What is the quarter-bridge strain gauge formula?
Vo / Vex = GF·ε / (4 + 2·GF·ε), which is close to GF·ε / 4 at ordinary strains. With GF = 2, 5 V excitation and 1000 µε, the exact output is 2.498 mV and the linear estimate is 2.5 mV.
From the strain, gauge factor and excitation, this strain gauge calculator returns the output of a quarter, half or full bridge. Run it as a microstrain calculator to work back from a measured voltage. The result uses the exact bridge equation, so the nonlinearity is included, and on a quarter bridge you can enter the lead resistance to account for long lead wires.
Positive for tension, negative for compression.
From the gauge package. Metal foil is usually near 2.
Commonly 120, 350 or 1000 Ω.
Optional: one conductor of the gauge cable.
Bridge output
2.4975 mV
Measures axial and bending strain together. A lone gauge has no temperature compensation; a dummy gauge in the adjacent arm (quarter-bridge type II) adds it without changing the output.
Positive strain is tension. The sign of the voltage depends on which output lead you call positive, so swap them if your reading is reversed.
Every arrangement at 1000 µε (GF 2, 5 V, ν 0.3)
| Bridge | × quarter | Output | mV/V | Nonlinearity |
|---|---|---|---|---|
| Quarter bridge: 1 active gauge | 1 | 2.498 mV | 0.4995 | 0.0999 % |
| Half bridge: axial + Poisson gauge | 1.3 | 3.248 mV | 0.6495 | 0.07 % |
| Half bridge: ±ε pair (bending) | 2 | 5 mV | 1 | 0 |
| Half bridge: 2 gauges, same strain, opposite arms | 2 | 4.995 mV | 0.999 | 0.0999 % |
| Full bridge: 2 axial + 2 Poisson (column) | 2.6 | 6.495 mV | 1.299 | 0.07 % |
| Full bridge: 2 axial + 2 Poisson (beam) | 2.6 | 6.5 mV | 1.3 | 0 |
| Full bridge: 4 active ±ε (bending or torsion) | 4 | 10 mV | 2 | 0 |
For a quarter bridge, Vo = Vex × GF·ε / (4 + 2·GF·ε). With a gauge factor of 2, 5 V excitation and 1000 µε, the output is 2.498 mV, against 2.5 mV from the common GF·ε / 4 shortcut. A half bridge with a ±ε pair gives 5 mV and a full bridge of four active gauges gives 10 mV.
The strain gauge equation is ΔR/R = GF × ε: a gauge bonded to a part changes resistance in proportion to the strain under it. Four arms in a Wheatstone bridge turn that small change into a differential voltage.
For a strain gauge Wheatstone bridge with four equal arms, the small-signal output is the sum of the arm changes, with adjacent arms subtracting:
The exact output depends on the arrangement. The quarter bridge is the one you will meet most often:
TN-507-1 writes each output as K·ε·(1 − η), where K is the small-signal sensitivity and η the nonlinearity. For the quarter bridge η = GF·ε / (2 + GF·ε). The same equation solved for strain is the correction for a reading:
The calculator uses the closed-form solution for every arrangement. It assumes a constant-voltage supply, a high-impedance amplifier, and equal arms across each output corner.
GF = 2 and ε = 1000 µε give ΔR/R = 0.002, which is 700 mΩ on a 350 Ω gauge. At 5 V: Vo = 5 × 0.002 / (4 + 2 × 0.002) = 2.498 mV. The linear estimate is 2.5 mV, so the nonlinearity is 0.0999 %, and each gauge dissipates 17.9 mW.
Two axial and two Poisson gauges on a column, ν = 0.30, GF = 2, 10 V excitation, ε = −500 µε. Then Vo = 10 × (−0.001 × 1.3) / (2 + (−0.001 × 0.7)) = -6.502 mV. The linear estimate is -6.5 mV. Nonlinearity is -0.035 %: negative because in compression the output magnitude runs above the linear line.
A 120 Ω quarter bridge at 5 V, GF = 2, wired with three conductors of 0.85 Ω each, reads 12 mV after zeroing, so V = 2.4 mV/V. The linear formula says 4800 µε. Correcting for the bridge alone gives 4823.2 µε, and including the leads gives 4857.3 µε. As a check, that strain produces 12 mV.
Output at 1000 µε with GF = 2 and ν = 0.30. The case numbers are those of TN-507-1, Table 1, and the names in the second column are NI’s. Nonlinearity of zero means the arm currents stay constant.
