Busbar temperature rise: test vs calculation

What the check has to answer

The temperature rise check asks a simple question: at the rated current, in the real installation, will any part of the busbar assembly exceed its allowable temperature? The answer depends on the bar, the joints, the enclosure, the ambient and the heat produced by everything else in the cabinet.

There are two ways to answer it. A calculation builds a thermal model and predicts the rise. A test applies the current to a representative assembly and measures the result. The two are not competing methods. They answer the same question with different levels of evidence, and the right choice depends on the stage of the design and the risk.

What a temperature rise test measures

A temperature rise test is usually run on a complete assembly or a representative section. The rated current is applied, the ambient is recorded and the temperatures are measured at the points that matter: the bar surface, the joints, the supports and the terminals. The test continues until the temperatures stabilise.

The stabilisation criterion matters. A common rule is that the temperature rise does not change by more than about 1 K per hour, but the exact criterion comes from the applicable standard. A test that stops too early can understate the final rise, because the joints and the enclosure heat more slowly than the bar.

The ambient is not a single number. The air at the bottom of the enclosure, the air at the top and the air outside the enclosure can all differ. The test report should state where the ambient sensors were placed and how the rise was calculated. A rise measured against a cool spot at the bottom of the cabinet looks better than one measured against the hot air at the top.

• Rated current and the actual duty cycle
• Ambient sensor positions and the reference ambient
• Thermocouple positions on bar, joints, supports and terminals
• Enclosure, ventilation and mounting as installed
• Time to stabilisation and the criterion used
• Correction to the reference ambient, if the test ambient differs

What a calculation can do

A calculation uses a heat balance. The heat generated by the current is compared with the heat removed by convection and radiation. The inputs are the bar geometry, the surface area, the ambient, the allowable temperature and the heat transfer coefficients. For a simple bar in free air, the result is quick and useful.

The difficulty is the coefficients. Natural convection depends on the orientation, the surface, the surrounding surfaces and the air movement. Radiation depends on the emissivity of the surface, the view to cooler surfaces and the temperature difference. In a sealed enclosure, the air is stratified, the walls are warm and the bar may see other heat sources. A single coefficient is an approximation.

A more detailed model, such as a finite element or computational fluid dynamics study, can include the enclosure and the air flow. It costs more and still needs validation. The value of any model is that it lets you compare options before a test: a wider bar, a different support spacing, a vent or a lower current.

Why calculation and test disagree

The most common reason is the joint. A calculation often assumes a contact resistance, while the test measures the real joint. A joint with a poor contact face, insufficient torque or the wrong plating can run much hotter than the bar beside it. The test finds it; the calculation may not.

The second reason is the enclosure. A sealed box traps heat. A vented box has a chimney effect that depends on the vent positions. A bar mounted against a wall radiates to a warm surface instead of a cool one. A calculation that assumes free air will overestimate the cooling and underestimate the rise.

The third reason is the nearby heat. A busbar next to a transformer, a contactor or a resistor sees heat from those components. The test includes them if the assembly is representative. The calculation often does not.

When a calculation is enough

A calculation is enough when the installation is simple and the risk is low. A short bar in open air, a bar with a large margin over the thermal limit or an early concept comparison can be handled with a heat balance. The result should be documented with the assumptions and a margin for the unknown.

It is also useful when the design is changing quickly. A test on a design that will change next week is wasted. Use the calculation to narrow the options, then test the chosen design once.

If the calculation shows the bar well below its limit and the joints are conventional, the risk of a surprise is small. If the calculation is close to the limit, treat it as a warning and plan a test.

When to test

Test when the assembly is enclosed, when the current is close to the thermal limit, when the joint design is new or when the customer or the standard requires design verification. Test also when the busbar is part of a switchgear assembly that will be certified as a whole, because the rating belongs to the assembly.

A test is only useful if the sample is representative. The bar section, the plating, the joint detail, the support spacing, the enclosure and the ventilation should match the production design. A test on a bare bar in a laboratory does not verify an assembly in a cabinet.

For a new joint, test the joint as well as the bar. A thermocouple on the bar surface will not show a hot joint if the joint is inside a clamp. Put sensors on the joint faces and on the terminal, and inspect the joint after the test.

Reading and using the report

A useful report states the current, the ambient, the sensor positions, the stabilisation criterion, the measured temperatures and the correction to the reference ambient. It should also state the pass limit and where the limit comes from. A report that gives only a maximum temperature is hard to compare with another supplier.

Check the ambient correction. If the test was run at 25 °C and the installation will see 45 °C, the rise can be added to the higher ambient, but only if the rise is roughly independent of ambient. That is a reasonable first approximation, not an exact rule.

For production, the type test is not repeated on every unit. The control moves to the inputs: the bar material and section, the plating thickness, the joint flatness, the torque and the contact resistance. If those are held, the thermal performance should follow.

1. 01

   Define the duty

   State the current, ambient, allowable temperature and the standard or customer limit.

2. 02

   Model first

   Use a heat balance to choose the section and the support layout.

3. 03

   Build a representative sample

   Match the bar, joints, plating, enclosure and ventilation to production.

4. 04

   Instrument the hot spots

   Put sensors on the bar, joints, terminals and supports, and record the ambient positions.

5. 05

   Run to stabilisation

   Use the criterion from the standard and record the time.

6. 06

   Control production

   Hold the material, plating, flatness and torque that the test validated.