Key takeaways
- Vibration damage starts at the stiffest connection and at the joint, not in the middle of a straight run.
- A support spacing that keeps the first mode away from the excitation is worth more than a thicker bar.
- Every bolted joint in a vibrating assembly needs a preload strategy, because fretting starts when clamp force falls.
- A test regime should include a resonance dwell, not only a sweep, because the dwell is what accumulates cycles.
Where vibration damage starts
A busbar in a vehicle, a generator set or a piece of construction equipment sees vibration from the engine, the road, the hydraulic pump or the process itself. The bar is a mass on a spring, and the connections at each end are the stiff points where the motion is resisted. That is where the stress concentrates.
The first failure is often a fatigue crack at a bend, a hole or the edge of a welded joint. The crack starts on the tension side, at a change of section or at a burr. A second failure mode is fretting at the bolted joint: the bar moves a few micrometres under the washer, the plating wears and the contact resistance rises. The joint then runs hotter and the clamp force falls further.
A third mode is a broken support or a cracked bracket. The bar may survive while the bracket that holds it fails. The whole assembly has to be considered, not only the bar. A support that is stiff enough at the bench can be the weakest part once the mass of the bar and the cable is included.
Find the modes before choosing a fix
A sine sweep with an accelerometer on the bar shows where the response peaks. Those peaks are resonances. The frequency and the shape of each mode tell you which span is moving and which support is doing the least work. A mode that moves the whole bar as a rigid body is a support problem. A mode that bows a single span is a stiffness problem.
The test should be run at a low level first, so the bar is not damaged before the survey is complete. The accelerometer position matters: a sensor at a node can miss the mode that is causing the problem. Use several positions or a roving hammer check to map the shape.
Once the modes are known, compare them with the excitation. An engine has a firing frequency and its harmonics. A road input is a broad spectrum with most energy at low frequency. A pump has a blade pass frequency. The design aim is to keep the first significant mode away from the frequencies that carry energy, or to damp the response if the mode cannot be moved.
A resonance is not a failure by itself
A mode that is excited at low amplitude may be harmless. The question is the strain at the critical point and the number of cycles it sees. Measure the response, then judge.
Support spacing and stiffness
A span behaves like a beam. Its natural frequency rises as the span shortens and as the section stiffens. Doubling the number of supports on a run raises the frequency by roughly a factor of four for the same section, because the span length is halved in the beam stiffness term. That is a large change for a small amount of metal.
Shortening the span is usually more effective than thickening the bar. A thicker bar adds mass as well as stiffness, and the frequency rises with the square root of the stiffness to mass ratio. A support adds stiffness without adding much mass to the vibrating span. The support also has to be stiff in the direction of the motion, not only strong.
The support should clamp the bar without crushing it. A soft pad can reduce the local stress but also lowers the support stiffness. A hard clamp on a thin bar can dent the copper. A shaped block or a saddle spreads the load. The mounting fastener should be rated for the vibration and should not loosen.
Strain relief at the connection
The connection between a busbar and a cable, a terminal or another bar is a stiffness change. A cable that is tied close to the lug makes the lug the fixed end of a vibrating span. A service loop in the cable moves the restraint away and reduces the strain at the lug.
The same idea applies to the bar. A flexible section or a braid between two rigid bars isolates the vibration and takes the movement in a part that is designed to bend. The flexible section should be supported so it does not whip, and its bend radius should suit the movement. A flexible part that is clamped tight at both ends has nowhere to move.
Welded and brazed joints deserve a separate look. A weld toe is a stress concentration, and a weld that is not fully penetrated is a crack starter. If the joint sees vibration, the weld should be inspected and, where possible, the design should move the stress away from the toe with a thicker section or a smoother transition.
Joint preload under vibration
A bolted joint that loses preload starts to fret. The relative motion is small, often only a few micrometres, but it is enough to wear the plating and generate oxide debris. The debris raises the contact resistance, the joint runs hotter and the preload falls further. The cycle is self-reinforcing.
