Improve Weld and Connection Inspection with the Hioki RM3545A

A welded connection can look consistent while its electrical performance varies. Low-resistance measurement adds a numerical result that can be compared across parts, production lots and process changes. The value becomes useful when the measurement location and conditions are controlled, so that a change in the reading reflects the assembly rather than a moving probe or changing fixture.

The RM3545A is intended for precision resistance measurement in development and production. This guide focuses on configuring a repeatable inspection process. It does not treat one resistance threshold as a universal definition of weld quality.

Choose the inspection approach around the part

Understand the 1 nΩ specification

HIOKI specifies 1 nΩ resolution in the 1000 μΩ range, with 1 A test current. Resolution describes the smallest displayed increment. It does not mean that any two parts differing by that amount can be distinguished reliably under all conditions.

For that range, a representative manufacturer accuracy specification is ±0.045% of reading ±0.010% of full scale, under the stated High-mode, OVC and SLOW2 conditions. Apply the complete expression and its conditions rather than quoting only the percentage of reading.

Compare the expected difference between parts with the repeatability and uncertainty of the complete measurement. If fixture movement produces a larger change than the defect being investigated, adding display digits will not solve the inspection problem. Establish a stable setup before setting a narrow comparator window.

Make probe placement repeatable

A four-terminal arrangement separates the current-driving connections from the voltage-sensing connections. This helps avoid including the current lead's voltage drop in the resistance result, but it does not make probe position irrelevant. The sensing points define the portion of the assembly included in the reading.

Document those points with a drawing or fixture specification. Keep contact location, force and surface condition consistent. Check whether nearby conductive paths allow current to bypass the joint of interest; the reading must represent the intended path if it is to support a useful decision.

Include repeated loading and unloading of the same part in a fixture trial. A stable reading without moving the probes does not establish repeatability after each production cycle. Use the trial to assess both the instrument settings and the mechanical arrangement.

Account for temperature and offset voltage

The RM3545A includes offset voltage compensation and temperature-related measurement functions. These tools address different effects and should be configured according to the material and test method.

A recently welded part may be cooling while it is measured. Comparing that result with a room-temperature reference without controlling the timing can mix thermal effects with connection differences. Define the interval between welding and testing, the temperature information needed and the basis for any correction.

Record correction settings alongside the result. A reviewer should be able to determine whether the reported value is the measured resistance at test temperature or a value adjusted to a reference temperature. Check that the selected correction model suits the actual assembly.

Select channel count and validate the full cycle

The RM3545A-1 is the single-channel model. The RM3545A-2 accepts up to two optional Z3003 multiplexer units for up to 20 four-terminal channels. Confirm the ordered switching hardware; the model suffix alone does not mean that every optional channel is installed.

Evaluate production throughput with the complete sequence: part positioning, contact verification, switching, settling, measurement, judgement and data transfer. An instrument's fastest measurement setting is not the same as the station's total cycle time, and faster settings may change the applicable measurement performance.

Define separate handling for an invalid contact, an interrupted measurement and a valid out-of-limit result. Save part identification, test-point identification, settings and judgement together. This prevents a communication or fixture problem from being mistaken for a defective joint.

Configuration support in Canada

Send RCC Electronics the expected resistance range, part geometry, number of test points and required cycle time. The RM3545A-1 product page is a starting point for single-channel requirements. Discuss fixture connections, integration needs and applicable calibration and service requirements before finalising a system.

FAQ: RM3545A weld and connection testing

Does a low resistance reading prove every aspect of weld quality?
No. Validate how the electrical result relates to the defects and performance requirements of the part. Other inspection methods may still be required by the production quality plan.

Is there one acceptable resistance limit for all welds?
No. Geometry, material and sensing locations affect the result. Establish limits using an appropriate validation study for the actual assembly and measurement method.

Does 1 nΩ resolution mean 1 nΩ accuracy?
No. Use the full accuracy specification and assess the fixture, temperature and repeatability contributions as part of the measurement system.

Should I select the fastest measurement setting?
Select the setting that meets both cycle-time and measurement requirements. Test it with representative parts and the intended switching and contact sequence.

What should a quotation request include?
Include resistance range, test-point count, fixture access, cycle time and communication requirements. State whether the system will evaluate prototypes or make automatic production decisions.

* Specifications are summarised from HIOKI RM3545A product information and the manufacturer application flyer. Confirm current settings, accessory compatibility and measurement limits. Application guidance does not establish universal acceptance limits.