Signal Quality · Noise · Instrumentation · 4-20 mA
Signal-to-Noise Ratio (SNR) in Instrumentation: Formulas, Worked Examples and Interactive Calculator
A complete plain-English guide to signal-to-noise ratio in process instrumentation: why SNR matters in 4-20 mA loops and PLC inputs, the voltage and power formulas derived from first principles, a dB-to-linear reference table, noise sources in field systems, worked calculation examples with full step-by-step working, and an interactive multi-mode calculator with quality verdict.
Voltage and Power Formulas Explained
dB Reference Table
Interactive Calculator
Noise Sources in Process Plants
Signal to Noise Ratio (SNR): Introduction
Every measurement signal in a process plant arrives at the PLC or DCS input card contaminated to some degree with noise. The thermocouple signal from a furnace picks up inductive interference from nearby power cables. The 4-20 mA signal from a pressure transmitter rides on a background of switching noise from a variable frequency drive sharing the same cable tray. The Modbus signal from an analyser is affected by ground loops created by equipment connected across different earth potential points.
Signal-to-noise ratio (SNR) is the number that quantifies how clean a signal is relative to the noise contaminating it. A high SNR means the measurement is reliable, stable and suitable for control. A low SNR means the measurement is unreliable, noisy and likely to cause control loop problems, false alarms and poor process visibility. Understanding SNR is essential for any instrumentation technician or engineer doing loop commissioning, signal troubleshooting or instrument selection.
This guide explains SNR completely: what it is, why the voltage formula uses 20 and the power formula uses 10, how to read and use dB values, a full dB-to-linear reference table, the common sources of noise in process plant instrument loops, worked calculation examples in both voltage and power modes, and a self-contained interactive calculator. For context on the 4-20 mA signal whose quality is most often evaluated with SNR, see our guide on the 4-20 mA current loop explained.
What this guide covers
What signal-to-noise ratio means in plain English
Why SNR is expressed in decibels (dB) and what dB means physically
The voltage SNR formula: SNR(dB) = 20 x log10(Signal / Noise) and why the factor is 20
The power SNR formula: SNR(dB) = 10 x log10(Signal / Noise) and why the factor is 10
How to convert between dB and linear ratio in both directions
Complete dB to linear reference table: -10 dB to +60 dB
Common sources of noise in process plant instrument loops
SNR quality thresholds for industrial instrumentation
Worked Example 1: 4-20 mA loop voltage SNR calculation
Worked Example 2: Power-domain SNR for a wireless transmitter
Worked Example 3: Converting a measured SNR to minimum detectable signal
Interactive calculator: voltage mode, power mode and dB-to-ratio conversion
Practical methods to improve SNR in field installations
Why SNR is expressed in decibels (dB) and what dB means physically
The voltage SNR formula: SNR(dB) = 20 x log10(Signal / Noise) and why the factor is 20
The power SNR formula: SNR(dB) = 10 x log10(Signal / Noise) and why the factor is 10
How to convert between dB and linear ratio in both directions
Complete dB to linear reference table: -10 dB to +60 dB
Common sources of noise in process plant instrument loops
SNR quality thresholds for industrial instrumentation
Worked Example 1: 4-20 mA loop voltage SNR calculation
Worked Example 2: Power-domain SNR for a wireless transmitter
Worked Example 3: Converting a measured SNR to minimum detectable signal
Interactive calculator: voltage mode, power mode and dB-to-ratio conversion
Practical methods to improve SNR in field installations
Advertisement
Advertisement
What Is Signal-to-Noise Ratio (SNR)?
Signal-to-noise ratio is the ratio of the power or amplitude of a wanted signal to the power or amplitude of the background noise. It tells you how much stronger the useful measurement signal is compared to the unwanted noise riding on top of it.
The plain-English definition
Imagine you are trying to hear someone talking to you across a noisy factory floor. The voice is the signal. The background machinery noise is the noise. If the voice is much louder than the noise, you can hear clearly (high SNR). If the machinery is almost as loud as the voice, you can barely make out what is being said (low SNR). Now replace the voice with a 4-20 mA measurement signal and the machinery with electrical interference, and you have exactly the SNR problem that instrumentation engineers face every day.Figure 1: SNR Visualised: Clean Signal vs Noisy Signal on the Same Measurement
Figure 1: The same 4-20 mA step measurement signal with high SNR (left, clean) vs low SNR (right, noisy). With high SNR the PLC input reads a stable, accurate value. With low SNR the noise masks the real signal, the displayed value jumps continuously and any derivative action in a PID controller amplifies the noise into violent valve movement.
