Testing ev battery packs & modules to ISO 16750-2
An aircraft bus is not a clean supply, and ISO 16750-2 is the document that says exactly how unclean it is allowed to get. You are testing cells, modules and full high-voltage packs on 40–1500 VDC, high current, proving capacity, cycle life and behaviour at the edges of the voltage window. The standard turns on supply-voltage code system (A–H) and cranking and starting profiles; it says nothing about EMC and radiated immunity — 4.13 points out to the EMC standards or high-voltage traction buses (LV 123 territory), which belong to a different qualification entirely. Reproducing it takes programmable DC source covering 0–36 V steady state with millisecond profile sequencing — cranking waveforms with 2 Hz ripple and 1 ms transient edges — plus superimposed AC ripple injection from 10 Hz to 200 kHz at up to 15 A peak-to-peak, slow ramps down to 0.5 V/min, and polarity reversal to −26 V. Micro interruptions need a series switch that opens in ≤ 10 µs to ≥ 10 MΩ, and load dump needs a pulse delivered from a defined 0.5–8 Ω source resistance at up to 202 V peak — a specified source-impedance pulse, not just a voltage setpoint.
What you are testing
- Device under test
- cells, modules and full high-voltage packs
- Bus
- 40–1500 VDC, high current
- Who buys this
- OEM and tier-one battery engineering teams
What ISO 16750-2 does not cover
- EMC and radiated immunity — 4.13 points out to the EMC standards
- high-voltage traction buses (LV 123 territory)
- 400 Hz aviation power
- reliability-duration testing
- grid ride-through
If your programme needs those, they come from a different standard and often different hardware.
The limits this test has to reproduce
A pack test spends most of its life inside the unlimited operating band. What decides the rig is the current at the top of the window and the behaviour at the edges.
| Parameter | Value | Clause |
|---|---|---|
| Superimposed alternating voltage | 10 Hz–200 kHz sweep, 2 % logarithmic steps; severity 1 Upp 6 V (12 V) / 10 V (24 V); severity 4 Upp 1 V at 30–200 kHz; ripple current limit 15 A (f1) / 10 A (f2) | Table 7 / Table 8 |
| Slow decrease and increase of supply voltage | UA → 0 V → UA at (0.5 ± 0.1) V/min, linear or steps ≤ 25 mV | §4.5.2 / Figure 6 |
| Micro interruption in supply voltage | tmicro 10 µs–2 s stepped from 10 µs; switch reaction ≤ 10 µs; open-switch resistance ≥ 10 MΩ | Table 9 / §4.6.1.2 |
| Reset behaviour at voltage drop | staircase in 5 % steps of USmin down to 0 V, ≥ 5 s per drop, functional check at each return to USmin | §4.6.2.2 / Figure 12 |
| Reversed voltage | −14 V (12 V) / −26 V (24 V) for 60 s (test case 2); −4 V for 60 s (test case 1, 12 V only) | Table 15 / Table 16 |
The clauses ev-battery programmes actually face, as published in ISO 16750-2, revision 2023 (fifth edition), replaces ISO 16750-2:2012. See every limit in ISO 16750-2. Certification work should be run against the published standard.
What decides the rig here
The requirements that separate a system that can run this programme from one that only meets the headline numbers.
Current, not voltage, sets the architecture
Pack-level discharge current drives busbar design, contact resistance and sense-lead discipline long before the voltage rating matters.
Regeneration is an economic decision
Dissipating pack discharge as heat is paid for twice — once at the meter and again in cooling load.
Channel count decides your calendar
Qualification throughput is a function of parallel channels, not system power. This is the number that slips schedules.
Other standards this DUT usually faces
Further reading on ISO 16750-2
Send us the clauses you have to satisfy
We will come back with the source and load capability that covers them — and tell you plainly if something in your spec needs hardware we would not supply.
Talk to an engineer