A bridge can show a voltage even when its sensor has stayed at the same resistance. Build a quarter Wheatstone bridge with an ordinary resistor standing in for a sensor. Add two known resistances to stand in for long leads, then move the voltmeter's sense point to see how a three-wire connection changes the offset.
A Wheatstone bridge is two voltage dividers that share an excitation voltage. The output is the difference between their midpoints. In a quarter bridge, one of the four resistors represents the active sensor. This fixture changes that resistor by a known 10 Ω step so you can distinguish the response to the sensor from the shift caused by wiring.
For an upper resistance U and a lower resistance L, the midpoint voltage is VEXC × L ÷ (U + L). Subtract the reference midpoint, nominally VEXC ÷ 2, to obtain the bridge reading. P and A are different places on the same current path.
| Condition | Active U | Active L at base | Base | With 10 Ω step |
|---|---|---|---|---|
| Direct, P | 1000 Ω | 1000 Ω | 0.00 mV | 12.44 mV |
| Two-wire, P | 1000 Ω | 1009.4 Ω | 11.70 mV | 24.02 mV |
| Three-wire, T | 1004.7 Ω | 1004.7 Ω | 0.00 mV | 12.38 mV |
| Unequal, T | 1004.7 Ω | 1006.9 Ω | 2.73 mV | 15.09 mV |
At P, both force-lead drops are below the red clip, so both resistances add to the lower arm. At T, the meter reads the remote midpoint through its own sense lead. One force-lead resistor is now above the sensed midpoint and the other is below. With equal force leads and equal nominal bridge resistors, their voltage drops cancel in the zero reading. With unequal leads, the remaining offset depends on their difference. The sense resistor carries only the small voltmeter current. The bridge output sees a source resistance of roughly 1 kΩ, including the two dividers and the sense resistor. A 10 MΩ meter therefore changes the differential voltage by roughly 0.01% in this fixture.
direct,base with the signed bridge millivolts and actual excitation volts. Small residual offset can come from the four resistor tolerances and meter offset.direct,plus10. This resistor changes the sensor stand-in's resistance without changing either lead. Remove it before starting the next base measurement.two_wire,base and two_wire,plus10, again inserting 10 Ω only for the second observation.three_wire,base and three_wire,plus10. Finally, add 2.2 Ω in series with RB alone and repeat both measurements as unequal. The three-wire sense path from A to T stays in place.Use voltage regulation for all readings and accept excitation from 4.90 V to 5.10 V. Pause for the display to settle after each power change. If the output changes sign unexpectedly, verify red is at P or T and black is at N. Preserve a small negative offset if it is repeatable. If it is unclear where the bridge starts to differ, return to the direct condition and check the four resistor values. Verify that the bridge meter references N and the excitation meter references G0, with the two nodes kept separate.
Copy the blank worksheet and enter the measured excitation in volts and bridge output in millivolts, including the sign. Download the ZIP with the editable diagram, complete program and tests. Run python3 bridge_leads.py observations.csv. The analyzer divides output by the actual excitation so that the result is in millivolts per volt. It reports each wiring condition's sensor response as the 10 Ω result minus the base result. It also reports base offset changes across the wiring conditions.
For Fluke 115 meters within one year of calibration at 18–28 °C and 0–90% relative humidity, the voltage accuracy used here is ±(0.5% of the reading + 2 counts). One count is 0.1 mV on the bridge meter and 1 mV on the excitation meter. The code evaluates each ratio at the four combinations of meter bounds, then carries the low and high limits into each subtraction. The bounds cover meter specification, with resistor mismatch and drift left as separate influences. If a difference interval includes zero, the meter readings do not resolve a nonzero difference under that model.
Compare the large two-wire base offset with the smaller equal-lead three-wire offset and the return of an offset when you add 2.2 Ω to RB. The lead resistance still changes the slope of the bridge response. The predicted equal-lead step is only about 0.06 mV smaller than the direct step at 5 V, below the display count of this setup. The experiment does not measure that small difference, calibrate strain, or compensate for the temperature of a real sensor.
