Skip to content
New product 12 V 100 Ah sodium-ion truck starting battery, 1500 CCA, Group 31T See the launch →

Designing a Sodium Module: Busbar, Compression, Thermal

Home / News and Blog / Designing a Sodium Module: Busbar, Compression, Thermal
Designing a Sodium Module: Busbar, Compression, Thermal
16 Nov 2026

Most module failures are not chemistry failures. They are three specific mistakes, and each one has a number attached that can be checked before the first build.

Failure 1: the busbar sized for average current

The design used the continuous rating from the datasheet and a busbar chosen for that current. In service, the pack spent most of its time at half that load — and the joints still got hot, because heat comes from resistance and duty, not from the average printed on the label.

The fix was not a bigger busbar. It was a shorter current path and a bolted joint with a specified torque, checked after a week of cycling. Contact resistance at a joint is a manufacturing variable, not a design constant.

Failure 2: the stack that lost preload

A prismatic module was assembled to the specified compression and passed its first capacity test. Eighteen months later the customer reported a 15% capacity loss with no fault code.

The end plates had deflected under load and the tie rods had relaxed. Preload had fallen below the window the cycle-life data was measured in. Nothing in the BMS measures compression — which is why this failure is always found late.

Failure 3: the thermal path that assumed air

A module in a sealed enclosure relied on convection. It worked in the lab at 22 °C and derated in summer, which the customer read as a battery fault.

The real number was the enclosure’s internal air temperature, not the ambient. Measuring it took an afternoon; discovering it took a season.

The checklist that would have caught all three

Specify joint resistance and verify it after assembly. Specify preload and verify it after a thermal cycle. Measure internal air temperature, not ambient, and design the derating curve against it.

None of these is exotic. All three are cheaper than a field return.

The numbers in the specification

Stack torque appears as a test condition, not a footnote: the 7 Ah pouch cycle data is quoted at 1 N·m initial torque. Prismatic stacks are specified around 300 kgf per cell face. If your fixture cannot hold those numbers over a thousand thermal cycles, the published cycle life does not apply to your module.

Where to go next

Questions we get asked about this

How much compression does a prismatic stack need?

Commonly around 300 kgf per cell face, maintained across life. The exact figure comes from the cell datasheet and the test conditions behind the cycle claim.

How do I verify busbar joints?

Measure joint resistance after assembly and again after a week of cycling; contact resistance drifts with thermal expansion.

Should I measure ambient or internal temperature?

Internal air temperature inside the enclosure. Ambient will mislead you by 10 °C or more in summer.

Products for this application

EV packs 22 models · specs and sizes 4850 — 48 V 50 Ah sodium-ion pack single product · full specification

Have a specification question this article does not answer? Send it with the numbers you already have.

Request a quote

Part of

More guides in this group