Pull tester gauge accuracy and calibration

A pull tester gauge used for BS 8539 site tests should read within 10% of the actual load. That is a site figure, not a laboratory one, but it is tight enough that an allowable resistance worked out from the readings can be trusted.

For scaffold tie tests, NASC TG4:25 asks for a gauge calibrated within the last 12 months to an accuracy of 5%. It does not say whether that is 5% of the reading or of full scale, so check which your calibration certificate states. Meet the 5% figure for scaffold tie anchors and whenever a specification calls up TG4:25. Use the 10% of reading figure for other BS 8539 tests, unless the specification sets another.

Ideally, gauges read directly in units of load, such as kilonewtons (kN), rather than in pressure.

A gauge's accuracy is often stated as a percentage of full scale, so the possible error in kilonewtons is the same at every reading. At low readings that fixed error becomes a large share of the reading.

Take a gauge with a 60 kN full scale and an accuracy of ±2% of full scale. Its possible error is ±1.2 kN at every reading:

Don't take readings below 20% of full scale on a gauge accurate to ±2% of full scale. Below that point the error can be more than 10% of the reading. Work out the test load first, for example with our test-load calculator. For a low test load, use a tester or gauge with a smaller full scale, or one whose accuracy is stated as a percentage of the reading.

On an analogue dial, no division should cover more than 5% of the full scale, and you can read to half a division. On a 50 kN gauge, that means divisions of 2.5 kN or less, read to 1.25 kN.

Digital readouts have largely replaced dials, and are usually far more accurate than this.

A gauge should keep showing the highest load the test reached after the reading has fallen back. Anchors relax under load, and an anchor that fails lets the load fall away, so without a peak reading the tester could record the wrong value. See first movement and load relaxation.

The Staht t25 and t60 show the live load and the peak load on the same screen.

Send a pull tester or gauge for calibration at least once a year, to a lab whose reference equipment is traceable to national standards. Send it sooner after a drop, or after any reading you doubt, such as an analogue needle that won't settle back to zero.

Keep the certificate that comes back with it, take it or the gauge's calibration chart to every test, and put the calibration date and the lab in the test report. See what goes in a site test report.

Every Staht t25 and t60 kit comes with a 12-month calibration certificate. Staht calibrates and services t25 and t60 testers and supplies a new 12-month certificate: see calibration and servicing and service and calibration. Annual calibration of a t25 or t60 can also be done digitally by a local calibration laboratory.

Each t25 and t60 calibration certificate records the tester's readings against a traceable reference across its range: check they cover the loads you test to.

For hydraulic test rigs, the G1 Pro digital gauge reads up to 1000 bar and is accurate to 0.25% of full scale (Staht Heavy Duty Test Kit brochure). State the calibration range you need when ordering. It reads pressure, so use it with a calibration chart that converts bar to kN for the ram it is fitted to.

A dial gauge for measuring movement must be accurate to 0.02 mm or better. Treat a dropped gauge, or one whose needle won't settle at zero, as out of calibration until it has been recalibrated. Stand it on its own base, never on the bridge, which flexes under load.