The SNR Formulas: Why Voltage Uses 20 and Power Uses 10
The difference between the voltage formula (factor of 20) and the power formula (factor of 10) confuses many people. The reason is straightforward once you understand the relationship between voltage and power.
Power SNR formula (fundamental definition):
SNR(dB) = 10 x log10(P_signal / P_noise)
Where:
P_signal = power of the useful signal (watts)
P_noise = power of the noise (watts)
log10 = base-10 logarithm
This is the original definition. Power is the fundamental quantity.
1 dB of SNR = a power ratio of 10^(1/10) = 1.259 (the signal is 25.9% stronger in power)
Where:
P_signal = power of the useful signal (watts)
P_noise = power of the noise (watts)
log10 = base-10 logarithm
This is the original definition. Power is the fundamental quantity.
1 dB of SNR = a power ratio of 10^(1/10) = 1.259 (the signal is 25.9% stronger in power)
Why voltage formula uses 20 (not 10):
Power is proportional to the SQUARE of voltage:
P = V² / R
So if voltage ratio is V_s / V_n, the power ratio is:
P_s / P_n = (V_s²/R) / (V_n²/R) = (V_s / V_n)²
Substituting into the power SNR formula:
SNR(dB) = 10 x log10((V_s / V_n)²)
= 10 x 2 x log10(V_s / V_n)
SNR(dB) = 20 x log10(V_s / V_n)
The factor of 2 from squaring the voltage ratio becomes the factor of 20. This is why you use 20 for voltage and current ratios, and 10 for power ratios. Both formulas give the same dB result when the physical quantity is the same. NEVER mix: do not apply the voltage formula to power values or vice versa.
P = V² / R
So if voltage ratio is V_s / V_n, the power ratio is:
P_s / P_n = (V_s²/R) / (V_n²/R) = (V_s / V_n)²
Substituting into the power SNR formula:
SNR(dB) = 10 x log10((V_s / V_n)²)
= 10 x 2 x log10(V_s / V_n)
SNR(dB) = 20 x log10(V_s / V_n)
The factor of 2 from squaring the voltage ratio becomes the factor of 20. This is why you use 20 for voltage and current ratios, and 10 for power ratios. Both formulas give the same dB result when the physical quantity is the same. NEVER mix: do not apply the voltage formula to power values or vice versa.
The two SNR formulas side by side
Voltage (or current) inputs: SNR(dB) = 20 x log10(V_signal / V_noise)
Use when: signal and noise are given in V, mV, µV, or mA, µA (current)
Power inputs: SNR(dB) = 10 x log10(P_signal / P_noise)
Use when: signal and noise are given in W, mW, µW, dBm (power)
dB to Linear Ratio Reference Table
The decibel scale is logarithmic, which means the relationship between dB and the actual signal ratio is not intuitive. This table lets you quickly convert between dB values and the linear ratio they represent, in both voltage and power domains.
| SNR (dB) | Voltage ratio (V_s/V_n) | Power ratio (P_s/P_n) | Quality (instrumentation) | Practical meaning in a 4-20 mA loop |
|---|---|---|---|---|
| -10 dB | 0.316 : 1 | 0.10 : 1 | TERRIBLE | Noise is stronger than signal. Reading is meaningless. Loop cannot control. |
| 0 dB | 1 : 1 | 1 : 1 | TERRIBLE | Signal equals noise. Displayed value jumps randomly. Alarm systems unreliable. |
| 6 dB | 2 : 1 | 4 : 1 | POOR | Signal twice the noise. Barely usable. PID derivative must be disabled. Unstable readings. |
| 10 dB | 3.16 : 1 | 10 : 1 | POOR | Minimum acceptable for monitoring only. Not suitable for closed-loop control. |
| 20 dB | 10 : 1 | 100 : 1 | ACCEPTABLE | Signal 10x noise. Adequate for most PI control. Derivative action still risks noise amplification. |
| 26 dB | 20 : 1 | 400 : 1 | ACCEPTABLE | Good practical threshold for field 4-20 mA signals in most process plant environments. |
| 30 dB | 31.6 : 1 | 1000 : 1 | GOOD | Signal 31.6x noise. Comfortable margin for all control modes including derivative. Stable HMI display. |
