Undisputed, the current safety standards for high voltage batteries address the chemical and thermal performance of battery cells during mechanical loads, i.e. pressure forces and intrusion. However, they do not represent the typical loads to the battery in vehicle crashes:​

• The battery intrusions specified in the standards, namely 50 % of the battery dimension, cannot be achieved with the typical battery on standard compression machines due to the high forces needed.
• The maximum forces specified in the standards, namely the thousandfold of the battery weight, are unrealistically high even for small batteries in mild hybrid vehicles (i.e. the 24 kg battery of the Mercedes-Benz S 400 HYBRID). The loads applied to the battery rarely exceed 200 kN.

Even with 240 kN applied to the battery package, the battery intrusion achieved is only approx 11 %, which is well below the targeted 50 %.

There are two main differences between the loads applied to the battery in a vehicle crash versus the quasi-static battery tests:​ 1. Due to the crash propagation, the load is applied indirectly by the surrounding structure and components via multiple and distributed load paths;​ 2. Due to the short period of the peak loads, the battery can withstand much higher dynamical forces than the maximum static loads.

In order to assess the safety performance of HV batteries in severe crashes more realistically, a comprehensive series of dynamical impact tests was conducted with all types and sizes of HVbatteries used in the current Mercedes-Benz hybrid and electric vehicles. The load profiles were derived from both, the relevant vehicle crashes, and the quasi-static battery standards, applying even higher loads and battery intrusions. The tests were conducted at the crash test facility of the TÜV SÜD, utilizing two different test methods:​

• The moving battery hitting an impactor attached to the rigid barrier;​
• The moving impactor hitting the battery attached to the rigid barrier.

Despite the high loads and the resulting major battery intrusions, no thermal or electric reactions occurred, neither short circuits, nor electrolyte leakages, nor fire or explosion. The shock-proof protection was ensured in all tests. Given the very realistic test method along with the high loads applied, a very high crash safety performance could be demonstrated for all the batteries. Furthermore, the tests confirmed that there are major differences in the load characteristic between the quasi-static battery test standards, and the dynamic crash loads. As a result, more realistic component tests for traction batteries must be specified as soon as possible.