Passive Components and Devices Circulators, Isolators, and Switches Informational

How does the insertion loss of a switch affect the overall system noise figure?

Q: How much NF degradation is acceptable from the switch?

The acceptable NF degradation depends on the overall NF budget and sensitivity requirement: for radio astronomy (system NF < 0.5 dB): switch IL must be < 0.1 dB (use MEMS or place switch after the cryogenic LNA). For satellite ground station (NF < 1.5 dB): switch IL < 0.3 dB (use low-loss FET or MEMS). For cellular base station (NF < 3 dB): switch IL < 0.5 dB (standard FET switches suffice). For handset (NF < 7 dB): switch IL < 0.5 dB (SOI CMOS standard). For test equipment (NF not critical): switch IL < 2 dB is acceptable. Rule of thumb: the switch should contribute < 10% of the total system NF budget. If total NF = 3 dB: switch can contribute < 0.3 dB.

Q: Does an SP8T switch have more loss than an SPDT?

Yes, but the increase depends on the architecture: (1) Series-shunt topology (each throw has a series FET in the signal path): IL is approximately the same regardless of the number of throws (the signal only passes through one series FET). But: the parallel OFF-state capacitances of the other throws load the common port, increasing the effective capacitance and slightly increasing IL. For an SP8T: IL ≈ SPDT IL + 0.1-0.3 dB (due to the loading from 7 OFF-state throws). (2) Distributed tree topology (cascade of SPDT switches): an SP8T = 3 stages of SPDT. IL = 3 × SPDT IL. If SPDT = 0.3 dB: SP8T = 0.9 dB. This is why monolithic SP8T switches (single-stage) are preferred over cascaded SPDT trees (the single-stage has lower total IL).

Q: Can I compensate for switch loss with more LNA gain?

Not directly. Adding more LNA gain after the switch does not reduce the NF (the NF is determined by the first components in the chain). More gain improves the signal level at downstream stages but does not reduce the noise already added by the switch and LNA. However: you can reduce the NF impact of the switch by adding a low-noise amplifier BEFORE the switch (between the antenna and the switch). This pre-switch LNA amplifies the signal before the switch loss is introduced. The NF contribution of the switch becomes: delta_NF = IL_sw / G_pre_LNA. For G_pre_LNA = 15 dB and IL_sw = 0.5 dB: delta_NF = 0.5/31.6 = 0.016 dB (negligible). Trade-off: the pre-switch LNA must handle the full antenna bandwidth (all bands), increasing susceptibility to out-of-band interferers. This architecture is used in military and SDR receivers where ultimate sensitivity is needed.

A switch placed in the RF signal path adds its insertion loss directly to the system noise figure when the switch is before the first amplifier (LNA). Using the Friis formula for cascaded noise figure: NF_system = NF_switch + NF_LNA + (NF_next - 1)/G_LNA + ... The NF of a passive component (switch, cable, filter) equals its insertion loss in dB: NF_switch = IL_switch (dB). For a switch with 0.5 dB insertion loss before the LNA: NF_system = 0.5 + NF_LNA + remainder. If NF_LNA = 1.0 dB: NF_system = 0.5 + 1.0 + small remainder = 1.5 dB (minimum). Without the switch: NF_system = 1.0 dB. The 0.5 dB switch loss directly degraded the system NF by 0.5 dB. Impact: (1) In a sensitive receiver (radio astronomy, deep-space communication): 0.5 dB NF degradation is unacceptable. Use the lowest-loss switch available (MEMS: 0.1 dB) or place the switch after the LNA. (2) In a cellular base station (NF ≈ 3-5 dB): 0.5 dB switch loss causes a 0.5 dB degradation in receiver sensitivity, equivalent to approximately 5% reduction in cell radius. This is significant across a network of thousands of base stations. (3) In a handset receiver (NF ≈ 5-8 dB): 0.5 dB switch loss is a smaller fractional degradation but still impacts the edge-of-cell performance. Mitigation: (a) Place the switch after the LNA (the LNA gain reduces the impact of downstream losses). (b) Use a bypass LNA before the switch (in some architectures). (c) Use the lowest-loss switch technology available for the specific frequency and requirements.

Category: Passive Components and Devices

Updated: April 2026

Product Tie-In: Circulators, Isolators, Switches

Switch Loss and Noise Figure

The placement and insertion loss of RF switches is a critical receiver design decision because every tenth of a dB of pre-LNA loss directly degrades the receiver sensitivity.

Parameter	Option A	Option B	Option C
Performance	High	Medium	Low
Cost	High	Low	Medium
Complexity	High	Low	Medium
Bandwidth	Narrow	Wide	Moderate
Typical Use	Lab/military	Consumer	Industrial

Technical Considerations

For a receiver chain: antenna → switch → LNA → mixer → IF amp. NF_system = NF_sw + (NF_LNA - 1)/G_sw + (NF_mixer - 1)/(G_sw × G_LNA) + ... Where G_sw = 10^(-IL_sw/10) (the "gain" of the switch is its transmission coefficient, which is less than 1). Since IL_sw is small (0.3-1.5 dB): G_sw ≈ 0.93-0.71 (linear). For IL = 0.5 dB: G_sw = 0.891. NF_sw = 1/G_sw = 1.122 (linear) = 0.5 dB. NF_system = 10×log10(1.122 + (NF_LNA_linear - 1)/0.891 + ...). For NF_LNA = 1.0 dB (linear = 1.259): NF_system = 10×log10(1.122 + (1.259-1)/0.891) = 10×log10(1.122 + 0.291) = 10×log10(1.413) = 1.50 dB. Without switch: NF = 10×log10(1.259) = 1.0 dB. Net degradation: 0.50 dB (exactly equal to the switch IL, as expected for the first component).

