Torque wrenches

The target torque should fall within 20 to 100 percent of the scale, since ISO 6789 only specifies accuracy in that span. In practice choose a wrench where your usual value sits mid-scale. For a 60 N.m bolt, a 20 to 100 N.m wrench is better than 60 to 340 N.m.

ISO 6789-1 allows 4 percent deviation for indicating types (Type I) and 6 percent for most setting click types (Type II) above 10 N.m. Wrenches are usually recalibrated every 5,000 cycles or 12 months, whichever comes first.

With click types, wind back to the lowest scale value (not below it) for storage so the spring does not fatigue under constant load. Do not use a torque wrench to loosen bolts, since breakaway torque often exceeds the scale and upsets calibration.

An adapter in line with the square drive axis does not change torque. A crowfoot or extension that lengthens the lever arm does, and the setting must be recalculated as T set = T required x L / (L + E). Always hold the handle at the marked load point.

Electronic types display actual values, can measure angle and store results for traceability, which torque-plus-angle procedures require. Mechanical click types are simple, need no battery and suit repeated tightening at one fixed value.

Prefer the value in the machine maker's documentation, since it accounts for the joint materials, gasket and lubrication. Tables by property class such as 8.8 or 10.9 are only a reference. Threads with oil or anti-seize need less torque than dry threads for the same clamp force.

Tighten in a star or diagonal cross pattern over several passes, for example 30 percent, 70 percent and then 100 percent of torque, followed by a circular check pass. This compresses the gasket evenly and avoids flange distortion. Fully tightening one bolt and moving to the next one along often causes leaks.

Impact can shift the spring mechanism or sensor without any visible sign. After a drop, check the wrench on a torque tester before using it on critical joints. Record the check result for traceability.