Employee
Power modules are the core component of high-power electronics systems. As such, their lifetime has an important effect on the reliability of the final product. In order to understand how real mission profiles affect the power modules, complex simulations have to be performed, often pushing the tools and the methodology to their limits due to very long and highly variable profiles. For this reason, a high level of expertise in simulation methodology and semiconductor characteristics is required. Knowing the behavior of power modules in real usage profiles has positive effects on the selection of the right power module and therefore on the reliability and optimization of the system costs.
The lifetime of the power module is the lifetime of the power system
A high-power system is composed of several components such as: capacitors, inductors, sensors, control boards, gate units, and of course, power modules. All of them have a limited lifetime and are sources of failure when they are not well suited for the application. For that reason, it is highly important to understand their behavior under specific operating conditions before selecting a component.
Power modules are not the exception, their lifetime is closely tied to their mission profile. Therefore, to estimate their lifetime, it is necessary to calculate the cycling load during the converter operation under real conditions. It is possible to estimate the lifetime of a power module using an experimental platform and running the system over a long period of time. However, this solution requires hardware implementation and long-time consuming tests. That is why different approaches using simulation tools allow users to accelerate and streamline lifetime calculations.
To simplify this step in the design process of industrial manufacturers, Infineon Power Simulation Platform (IPOSIM) offers automated Lifetime Estimation as a new premium feature. Now designers can digitally estimate the lifetime of power modules according to their application requirements.
The IPOSIM Lifetime Estimation service is a fully automated process which synthetizes all our semiconductor expertise, combined with state-of-the-art methodology and advanced programming knowledge. This service gives designers access to Infineon‘s power electronics expertise online, 24/7, and as often as needed.
Figure 1: Automated lifetime estimation process
The process (Figure 1) starts with relevant information about the desired mission profile, cooling conditions, and the power module to be tested. Considering this information, the calculations to estimate the lifetime of the device will run automatically in IPOSIM. Once finalized, users can download a customized PDF report containing the most relevant results, including the possible number of cycles for the selected device.
Struggling to select the right power module for a traction application? Not anymore
Let’s imagine we are designing a power converter for a traction application. We have already chosen a topology, fixed the operating conditions, defined the cooling condition, and we found out that we need a power module with the following ratings (1.7kV@1.2kA). It is time to decide which power module fits best to this application.
For this example, we will make some assumptions just to simplify the design process. Let’s say then that the only package that can be used is a PrimePACK™. After several technical considerations, the following three devices are our candidates:
- FF1200R17IP5
- FF1400R17IP4
- FF1500R17IP5R
What about the mission profile? Well, let us assume that the round way trip of the train takes something about two hours and we need to guarantee at least ten thousand cycles. Figure 2 depicts the mission profile, translated into power electronics variables, of a standard cycle of a train operation.
Figure 2: Mission profile of a traction application
We used the IPOSIM Lifetime Estimation service to estimate the lifetime of these three devices. Once the analysis concludes, IPOSIM generates a PDF report highlighting the most relevant results as depicted in the Figure 3. The power losses allow us to get an idea about the conversion efficiency. With the temperature results, we can validate if the selected device fits well to our cooling conditions and guarantees a proper operation during the complete mission profile. Lastly, the histograms depict the lifetime consumption depending on the temperature swing.
Figure 3: Simulation results for a traction application using the FF1500R17IP5R power module
Selection of the right power module
We have seen some remarkable results; however, we have not yet selected the right component based on the number of cycles criteria. That is why the report includes a final table summarizing the possible number of cycles per device. The following table shows a comparison between the evaluated devices.
|
Power cycling IGBT |
Power cycling Diode |
Thermal cycling |
|
|
FF1200R17IP5 |
10609 |
3236 |
n/a |
|
FF1400R17IP4 |
n/a |
n/a |
n/a |
|
FF1500R17IP5R |
30078 |
10536 |
n/a |
As can be seen, even if the three devices are rated in terms of electrical parameters, just the FF1500R17IP5R achieved the desired number of cycles (>10k cycles). It is important to point out that the FF1400R17IP4 did not generate any result. The reason is that with this power module under the evaluated conditions, the temperature increased more than the maximum limits. Therefore, the lifetime was not calculated because this device is not characterized over the temperature ranges.
How to get started with Lifetime Estimation
In this short article, we explained how and why IPOSIM Lifetime Estimation service enables fast and precise decision making during the design phase, saving time and effort for design teams. This service starts with one of the most commonly used topologies in the power electronics design: a module-based, three-phase two-level inverter. Other topologies will also be added in the near future.
For more information visit IPOSIM webpage here, where you can find a detailed user manual and several guides on how to use the service.
Links to explore further:
