Battery Testing: Durability and reliability testing for batteries

Complete durability solutions for battery testing and secure vehicle integration

Very recently, I had to move from one place to another. You know how it goes: heavy! Of course, I had to put some stuff (carefully) in the boxes and I had to bring those with my car rather than putting them on truck. Sure, I had to do this several times. Now, I really wonder why we rented the truck. Anyway, topic for another time…

I should also admit that some people have the tendency of questioning many things (even life itself) while doing physically demanding type of jobs – count me in!

Then I realized that my car, which is a plug-in-hybrid, has a relatively small luggage compartment in a sedan car. I know that it’s due to the fact that battery electric cars have larger batteries and packaging those are not always very straightforward. But, are those that big that you can easily notice that they are taking some space from your luggage compartment?

Then, I made some research in order to see how big those battery packs might be on an electric car. Believe me, they can go very high! Here is one example I could find which shows the weight distribution of Tesla Model S.

So, the battery mass is 28.59% of the whole vehicle mass – it’s almost 1/3 of the whole vehicle!

If you’re also a durability addicted, same as me, I’m pretty sure that you also have the same question in mind:

How to validate battery packs?

Let’s turn this to even more complicated situation by multiplying the question.

1. How to safely integrate battery packs onto vehicles and make sure the whole vehicle survives for expected lifetime?
2. How to produce reliable batteries before implementing those onto vehicles?

At a first glance, we might think that these two questions are almost the same. But they’re not! The first question is more related to whole vehicle durability where we are more interested in battery integration. At this stage, it would be important to also take care of the location you put the battery on. For such heavy batteries (remember Tesla Model S weight distribution), battery mounts carry heavy loads and transmit them to the vehicle body. And in dynamic conditions while for instance driving on a rough road, high level of forces would be observed on and around batteries. To solve those kinds of potential problems, we need to have a structured durability engineering process where we cover the whole vehicle level. Typically, for those situations, we see high amplitudes of forces and displacements which are causing fatigue problems. And we mostly experience those high levels up to certain frequency level (traditional durability or fatigue characteristic; high amplitudes, low frequency).

The second question is more related to battery itself. So, as well as the road induced vibrations, we also need to think of the battery induced vibrations where electronics inside the battery packs can also act like vibration exposer factor. And again, in dynamic conditions, we might face unexpected durability failures due to both high loads transmitted to battery via battery mounts and due to the vibrations induced by the battery itself. For such situations, we should also think of vibration fatigue where the high frequency content of the loads might also play a critical role for potential durability failures.

In this blog, I’ll focus on the second part where we’ll see how we can bring reliable batteries to implement on vehicles.

Validate your design according to the standards

The first thing you can do is to validate your battery according to the pre-defined standards. There are already several standards available for battery testing such as IEC 62660, ISO 12405, SAE J2380 etc. These would already give you some battery testing profiles that you can use on your shaker while testing the batteries. These standards would already give you all you need:

• vibration mode (sine or random),
• axis to apply the loads,
• acceleration level,
• test duration…

It seems like you have everything to validate your battery design before implementing on your car!

These types of tests are executed on a shaker where you need a closed loop controlling strategy. Because you need to make sure that both the measured input and the output are on the desired level.

As Siemens, we help our customers with hardware and software solutions in order to reliably execute closed loop vibration qualification testing on shakers. And I should also indicate that solution we offer is sufficiently reliable and even used for satellite testing.

Are we good to implement the batteries which pass these standards? Well, I’d handle this very carefully, since I have another question in mind: Are the standards severe as my real customer usage?

If not, be ready for some durability and fatigue-based issues you might hear from field while customers start to drive your car. It’s also possible that you see zero issues from the field. But that might also be an engineering issue since the design might be over-engineered.

Go beyond standards: Design right vibration qualification tests!

As mentioned, standards give you the opportunity to test your batteries with some pre-defined test profiles. But you should also make sure that you test your better with right loads.