| 40 dB | 100 : 1 | 10000 : 1 | GOOD | Signal 100x noise. Very stable. Suitable for precision measurement and safety systems. |
| 50 dB | 316 : 1 | 100000 : 1 | EXCELLENT | Excellent signal quality. Typical of a well-installed 4-20 mA loop with proper shielding and grounding. |
| 60 dB | 1000 : 1 | 1000000 : 1 | EXCELLENT | Outstanding. Typical benchmark for clean laboratory instrumentation or industrial twisted-pair with perfect installation. |
Quick memory aid: the 20 dB rule and the 6 dB rule
Every 20 dB increase in SNR means the voltage ratio multiplied by 10 (signal becomes 10x stronger relative to noise). Every 6 dB increase in SNR means the voltage ratio doubled. So: 20 dB = 10:1 voltage ratio. 40 dB = 100:1. 60 dB = 1000:1. For power: every 10 dB increase means the power ratio multiplied by 10. So: 10 dB = 10:1 power ratio. 20 dB = 100:1. 30 dB = 1000:1.Converting Between dB and Linear Ratio
Convert dB to voltage ratio and power ratio:
Voltage ratio = 10^(SNR_dB / 20)
Power ratio = 10^(SNR_dB / 10)
Example: SNR = 34 dB
Voltage ratio = 10^(34/20) = 10^1.7 = 50.1 : 1
Power ratio = 10^(34/10) = 10^3.4 = 2512 : 1
Convert voltage ratio to dB: SNR(dB) = 20 x log10(voltage ratio)
Example: voltage ratio = 75 : 1
SNR(dB) = 20 x log10(75) = 20 x 1.875 = 37.5 dB
Convert power ratio to dB: SNR(dB) = 10 x log10(power ratio)
Example: power ratio = 500 : 1
SNR(dB) = 10 x log10(500) = 10 x 2.699 = 27.0 dB
Power ratio = 10^(SNR_dB / 10)
Example: SNR = 34 dB
Voltage ratio = 10^(34/20) = 10^1.7 = 50.1 : 1
Power ratio = 10^(34/10) = 10^3.4 = 2512 : 1
Convert voltage ratio to dB: SNR(dB) = 20 x log10(voltage ratio)
Example: voltage ratio = 75 : 1
SNR(dB) = 20 x log10(75) = 20 x 1.875 = 37.5 dB
Convert power ratio to dB: SNR(dB) = 10 x log10(power ratio)
Example: power ratio = 500 : 1
SNR(dB) = 10 x log10(500) = 10 x 2.699 = 27.0 dB
SNR Calculator: Voltage Mode, Power Mode and dB Converter
The calculator below handles all three common SNR calculation needs. Select the mode using the tabs, enter your values with the correct units, and click Calculate for an instant result with quality verdict and full formula shown.
Signal-to-Noise Ratio Calculator
Voltage mode · Power mode · dB to linear converter · Quality verdict
The useful measurement signal amplitude
Root-mean-square noise amplitude on the signal
✔ Result
SNR (dB)
Voltage ratio
Power ratio
Formula used
20×log10
Received useful signal power
Background noise power level
✔ Result
SNR (dB)
Power ratio
Voltage ratio
Formula used
10×log10
Enter positive dB to convert to ratio. Negative dB means noise is stronger than signal.
dB
✔ Result
Voltage ratio
Power ratio
SNR (dB)
Quality
Advertisement
Advertisement
Worked Examples: Full Step-by-Step SNR Calculations
Example 1: 4-20 mA Loop Signal Quality Check (Voltage Mode)
A commissioning engineer measures the 4-20 mA signal on a pressure transmitter loop. The signal at 50% process value is 12 mA (across a 250-ohm barrier resistor = 3.0 V). A clamp meter in AC mode on the signal wire reads 0.03 V RMS of electrical noise. Calculate the SNR and assess signal quality.
Given: V_signal = 3.0 V (DC), V_noise = 0.03 V (RMS AC). Formula: SNR = 20 x log10(S/N)
Step 1: Calculate the voltage ratio
Ratio = V_signal / V_noise
= 3.0 / 0.03
Ratio = 100 : 1
Step 2: Calculate SNR in dB SNR(dB) = 20 x log10(100)
= 20 x log10(10²)
= 20 x 2
SNR = 40.0 dB
Step 3: Assess quality 40 dB is in the GOOD range (30 to 50 dB).
Signal is 100x stronger than noise in voltage.
VERDICT: GOOD. Signal suitable for all control modes including PID with derivative. If noise were instead 0.3 V (10x higher), SNR = 20 dB: ACCEPTABLE but derivative must be disabled. If noise were 1.0 V: SNR = 9.5 dB: POOR. Signal barely usable for monitoring only.