Performance Analysis

(1) Before the LNA (highest NF impact): the switch IL adds directly to the system NF. This is the simplest architecture (single LNA after the switch selects the antenna/band). Used when: simplicity and cost are priorities, the switch IL is low (< 0.3 dB), or the system NF requirement is relaxed. (2) After the LNA (lowest NF impact): the NF degradation is: delta_NF = IL_sw / G_LNA (approximately, in dB). For IL_sw = 0.5 dB and G_LNA = 20 dB: delta_NF = 0.5/100 = 0.005 dB (negligible). However: the LNA amplifies the signal AND noise before the switch. If the LNA saturates due to a strong interferer: the system is more susceptible to blocking. Also: a separate LNA is needed for each antenna/band input before the switch, increasing cost and DC power. (3) Bypass switch: in some architectures, a bypass switch routes the signal around the LNA for high-power signals (to prevent LNA saturation). The bypass path has higher NF (no amplification) but prevents compression. The LNA path is used for weak signals (lower NF).

Design Guidelines

Consider a cellular base station receiver at 1900 MHz: Component chain: antenna → cable (0.5 dB) → filter (0.8 dB) → switch (0.3 dB) → LNA (gain 20 dB, NF 1.0 dB) → mixer (NF 8 dB, conversion gain -6 dB) → IF amp. NF calculation: NF_cable = 0.5 dB. NF_filter = 0.8 dB. NF_switch = 0.3 dB. Pre-LNA loss total = 0.5 + 0.8 + 0.3 = 1.6 dB. NF_system ≈ 1.6 + 1.0 + (8-1)/20 + ... ≈ 1.6 + 1.0 + 0.35 ≈ 3.0 dB. If the switch IL could be reduced from 0.3 dB to 0.1 dB (using MEMS): NF_system ≈ 1.4 + 1.0 + 0.35 ≈ 2.8 dB. Improvement: 0.2 dB. Over 50,000 base stations: 0.2 dB better NF increases the coverage area by approximately 4-5%, saving significant infrastructure cost. This is why switch insertion loss is a critical specification in cellular systems.

Performance verification: confirm specifications against the application requirements before finalizing the design
Environmental factors: temperature range, humidity, and vibration affect long-term reliability and parameter drift
Cost vs. performance: evaluate whether the application demands premium components or standard commercial grades

Implementation Notes

When evaluating how does the insertion loss of a switch affect the overall system noise figure?, engineers must account for the specific requirements of their target application. The optimal choice depends on the frequency range, power level, environmental conditions, and cost constraints of the overall system design.

Common Questions

Frequently Asked Questions

How much NF degradation is acceptable from the switch?

The acceptable NF degradation depends on the overall NF budget and sensitivity requirement: for radio astronomy (system NF < 0.5 dB): switch IL must be < 0.1 dB (use MEMS or place switch after the cryogenic LNA). For satellite ground station (NF < 1.5 dB): switch IL < 0.3 dB (use low-loss FET or MEMS). For cellular base station (NF < 3 dB): switch IL < 0.5 dB (standard FET switches suffice). For handset (NF < 7 dB): switch IL < 0.5 dB (SOI CMOS standard). For test equipment (NF not critical): switch IL < 2 dB is acceptable. Rule of thumb: the switch should contribute < 10% of the total system NF budget. If total NF = 3 dB: switch can contribute < 0.3 dB.

Does an SP8T switch have more loss than an SPDT?

Yes, but the increase depends on the architecture: (1) Series-shunt topology (each throw has a series FET in the signal path): IL is approximately the same regardless of the number of throws (the signal only passes through one series FET). But: the parallel OFF-state capacitances of the other throws load the common port, increasing the effective capacitance and slightly increasing IL. For an SP8T: IL ≈ SPDT IL + 0.1-0.3 dB (due to the loading from 7 OFF-state throws). (2) Distributed tree topology (cascade of SPDT switches): an SP8T = 3 stages of SPDT. IL = 3 × SPDT IL. If SPDT = 0.3 dB: SP8T = 0.9 dB. This is why monolithic SP8T switches (single-stage) are preferred over cascaded SPDT trees (the single-stage has lower total IL).

Can I compensate for switch loss with more LNA gain?

Not directly. Adding more LNA gain after the switch does not reduce the NF (the NF is determined by the first components in the chain). More gain improves the signal level at downstream stages but does not reduce the noise already added by the switch and LNA. However: you can reduce the NF impact of the switch by adding a low-noise amplifier BEFORE the switch (between the antenna and the switch). This pre-switch LNA amplifies the signal before the switch loss is introduced. The NF contribution of the switch becomes: delta_NF = IL_sw / G_pre_LNA. For G_pre_LNA = 15 dB and IL_sw = 0.5 dB: delta_NF = 0.5/31.6 = 0.016 dB (negligible). Trade-off: the pre-switch LNA must handle the full antenna bandwidth (all bands), increasing susceptibility to out-of-band interferers. This architecture is used in military and SDR receivers where ultimate sensitivity is needed.

Keep Reading