Worked Example: Pack-Level FMEA

Busbar Joint Failure Modes and Contact Resistance

This is a worked example accompanying Failure Mode and Effects Analysis (FMEA) in Battery Design. It works through a pack-level DFMEA line on a bolted busbar joint, a component that is mechanically simple but disproportionately responsible for field failures because it sits at the intersection of high current, thermal cycling, and vibration.

Item and function

Bolted busbar joint connecting adjacent modules within the pack. Function: carries the full pack current path with minimal added resistance, while tolerating thermal expansion and vibration over the vehicle’s life.

Failure mode

Joint loosens over vibration and thermal cycling.

As with the separator example, this is written as a specific mechanism rather than “busbar fails.” A busbar can equally fail by fatigue cracking at a stress concentration, or by corrosion at the contact interface; each of those is a distinct line with its own cause and controls, even though several converge on the same effect of increased contact resistance.

Effect

As clamp force reduces, contact resistance at the joint rises. This produces localised heating at the joint under load, which is worst during high-current events such as fast charging or hard acceleration. Sustained or repeated heating can degrade the joint further (a self-reinforcing cycle, since heat itself accelerates stress relaxation in the clamped materials), and in a severe case can reach temperatures sufficient to damage surrounding insulation or, in combination with a fault elsewhere in the HV system, contribute to arcing.

Severity: 8. Not scored at the top of the scale because a single degrading joint typically presents first as a detectable performance or thermal anomaly rather than an immediate hazard, but it is scored high because the failure path does lead toward a genuine safety-relevant outcome (localised heating adjacent to HV conductors and insulation) if it goes undetected and unaddressed.

Cause

Insufficient initial clamp torque, or relaxation of the clamped joint over time as the busbar material and any compliant elements (such as Belleville washers, where fitted) settle under sustained thermal cycling.

A related but distinct cause worth its own line is dissimilar-metal contact at the joint (for example aluminium busbar against a copper cell tab) creating galvanic conditions that accelerate loosening through corrosion products building up at the interface; see Aluminium vs Copper Busbars.

Occurrence: 4. Bolted joints in automotive service routinely see years of thermal cycling and vibration, and torque relaxation in clamped joints is a well documented mechanism in bolted-joint engineering generally, not unique to batteries. A score of 4 reflects that this is a known and moderately likely mechanism rather than a rare one, tempered by the fact that a properly specified joint with adequate clamp force margin should not reach a failure condition within normal service life.

Current controls

Torque audit at assembly, verifying every joint is tightened to specification with a calibrated tool
Periodic DCIR-style resistance check in service, where the BMS or a service procedure can flag joints with rising resistance before they become a thermal problem
Visual/thermal inspection during scheduled service intervals

Detection: 5. A mid-scale score. Torque audit at assembly is a good control against an under-torqued joint from new, but it does nothing to catch loosening that develops later in service. In-service DCIR-style monitoring is genuinely useful if it exists and is checked, but many pack designs do not continuously monitor individual joint resistance in normal operation, meaning detection in practice often relies on the failure reaching a point where it affects a measurable higher-level parameter, or on a scheduled inspection catching it in time.

RPN and Action Priority

RPN = 8 × 4 × 5 = 160.

Under Action Priority scoring, Severity 8 combined with Occurrence 4 and Detection 5 bands as High, since both Occurrence and Detection sit above the low thresholds that would otherwise let a Severity-8 line band as Medium.

Recommended action

Specify Belleville washers or an equivalent compliant element at the joint to maintain clamp force despite thermal cycling and material settling, rather than relying on a rigid bolted joint alone
Add a torque-mark or torque-seal inspection point so under-torqued or disturbed joints are visually identifiable at any later inspection, not just at initial assembly
Where feasible, add continuous or periodic resistance monitoring at high-current joints as part of the BMS diagnostic set, rather than relying solely on scheduled physical inspection

After implementing the compliant element and torque-mark inspection, Occurrence would be expected to drop toward 2 to 3 (joints are less likely to loosen undetected) and Detection would improve toward 3 (a disturbed torque mark is straightforward to spot). Severity remains 8, since the consequence of an undetected failure is unchanged; only the likelihood of it occurring and going unnoticed has improved. This again illustrates why Action Priority tracks the mitigation of genuine risk more faithfully than RPN, which would show a large numeric drop (160 down to roughly 48 to 72) that could be mistaken for a bigger risk reduction than the Severity-driven picture actually supports.

Where this connects

Download the FMEA Worksheet Template to work through this and your own examples.