#!/usr/bin/env python3
"""Compare two- and three-wire quarter-bridge measurements."""
import argparse
import csv
import json
import math
from pathlib import Path
import sys
FIELDS = ("wiring", "sensor", "excitation_v", "bridge_mv")
WIRINGS = ("direct", "two_wire", "three_wire", "unequal")
SENSOR_STATES = ("base", "plus10")
EXPECTED = {(wiring, sensor) for wiring in WIRINGS for sensor in SENSOR_STATES}
def bounded(raw, minimum, maximum, label):
try:
value = float(raw)
except (TypeError, ValueError):
raise ValueError(f"{label}: enter a number") from None
if not math.isfinite(value) or not minimum <= value <= maximum:
raise ValueError(f"{label}: expected {minimum:g} to {maximum:g}")
return value
def nominal_ratio(wiring, sensor):
"""Nominal differential output in millivolts per volt."""
upper = 1000.0
lower = 1000.0 + (10 if sensor == "plus10" else 0)
if wiring == "two_wire":
lower += 2 * 4.7
if wiring in ("three_wire", "unequal"):
upper += 4.7
lower += 4.7 + (2.2 if wiring == "unequal" else 0)
return 1000 * (lower / (upper + lower) - 0.5)
def reading(key, excitation, bridge):
excitation = bounded(excitation, 4.90, 5.10, "excitation_v")
bridge = bounded(bridge, -100.0, 100.0, "bridge_mv")
# Fluke 115 DC: 0.5% of reading plus two counts at the selected range.
excitation_error = 0.005 * excitation + 0.002
bridge_error = 0.005 * abs(bridge) + 0.2
corners = [
b / e
for b in (bridge - bridge_error, bridge + bridge_error)
for e in (excitation - excitation_error, excitation + excitation_error)
]
return {
"excitation_v": excitation,
"bridge_mv": bridge,
"measured_mv_per_v": bridge / excitation,
"meter_interval_mv_per_v": [min(corners), max(corners)],
"nominal_mv_per_v": nominal_ratio(*key),
}
def read_worksheet(stream):
reader = csv.reader(stream, strict=True)
if tuple(next(reader, ())) != FIELDS:
raise ValueError("use the exact worksheet headings")
points = {}
for number, row in enumerate(reader, start=2):
key = tuple(row[:2])
if len(row) != len(FIELDS) or key not in EXPECTED:
raise ValueError(f"line {number}: unknown wiring, sensor state or column count")
if key in points:
raise ValueError(f"line {number}: repeated reading {key}")
try:
points[key] = reading(key, row[2], row[3])
except ValueError as error:
raise ValueError(f"line {number}: {error}") from error
if points.keys() != EXPECTED:
raise ValueError("need all eight observations")
return points
def subtract(earlier, later):
old_low, old_high = earlier["meter_interval_mv_per_v"]
new_low, new_high = later["meter_interval_mv_per_v"]
lower = new_low - old_high
upper = new_high - old_low
return {
"difference_mv_per_v": later["measured_mv_per_v"] - earlier["measured_mv_per_v"],
"meter_interval_mv_per_v": [lower, upper],
"nominal_difference_mv_per_v": later["nominal_mv_per_v"] - earlier["nominal_mv_per_v"],
"interval_contains_zero": lower <= 0 <= upper,
}
def analyze(stream):
points = read_worksheet(stream)
return {
"wiring": {
kind: {
"points": {sensor: points[kind, sensor] for sensor in SENSOR_STATES},
"sensor_change": subtract(points[kind, "base"], points[kind, "plus10"]),
}
for kind in WIRINGS
},
"two_wire_minus_direct_at_base": subtract(points["direct", "base"], points["two_wire", "base"]),
"three_wire_minus_direct_at_base": subtract(points["direct", "base"], points["three_wire", "base"]),
"unequal_minus_equal_three_wire_at_base": subtract(points["three_wire", "base"], points["unequal", "base"]),
}
def main():
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument("worksheet", type=Path)
args = parser.parse_args()
try:
with args.worksheet.open(encoding="utf-8-sig", newline="") as stream:
result = analyze(stream)
except (OSError, ValueError, csv.Error) as error:
print(f"Worksheet error: {error}", file=sys.stderr)
return 2
print(json.dumps(result, indent=2, allow_nan=False))
return 0
if __name__ == "__main__":
sys.exit(main())
Two 4.7 Ω force leads cancel in the nominal zero reading. If you increase only RB by 2.2 Ω, the lower arm increases while the upper arm stays the same. The sensed midpoint rises relative to the reference. A separate sense wire keeps the voltage drop in its own path small, but it cannot cancel unequal resistances in the current path.
See the four-wire resistance experiment for a different use of separate current and voltage connections. The program's tests use synthetic rows. Use the worksheet to record your own bench readings.