| Case | Arrangement | NI name | Vo / Vex | mV/V | Output at 5 V | Nonlinearity |
|---|---|---|---|---|---|---|
| 1 | Quarter bridge: 1 active gauge | Quarter-bridge I and II | GF·ε / (4 + 2·GF·ε) | 0.4995 | 2.498 mV | 0.0999 % |
| 2 | Half bridge: axial + Poisson gauge | Half-bridge I | GF·ε(1 + ν) / (4 + 2·GF·ε(1 − ν)) | 0.6495 | 3.248 mV | 0.07 % |
| 3 | Half bridge: ±ε pair (bending) | Half-bridge II | GF·ε / 2 | 1 | 5 mV | 0 |
| 4 | Half bridge: 2 gauges, same strain, opposite arms | none | GF·ε / (2 + GF·ε) | 0.999 | 4.995 mV | 0.0999 % |
| 5 | Full bridge: 2 axial + 2 Poisson (column) | Full-bridge III | GF·ε(1 + ν) / (2 + GF·ε(1 − ν)) | 1.299 | 6.495 mV | 0.07 % |
| 6 | Full bridge: 2 axial + 2 Poisson (beam) | Full-bridge II | GF·ε(1 + ν) / 2 | 1.3 | 6.5 mV | 0 |
| 7 | Full bridge: 4 active ±ε (bending or torsion) | Full-bridge I | GF·ε | 2 | 10 mV | 0 |
GF = 2. The error in percent is roughly equal to the strain in percent, as TN-507-1 notes. In compression the output runs high instead of low.
| Strain | Linear | Exact, tension | Error, tension | Exact, compression | Error, compression |
|---|---|---|---|---|---|
| 100 µε | 0.05 mV/V | 0.049995 mV/V | 0.01 % | -0.050005 mV/V | -0.01 % |
| 500 µε | 0.25 mV/V | 0.24988 mV/V | 0.05 % | -0.25013 mV/V | -0.05 % |
| 1,000 µε | 0.5 mV/V | 0.4995 mV/V | 0.0999 % | -0.5005 mV/V | -0.1 % |
| 2,000 µε | 1 mV/V | 0.998 mV/V | 0.2 % | -1.002 mV/V | -0.2 % |
| 5,000 µε | 2.5 mV/V | 2.4876 mV/V | 0.498 % | -2.5126 mV/V | -0.503 % |
| 10,000 µε | 5 mV/V | 4.9505 mV/V | 0.99 % | -5.0505 mV/V | -1.01 % |
| 20,000 µε | 10 mV/V | 9.8039 mV/V | 1.96 % | -10.204 mV/V | -2.04 % |
| 50,000 µε | 25 mV/V | 23.81 mV/V | 4.76 % | -26.316 mV/V | -5.26 % |
| 100,000 µε | 50 mV/V | 45.455 mV/V | 9.09 % | -55.556 mV/V | -11.1 % |
Vo / Vex = GF·ε / (4 + 2·GF·ε), which is close to GF·ε / 4 at ordinary strains. With GF = 2, 5 V excitation and 1000 µε, the exact output is 2.498 mV and the linear estimate is 2.5 mV.
Very little. At 1000 µε with GF = 2 and 5 V excitation, a quarter bridge gives 2.5 mV, a half bridge with a ±ε pair gives 5 mV, and a full bridge of four active gauges gives 10 mV, so an instrumentation amplifier or a high-gain ADC is normally needed.
The ratio of fractional resistance change to strain: GF = (ΔR/R) / ε. Metal foil gauges are usually close to 2. Use the value printed on the gauge package, because it is measured for that lot and scales every result directly.
When only one arm changes, the current through that side of the bridge changes as well, so the output is not quite proportional to ΔR. The error is 0.0999 % at 1000 µε and 0.99 % at 10,000 µε for GF = 2. Bridges where the resistance changes in each divider cancel, such as a ±ε pair, keep the arm currents constant and are exactly linear.
A quarter bridge is the simplest, but one gauge alone has no temperature compensation and reads axial and bending strain together. A half bridge with its gauges in adjacent arms, or a full bridge, compensates temperature and multiplies the output. Pick the arrangement by what you want to reject: a ±ε pair or a four-gauge beam bridge reads bending only, and the full-bridge column reads axial strain only.
With two wires, both leads are in the gauge arm. About 1.7 Ω of lead on a 120 Ω gauge unbalances the bridge by the equivalent of 7000 µε and costs 1.4 % of the sensitivity, and the offset drifts with cable temperature. With three wires, one lead sits in each adjacent arm, so the bridge stays balanced and the sensitivity loss is 0.7 %.
Divide the output change by the excitation to get Vo / Vex, then solve the bridge equation. For a quarter bridge, ε = 4(Vo/Vex) / (GF·(1 − 2·Vo/Vex)). Reading 12 mV on 5 V with GF = 2 gives 4823.2 µε, compared with 4800 µε from the linear formula.
The output scales directly with excitation, but so does the heat in each gauge: each arm of a balanced bridge sees half the excitation, so a 350 Ω gauge dissipates 17.9 mW at 5 V and 71.4 mW at 10 V. TN-502 sets the limit by power density, the power divided by the grid area: for moderate-accuracy static work, 2 to 5 W/in² on thick steel but only 0.2 to 0.5 W/in² on fiberglass or epoxy. A higher-resistance or larger gauge lets you use more voltage.
See all free electronics tools →
ProtoFlow is free AI PCB design software: describe a circuit and get a schematic with real parts, PCB layout with an autorouter, SPICE simulation and Gerber files. Native KiCad, Altium and EAGLE import and export. Free on macOS, Windows and Linux.
Free to start. No card required.