Battery testing with a right profile is vital in order to avoid over-testing or under-testing. Under-testing might lead to late durability issues which would be very expensive to fix. Over-testing might lead to high material cost and over engineering effort. Basically, we can count on three important criteria when creating vibration test profiles for battery validation:

• Rely on realistic loads – from real-life measurements rather than using just the standards
• Correctly determine – profile should be damage equivalent with real customer usage
• Accelerate test schedules – keep the damage but shorten the test duration

To help create the most optimum test profiles, we help our customers with our Simcenter SCADAS hardware and Simcenter Testlab software solutions. Simcenter SCADAS will help you acquire multi-physics data from real life conditions where you can see how your battery is affected by both road loads and battery induced loads. At the same time, Simcenter Testlab will allow you to configure your channels before starting the recording, analyze damage potential on acquired vibration signals and create synthesis test profiles that you can use for your battery. This approach is called Mission Synthesis. The workflow of Mission Synthesis is:

First question we need to ask is “What will happen to my product once customers use it?”.

To answer this question, we need to identify the operational environments of the vehicle. The example below shows four effective environments respect to distribution in total lifetime from a specific market.

We need to acquire relevant data from this different environments (missions). Typically, test engineers acquire vibration data from mounting locations of the test component to test it on a shaker. And they calculate Power Spectral Density (PSD) from acquired vibration data.

Simcenter SCADAS data acquisition hardware is designed to enable multi-domain acquisition campaigns to acquire analogue and digital data from a wide range of sensors including vibration, strain, force, displacement, microphone, video, GPS, temperature, pressure, voltage and so forth.

Acquired vibration data can already be used as an input to shaker table via PSD calculation. This refers tothe bullet saying: ‘relying on real life conditions, rather than standards’. But this is not enough to create a damage-equivalent and accelerated test profile. Now you may think: “How to quantify the impact of severe environments to my product?”

To better understand the durability impacts of different environments, we should go more into vibration fatigue, where we calculate the responses based on an excitation. So, we should focus on Fatigue Damage Spectrum (FDS), Maximum Response Spectrum (MRS) and Shock Response Spectrum (SRS). These calculations are for assessing the fatigue damage potential linked to high cycle fatigue. Assuming that the resonant frequencies from vibrations signal are considered as SDOF (Single Degree of Freedom) systems, where we take frequency and Q-factor damping into account. The responses (relative displacements) are calculated from each of these SDOF systems.

After understanding the damage potential, next step would be for test acceleration where we create random PSD or sine sweep profiles (or even combined) from calculated response spectrums (FDS, MRS and SRS). Simcenter Testlab enables an automated environment where the test duration is user definable while the software already considers the damage potential.

As a final step in the process, we mount the battery on a shaker. We fix it in the same direction as it will undergo the main vibrations in real-life. We load the synthesized test as calculated in the Simcenter Testlab Mission Synthesis software and start the vibration control test using the Simcenter Testlab vibration control modules by also using Simcenter SCADAS hardware.

We can easily think of two use cases for Mission Synthesis:

1. OEM perspective: OEM can use FDS to compare qualification evidence of potential battery suppliers.
2. Supplier perspective: Different OEM customers may give different qualification specifications (even with different types: PSD, sine sweep and different test duration…). Supplier can create single qualification for their battery testing by comparing different qualifications.

Conclusion

As mentioned, standards give you the opportunity to test your batteries with some pre-defined test profiles. But you should also make sure that you test your better with right loads. Efficient vibration qualification battery testing is vital, because of:

• Expectation for high quality, reliability and short test duration,
• Different environmental vibration (e.g. transportation, road impact, battery induced vibrations etc.) during lifetime,
• Under-testing will increase the chance of mechanical failure and add cost. Over-testing would lead to over-design and to over-cost.

Questions? Let me know, I’ll gladly reply any possible question. Just drop me an e-mail safak.has@siemens.com.