contact us
products
- Main Products
- IEC Test Probe
- UL Test Probe
- Material Flammability Tester
- IP Code Tester
- Impact Test Apparatus
- Security Testing Machine
- Lamp Cap Gauge Tester
- Lampholders Gauge Tester
- Plug & Socket Tester
- Electrical Safety Tester
- LED Test Instruments
- Environmental Test Equipment
- Instrument Accessories
- Weighing Sensor
- Others Testing Equipment
technical articles
Company News
Weighing Sensor Span and Zero: Why One Calibration Point Is Not Enough
A weighing system has two adjustable quantities and they fail independently. Zero is where the reading sits with no load. Span is how much the reading changes per kilogram. A single calibration mass adjusts one of them and assumes the other, and the assumption is where the error hides.
Two points are the minimum because two points define a line. One point defines a line only if you already know where it crosses the axis.
What the two adjustments do
| Adjustment | What it changes | How it drifts |
| Zero | The intercept of the line | Creep, temperature, mechanical set |
| Span | The slope of the line | Gauge factor, spring element, electronics gain |
| Linearity | The shape between the points | Not adjustable on most systems |
| Hysteresis | Difference between loading and unloading | Mechanical, not adjustable |
| Resolution | The smallest step the indicator shows | Set by the ADC, not by calibration |
| Repeatability | Spread of repeated readings | The real limit of the system |
Zero drift is the common one and it is usually thermal or mechanical. Span drift is rarer and more serious, because a span error grows with load: a 0.1 percent span error is 10 g on a 10 kg load and nothing at zero, so a zero check never finds it.
Why one point cannot separate them
Calibrate with a 10 kg mass and adjust until the display reads 10.000 kg. If the zero was already 20 g high, the adjustment has just changed the span to compensate, and the system now reads correctly at 10 kg and 20 g high at zero. Every load between is wrong by a fraction of that 20 g, and nothing in the procedure notices.
Two points, one near zero and one near full scale, separate the two: the zero point sets the intercept, the full-scale point sets the slope.
A calibration that holds
- Warm up before calibrating. Electronics drift for the first 30 minutes, and calibrating cold bakes the drift into the span.
- Exercise the sensor. Load to full scale three times before the calibration run, so the spring element and the fixtures settle.
- Use masses traceable to a standard, and record their actual values, not their nominal ones.
- Check the midpoint. A point at 50 percent that reads 2 counts off is linearity, and it is information, not a failure.
Monitoring between calibrations
Keep a check weight near the working load and use it daily. Plot the reading: a slow trend is drift and can be compensated, a step is a fault. Record the ambient temperature alongside, because most zero drift in an industrial installation is thermal and the correlation is obvious once the two curves are on the same chart.
If the sensor is used outdoors or in a cold store, specify the compensated temperature range and check at both ends of it. A weighing sensor with a stated temperature effect on zero and on span can be corrected in the indicator, and the site index lists the sensors by capacity and by protection class.
Standards referenced: OIML R76.