= 3.0 / 0.03
Ratio = 100 : 1
Step 2: Calculate SNR in dB SNR(dB) = 20 x log10(100)
= 20 x log10(10²)
= 20 x 2
SNR = 40.0 dB
Step 3: Assess quality 40 dB is in the GOOD range (30 to 50 dB).
Signal is 100x stronger than noise in voltage.
VERDICT: GOOD. Signal suitable for all control modes including PID with derivative. If noise were instead 0.3 V (10x higher), SNR = 20 dB: ACCEPTABLE but derivative must be disabled. If noise were 1.0 V: SNR = 9.5 dB: POOR. Signal barely usable for monitoring only.
Example 2: Wireless Transmitter Power SNR (Power Mode)
A WirelessHART transmitter reports a received signal strength of 0.5 mW. The ambient RF noise floor in the plant is measured at 0.001 mW. Calculate the power SNR and determine if the wireless link is reliable.
Given: P_signal = 0.5 mW, P_noise = 0.001 mW. Formula: SNR = 10 x log10(S/N)
Step 1: Calculate the power ratio
Power ratio = P_signal / P_noise
= 0.5 / 0.001
Power ratio = 500 : 1
Step 2: Calculate SNR in dB SNR(dB) = 10 x log10(500)
= 10 x 2.699
SNR = 27.0 dB
Step 3: Equivalent voltage ratio Voltage ratio = sqrt(Power ratio) = sqrt(500)
Voltage ratio = 22.4 : 1
Step 4: Assess quality 27 dB is in the ACCEPTABLE to GOOD boundary range.
WirelessHART requires SNR > 10 dB for reliable communication.
VERDICT: ACCEPTABLE. Wireless link reliable. No immediate action required. Monitor if process equipment changes or additional RF sources are added nearby.
= 0.5 / 0.001
Power ratio = 500 : 1
Step 2: Calculate SNR in dB SNR(dB) = 10 x log10(500)
= 10 x 2.699
SNR = 27.0 dB
Step 3: Equivalent voltage ratio Voltage ratio = sqrt(Power ratio) = sqrt(500)
Voltage ratio = 22.4 : 1
Step 4: Assess quality 27 dB is in the ACCEPTABLE to GOOD boundary range.
WirelessHART requires SNR > 10 dB for reliable communication.
VERDICT: ACCEPTABLE. Wireless link reliable. No immediate action required. Monitor if process equipment changes or additional RF sources are added nearby.
Example 3: Finding the Minimum Detectable Signal from SNR
A thermocouple amplifier has a noise floor of 0.5 µV RMS. The minimum required SNR for the temperature measurement is 30 dB. What is the minimum thermocouple signal level that will give acceptable measurement?
Given: V_noise = 0.5 µV RMS. Required SNR = 30 dB. Find: minimum V_signal.
Step 1: Convert required SNR to voltage ratio
Voltage ratio = 10^(SNR_dB / 20)
= 10^(30 / 20)
= 10^1.5
Voltage ratio = 31.62 : 1
Step 2: Calculate minimum signal from ratio and noise V_signal_min = Voltage ratio x V_noise
= 31.62 x 0.5 µV
V_signal_min = 15.8 µV RMS
Step 3: What does this mean in process terms? A Type K thermocouple produces approximately 41 µV per degree Celsius.
Minimum detectable temperature change:
dT_min = V_signal_min / (thermocouple sensitivity)
= 15.8 µV / 41 µV per °C
Minimum detectable temperature change = 0.39°C If the amplifier noise floor rises to 2 µV (due to poor shielding), the minimum detectable temperature change becomes 1.54°C for the same 30 dB SNR. This is why thermocouple shielding and amplifier selection are so important.
= 10^(30 / 20)
= 10^1.5
Voltage ratio = 31.62 : 1
Step 2: Calculate minimum signal from ratio and noise V_signal_min = Voltage ratio x V_noise
= 31.62 x 0.5 µV
V_signal_min = 15.8 µV RMS
Step 3: What does this mean in process terms? A Type K thermocouple produces approximately 41 µV per degree Celsius.
Minimum detectable temperature change:
dT_min = V_signal_min / (thermocouple sensitivity)
= 15.8 µV / 41 µV per °C
Minimum detectable temperature change = 0.39°C If the amplifier noise floor rises to 2 µV (due to poor shielding), the minimum detectable temperature change becomes 1.54°C for the same 30 dB SNR. This is why thermocouple shielding and amplifier selection are so important.
