dielectric substrate and narrow conducting strip of a microstrip line, so waveguides can carry orders of magnitude more power (from hundreds of watts to megawatts in radar transmitters) before arcing or thermal failure; (ii) loss — a waveguide's fields are largely confined within a low-loss air-filled interior bounded by conductor walls, whereas a microstrip's fields exist partly in the (typically lossier) dielectric substrate and are concentrated at a narrow conducting strip's sharp edges (encouraging higher current density and thus higher conductor loss); consequently waveguides exhibit substantially lower attenuation per unit length than microstrip at the same frequency in this range. These two properties — high power handling and low loss — are exactly what is required in high-power radar transmitters, satellite uplink chains, and long-haul microwave links operating in the 1-30 GHz range, which is why waveguide remains the preferred medium there despite its greater bulk, weight and machining cost compared with microstrip.

Crossover behaviour and design implication: the two arguments above are, in effect, two ends of the same underlying trade-off between physical size and electrical performance, and neither technology is universally superior — the appropriate choice depends on where in the spectrum a given subsystem operates. In practice many mm-wave systems above 60 GHz are still front-ended by short transitional waveguide sections (for the antenna feed or the highest-power stage) that are subsequently converted to microstrip via a waveguide-to-microstrip transition as soon as the signal reaches lower-power amplification or mixing stages, deliberately combining waveguide's power-handling strength at the antenna interface with microstrip's compactness and integration advantage everywhere else in the signal chain — illustrating that the 1-30 GHz versus 60+ GHz comparison given above is best read as a guideline for which technology dominates a frequency band's transmission-line budget, rather than an absolute rule excluding the other technology entirely from that band.