Resistance and continuity
Bolted connection resistance testing uses a low-resistance ohmmeter to measure the resistance of bus joints, lugs and splices across switchgear, busway, transformers, breakers and other equipment. A joint with too much resistance overheats under load, so the readings are compared with similar joints and with earlier results to find loose or contaminated connections before they fail.
A bolted connection resistance test measures the electrical resistance across a joint where two conductors are held together by bolts, such as a bus-to-bus splice, a cable lug on a terminal pad, a busway plug-in joint, a transformer bushing connection, or a grounding lug. The measurement is made with a low-resistance ohmmeter, often called a micro-ohmmeter, which passes a steady direct current through the joint and reads the voltage drop across it with separate sense leads. The result is a very small resistance. Because the instrument measures only the joint and a short stretch of conductor on either side, the test isolates the quality of the connection from the rest of the circuit. It is called for across a wide range of equipment in the published acceptance and maintenance standards: switchgear and switchboards, metal-enclosed and outdoor bus, assembled busway, circuit breakers and switches, transfer switches, transformers and reactors, capacitors, drives, chargers, batteries, motor starters, grounding systems and other assemblies where a bolted path carries load current.
Bolted joints are one of the commonest places for a power system to develop trouble. Thermal cycling loosens bolts, aluminum and copper creep under clamping force, oxide films and contamination build up on contact surfaces, and installation errors such as improper torque, wrong hardware or dissimilar metals leave a poor contact from the start. A joint that carries current through a small effective contact area behaves like a resistor, and the heat produced rises with the square of the current. A joint that is only slightly warm at light load can become very hot at full load, which accelerates oxidation, which raises the resistance further, and the loop continues until insulation is damaged, a connection burns open or an arc starts. The consequences can be an unplanned outage, a fire, or an arc flash incident. Measuring joint resistance finds the weak connections while they are still cheap to fix, before an infrared survey shows a hot spot or a failure shows the same thing less gently. It also verifies the quality of installation work at acceptance.
The test relies on Ohm's law applied at very small values. A known direct current is forced through the joint, the voltage drop is measured across it with separate sense probes, and the ratio gives resistance. Four-wire sensing is essential: the current leads carry the test current, and the voltage is read at probes placed on either side of the joint, so lead and probe contact resistance does not enter the result. The resistance of a joint has two sources, the bulk resistance of the short length of conductor between the probes and the contact resistance at the interface. In a sound joint the contact part is small and close to that of identical joints. A loose, contaminated, corroded or under-torqued connection has a smaller effective contact area, so its contact resistance is higher. Because a joint's reading is affected by the geometry and material, the best reference is another joint of the same design, such as the same position on the other phases, and a high reading relative to its siblings stands out clearly. The probes must be placed the same distance apart and on the same surfaces for the comparison to hold.
The equipment is de-energized, isolated from every source, proven dead and grounded under the owner's lockout and work-permit procedures, with stored energy released and any control power that could re-energize the circuit removed. The instrument is selected for low-resistance work, its calibration is confirmed, and its leads are inspected. Current leads are clamped on either side of the joint, usually beyond the sense points, and the voltage probes are placed across the joint so that the measured path includes the connection and the minimum of metal around it. The contact surfaces under the probes are cleaned so that the probe contact is not the thing being measured. The direct current is applied, the reading is allowed to settle, and the result is recorded. Where a bus has several joints in a row, each is measured in turn using the same probe spacing. Joint conditions also matter: the circuit must be at rest, the conductor temperature is noted, and a joint that was recently tightened should be given its settling time before the final reading is taken. Joints that look off are inspected, cleaned or retorqued under the manufacturer's instructions and measured again, and the work and the before and after readings are written down. A calibrated torque tool is sometimes used instead for joints where measurement is impractical, but it is a different method.
The record identifies each joint tested by location and phase, together with its resistance reading, the instrument and its calibration status, the probe spacing or reference points used, and the date. It notes the condition of the equipment and the joint when found, for example clean, discolored, corroded or showing signs of heating, the ambient and conductor temperature, and any cleaning, retorquing, hardware replacement or re-measurement done during the visit with the values before and after. Information about the equipment class, the conductor material and the type of joint helps the next person judge the numbers, and it gives later tests a baseline to be trended against.
This page states no limit. Joints are judged mainly by comparison. The first comparison is between like joints, for instance the same position on each phase of a bus, where a reading that stands apart from the rest by a published proportion calls for attention. The second is with the manufacturer's data for the joint or equipment, where it is supplied. The third is trending against the baseline from installation and earlier maintenance visits, where a steady rise matters even when the joint still resembles its neighbors. A published value may also exist for particular equipment, and the governing tables are found in ANSI/NETA ATS-2025 for acceptance work and ANSI/NETA MTS-2023 for maintenance work, together with the way the standards define a significant difference between similar joints. A joint that fails the comparison is cleaned, retorqued and retested. Thermographic survey results and visual inspection add context to the decision. For the actual limits, consult your own licensed copies of those standards and the manufacturer's instructions.
Acceptance testing proves the installation work and sets the baseline for every joint, and it is the right moment to correct hardware or torque errors. Maintenance testing is a trend exercise, where the useful information is the change since the last reading, so repeatable probe positions and instruments matter most. The equipment class changes the practice. On assembled busway the joints are numerous and the housing limits access, so the test is carried out at each joint accessible through the inspection openings. On metal-enclosed switchgear and outdoor bus structures, joints sit behind barriers and the surveys are organized by compartment and phase. On breakers and switches, the bolted connections to the bus and cables are separate from the moving contacts, which are tested under contact resistance. On transformers, reactors and capacitors the bolted terminals are measured separately from the windings. On batteries the intercell connections are a close relative of this test. Some programs use infrared thermography between resistance surveys, as it can see joints under load that are hard to reach de-energized.
The equipment classes and standard sections that call for this test, and the UltraDb forms that carry it.
Current passing through extra resistance produces heat, and the heat grows with the square of the current. A joint that looks only warm at light load can run far hotter at full load, which accelerates oxidation and raises the resistance further, creating a runaway that ends in damage or failure.
A calibrated torque check confirms clamping force but does not show the quality of the contact surfaces, which can be contaminated or oxidized even at the right torque. Resistance measurement confirms that the joint conducts as it should, and many programs use both, with resistance as the electrical proof.
By comparison. A reading that sits well apart from other joints of the same design, such as the same position on the other phases, or that has risen since the last test, points to a poor connection. The standards define the size of difference that calls for action; check your licensed copy.
Bolted connection resistance covers fixed joints such as bus splices and lugs, while contact resistance refers to the separable moving contacts inside breakers, switches and contactors. Both use a low-resistance ohmmeter, but the equipment, the access and the way the results are compared differ.
UltraDb references ASTM, ANSI, IEEE, NETA, NFPA, NEMA, FAA, TIA, and IEC standards descriptively, to identify the published methods its forms are built from. UltraDb and Verlecta are not affiliated with, endorsed by, certified by, or licensed by NETA, ASTM, ANSI, IEEE, NFPA, NEMA, FAA, TIA, IEC, or Megger. Standard names and section numbers are the property of their respective organizations. Always test to the edition your contract specifies. Where a standard is sold through the ANSI webstore, Verlecta may earn a commission on a purchase made through our link, at no extra cost to you. A commission never decides which standards we cite or how a form is built.