The defences are a stable preload and a joint that resists slip. A Belleville washer keeps the clamp force over a wider range of relaxation. A larger washer spreads the load and reduces the bearing stress. A thread locking compound or a locking feature stops the fastener rotating, but it does not restore a preload that has been lost.
The joint should also be checked for slip. If the lateral load from vibration exceeds the friction capacity of the joint, the bar will move however much preload is applied. A dowel, a key or a formed location can take the lateral load and leave the bolt to do its clamping job. That is often a better answer than a higher torque.
A test regime that matches service
The test should reproduce the frequency range, the amplitude and the duration of the service. A generic sweep at a low level is a survey, not a qualification. The qualification run should include the dwell at each resonance, because the dwell is where the cycles accumulate.
For road vehicles, ISO 16750-3 gives mechanical load profiles that are commonly used as a starting point. IEC 60068-2-6 covers sinusoidal vibration and IEC 60068-2-64 covers broadband random vibration. The programme standard, the mounting position and the vehicle class decide which profile applies.
A combined test is more realistic than vibration alone. Temperature changes the preload, the damping and the material strength. A joint that passes a vibration test at room temperature can fail when the same profile is run at the top of the service range. If the equipment sees both, the test should combine them or run them in sequence.
| Test step | Typical call | What it shows |
|---|---|---|
| Sine sweep | Low level across the service band | Resonances and mode shapes |
| Resonance dwell | Hold each mode for an agreed time | Cycle accumulation at the worst point |
| Random vibration | Profile from the application standard | Response to a broad spectrum |
| Thermal cycling | The service temperature band | Preload loss and differential movement |
| Combined test | Vibration with temperature | Interaction of load and movement |
| Post-test check | Resistance, torque and visual | Whether the part still works |
Illustrative steps. The levels and durations come from the programme standard and the mounting position.
Design review list
A vibration review is mostly a set of questions about stiffness, mass and load paths. The answers are often visible on the layout before any test is run. The test then confirms the design or finds the mode that the layout missed.
The review should include the supports, the joints, the connections and the cable. A bar that is well supported but has a cable tied tight to the lug is still a fatigue risk. A joint that is well preloaded but has no lateral location is still a fretting risk.
- List the excitation frequencies and the mounting position.
- Check the first mode against the excitation band.
- Review the support stiffness in the direction of motion.
- Give every bolted joint a preload and a slip strategy.
- Leave a service loop at cable and lug connections.
- Include a resonance dwell and a post-test resistance check.
Frequently asked questions
Where do busbars usually crack under vibration?
At a change of section, such as a bend, a hole or a weld toe, and at the connection where the bar is restrained. The crack starts on the tension side and grows with the number of cycles.
Does a thicker busbar always reduce vibration?
No. A thicker bar adds stiffness but also mass. Shortening the span or adding a support is often more effective, and it avoids adding weight and cost.
What causes a busbar joint to fret?
Relative motion under the clamp load. It starts when the joint slips or loses preload, and it wears the plating and generates oxide debris that raises the contact resistance.
Which standard covers busbar vibration testing?
It depends on the application. ISO 16750-3 is common for road vehicles, and IEC 60068-2-6 and IEC 60068-2-64 cover sinusoidal and random vibration. The programme standard decides the profile.
Sources and further reading
- ISO 16750-3:2012, Road vehicles, environmental conditions and testing for electrical and electronic equipment, mechanical loads
- IEC 60068-2-6:2007, Environmental testing, test Fc: vibration, sinusoidal
- IEC 60068-2-64:2008, Environmental testing, test Fh: vibration, broadband random
Qumek Engineering Desk
Engineering team
The engineering team behind Qumek drawings, DFM notes, tolerances and inspection standards. Every guide is reviewed against the drawings and processes used on live programmes.
Reviewing a busbar for a mobile application?
Send the layout, the mounting points and the vibration profile. The engineering desk will mark the modes and the joints that need attention.
Send a layout