Common Sources of Noise in Process Plant Instrument Loops
| Noise source | Typical frequency | Signals affected | Typical SNR impact | Prevention method |
|---|---|---|---|---|
| Variable frequency drives (VFDs) | 1 kHz to 100 kHz switching harmonics | 4-20 mA, thermocouple, RTD, pulse signals in nearby cables | Severe: can reduce SNR by 20-30 dB in adjacent cable | Separate VFD power cables from signal cables by minimum 300 mm. Use ferrite beads on signal cables. Route at 90 degrees where crossings are unavoidable. |
| Power cables (50/60 Hz mains) | 50 Hz or 60 Hz and harmonics | Low-level signals: thermocouples, RTDs, strain gauges, pH sensors | Moderate: 10-15 dB reduction if cables run parallel | Maintain minimum separation (300 mm for 4-20 mA, 600 mm for thermocouple). Use twisted-pair shielded cable. Terminate shield at one end only. |
| Electromagnetic interference (EMI) from motors | Broadband up to several kHz | All analog signals, especially unshielded or poorly grounded | Significant: depends heavily on proximity and shielding | Proper cable shielding (drain wire to earth at control room end). Keep signal cables away from motor starter wiring. Use differential signal inputs where possible. |
| Ground loops | 50/60 Hz (mains frequency) | All analog signals when grounded at both ends | Moderate to severe: 5-25 dB loss depending on potential difference | Ground signal cable shields at one end only (control room/panel end). Use galvanic isolators for long cable runs across different earth potential areas. |
| Radio frequency interference (RFI) | MHz to GHz range | Electronic transmitters, HART signals, wireless devices | Variable: can disrupt HART communication overlaid on 4-20 mA | Ferrite clamp-on filters on signal cable entries to enclosures. Proper bonding of enclosures to earth. Follow ATEX/IEC 61000 EMC guidelines. |
| Thermal noise (Johnson noise) | Broadband (white noise across all frequencies) | All low-level signals: thermocouples, millivolt sensors | Fundamental limit: cannot be eliminated, only reduced | Use lower impedance sources. Reduce bandwidth with filtering. Lower operating temperature of front-end amplifier. P_noise = 4kTBR (Boltzmann constant × temperature × bandwidth × resistance). |
Practical Methods to Improve SNR in Field Installations
| Method | SNR improvement | Cost / effort | When to apply |
|---|---|---|---|
| Correct shield termination (drain wire to single-end earth) | 10-30 dB typical | Very low | Always. Check every cable in the loop. The single most common cause of poor SNR in field installations. |
| Separate signal and power cables | 10-20 dB | Low (at installation) | During design and cable routing. Minimum 300 mm separation for 4-20 mA. 600 mm for millivolt signals. |
| Use galvanic isolation on long cable runs | 15-25 dB (eliminates ground loops) | Medium | Whenever signal cable spans more than 50 m or crosses between separate earthing areas. |
| Switch from single-ended to differential input | 20-40 dB (common-mode rejection) | Medium (input card change) | In electrically noisy environments where all single-ended inputs are affected. |
| Low-pass filter on PLC/DCS analog input | 6-15 dB on high-frequency noise | Very low (software parameter) | As a supplement to hardware measures. Set filter time constant to minimum that does not affect process control response. |
| Use 4-20 mA instead of voltage signal | Immune to resistive voltage drops | Low | For any signal run over more than a few metres. Current signals are inherently more noise-immune than voltage signals. |
Advertisement
Advertisement
Further Reading and External Resources
Trusted external resources on SNR and signal quality in instrumentation
- IEC 61000: Electromagnetic Compatibility (EMC) Standards. The international standard series governing electromagnetic interference, emissions and immunity requirements for industrial instrumentation and control equipment.
- Emerson: WirelessHART Technology and SNR in Industrial Wireless. Technical guidance on signal strength, SNR requirements and installation best practices for WirelessHART field devices from a leading supplier.
- AutomationForum: SNR Calculator for Instrumentation. The reference article that inspired this guide, from a respected instrumentation engineering knowledge base.
- Analog Devices: The Data Conversion Handbook. Comprehensive technical reference on SNR, noise, ADC resolution and signal chain analysis from one of the world's leading analog integrated circuit manufacturers.
Frequently Asked Questions: Signal-to-Noise Ratio in Instrumentation
What is a good SNR value for a 4-20 mA instrument loop?
For a 4-20 mA loop used in process control, an SNR of 30 dB or higher is considered good. This means the signal is at least 31.6 times stronger than the noise (voltage ratio). SNR of 20 dB (10:1 voltage ratio) is acceptable for PI control but derivative action should be disabled. Below 20 dB the signal quality is marginal and below 10 dB the signal is unreliable for any control purpose.
Why does the voltage SNR formula use 20 and the power formula use 10?
Both formulas produce the same dB result for the same physical signal, because power is proportional to the square of voltage (P = V²/R). When you express the voltage ratio in dB using the power formula, you get 10 × log10(V²/V_n²) = 10 × 2 × log10(V/V_n) = 20 × log10(V/V_n). The factor of 2 from squaring the voltage becomes the factor of 20. Never mix the formulas: apply 20 × log10 to voltage or current ratios, and 10 × log10 to power ratios.
What causes poor SNR in a 4-20 mA loop?
The most common causes in process plants are: incorrect cable shield termination (most common: shield grounded at both ends creating a ground loop), signal cable running parallel to power cables or VFD cables, missing or damaged cable shield, ground potential differences between transmitter and panel earth points, and electromagnetic interference from nearby motors or switchgear. Correct shield termination to earth at one end only is the single fix that solves the majority of poor SNR cases at zero cost.
What is the difference between SNR and SINAD?
SNR compares signal power only to noise power. SINAD (Signal to Noise And Distortion) compares signal power to the combined power of both noise and harmonic distortion products. SINAD is used in audio and ADC testing where harmonic distortion from nonlinearity is a significant impairment alongside random noise. In process instrumentation, SNR is the standard measure. SINAD is mainly relevant when evaluating the data conversion quality of analog input modules in PLC and DCS systems.
How does SNR affect PID control loop performance?
SNR directly affects how much noise reaches the PID derivative action (Kd). Derivative action amplifies the rate of change of the error signal. If noise is present, the derivative term amplifies the noise into high-frequency control output variations that drive the control valve back and forth at high frequency. This is called valve chatter or hunting. The fix is either to increase SNR (fix the noise source) or disable derivative action and use a PI controller instead. Proportional and integral action are much less sensitive to noise than derivative action.
How is SNR measured in the field on a 4-20 mA loop?
Connect a true-RMS multimeter across the loop burden resistor (or directly in series with the loop if the meter has a suitable current measurement mode). Set the meter to DC and record the signal level. Then switch the meter to AC and record the RMS noise voltage. Calculate SNR = 20 × log10(DC signal / AC noise). Alternatively, use a loop calibrator with a spectrum analyser function, or simply observe the stability of the value in the DCS historian: a stable reading indicates good SNR, a noisy dancing reading indicates poor SNR.
Related articles on this site
- 4-20 mA Current Loop Explained: How It Works, Wiring and Troubleshooting
- Analog vs Digital Signals in Instrumentation: A Complete Guide
- 4-20 mA Signal Conversion: Percentage to mA Calculator and Formula
- HART Protocol: How It Works and How to Use a HART Communicator
- PLC Analog Scaling and Raw Count: Formula, Calculator and Brand Reference
- Instrument Loop Checking: A Complete Step-by-Step Procedure
What we learn today
- SNR(dB) = 20 x log10(V_signal / V_noise) for voltage signals. SNR(dB) = 10 x log10(P_signal / P_noise) for power signals. The factor of 20 vs 10 comes from the fact that power is proportional to the square of voltage: squaring the voltage ratio inside the logarithm doubles the multiplier from 10 to 20. Never mix the two formulas.
- SNR quality thresholds for process instrumentation: below 10 dB is unacceptable, 10-20 dB is poor (monitoring only), 20-30 dB is acceptable for PI control, 30-40 dB is good for all control modes, above 40 dB is excellent. Every 20 dB increase doubles the voltage ratio ten times (×10). Every 6 dB increase doubles the voltage ratio.
- The most common cause of poor SNR in 4-20 mA field loops is incorrect shield termination: shield grounded at both ends creates a ground loop at mains frequency. The fix is always to ground the shield drain wire at one end only (the control room or panel end). This single correction resolves the majority of noise problems at zero cost.
- SNR directly affects PID derivative action: derivative amplifies the rate of change of the error signal, which means it also amplifies any noise, causing valve chatter. For loops with SNR below 30 dB, disable derivative action and use a PI controller. Fix the noise source (shielding, routing, grounding) before enabling derivative action.