Tip / Sign in to post questions, reply, level up, and achieve exciting badges. Know more

cross mob

All you need is a simple cross-reference table” (or how to determine equivalent IGBT and MOSFET ratings)

lock attach
Attachments are accessible only for community members.

All you need is a simple cross-reference table” (or how to determine equivalent IGBT and MOSFET ratings)

_Klaus
Moderator
Moderator
Moderator
10 replies posted First comment on blog 10 sign-ins

How it all started

Some months ago, one of my marketing colleagues asked me to publish a cross-reference table on our webpage – one that maps the part numbers of our discrete IGBTs to part numbers of equivalent silicon carbide MOSFETs.

“Klaus, I need a cross-reference table for IGBTs and silicon carbide MOSFETs,” he said, when we passed each other in the corridor.
“What exactly do you want to see in that table?” I asked, assuming it was going to be an easy task – a mistake, as I was about to learn a minute later.
“Just the mapping between the nominal current of our IGBTs and the on-state resistance of our MOSFETs,” he responded.
“OK, but for what application and topology,” I asked, wanting to understand better.
He looked at me quizzically, “A general mapping is more than enough. Just a simple cross-reference table, no need to overcomplicate it.”

I somehow figured out that I should stop asking him questions to further clarify the design goals and boundary conditions. I was not supposed to “overcomplicate” it, anyway.

A general mapping

Based on the above-mentioned “requirements,” I did a series of calculations, compared them to existing measurement data, and summarized the findings in a training video. While the training video clearly states that an optimized mapping is to be determined case by case, it also provides a basic rule of thumb:

Use a MOSFET with a DC-current rating of 50% - 100% compared to the nominal current of the IGBT.

This rule of thumb was determined, assuming the use of 1200 V switches in a half-bridge converter. As boundary conditions, a bus voltage Vbus of 600 V and a duty cycle d=50% were selected. The bus voltage lies between the switched voltage of three-level and two-level circuits, and the duty cycle between inverter and rectifier operation. For the cooling path, through-hole packages, high-performance insulators, and a heat sink with forced-air cooling were assumed.

I know that my colleague from marketing does not want to hear it, but as long as we stay on such a general level, a clear-cut mapping of IGBT and MOSFET part numbers is not possible. Thus, in the training video, instead of listing specific replacement devices, a range of potential replacements had to be provided. In short, the general answer to such a general question is “it depends.” It depends on the individual case. You can probably imagine the frustration of my colleague hearing those two words from an engineer. But let’s forget about the day-to-day teasing between technical and commercial folks for now.

Simulate your individual case with IPOSIM

What about your individual case? I would encourage you to have a look at IPOSIM, the Infineon Online Power Simulation Tool. It is a PLECS-based simulation platform, which is available online, that allows you to determine the losses and temperatures of a power electronic system. A big advantage compared to offline simulations on your own computer is its convenience. You can select from different circuits, modulation schemes, components and calculation parameters with just a few clicks, in just a few seconds.

Here, so-called parameter-sweep simulations come in handy. They allow you, for instance, to determine the maximum output current and losses of a circuit for different switching frequency values and semiconductor devices. That is about everything we need to identify an equivalent replacement device. Just perform the calculations for all the replacement candidates and choose the one that is closest to the reference.

DIY with spreadsheets

Do-it-yourselfers might also use a spreadsheet, for instance to consider further aspects such as cost. It is not very convenient to take that route, initially. You need to derive analytic formulas for the loss and temperature calculations, implement them in your spreadsheet, and verify them in some way. However, once that is done, you have a powerful, flexible, scalable and extendable tool on hand which can also be used by less experienced engineers.

The analytic calculation of power losses and semiconductor temperatures for a given operation point is usually quite straightforward, and requires only basic arithmetical operations. But what if we need to determine the maximum output current of a circuit – same as before, but this time with a spreadsheet? Normally, solving equations analytically for the output current is impractical. An iterative approach is the preferred choice because it is easier to implement and less error-prone.

Figure 1 shows a very simple example implemented in Microsoft Excel. Starting with an arbitrarily chosen initial current value, the semiconductor losses and temperatures are calculated. Depending on how close the device temperature is to its limit, the current value is increased or decreased. Then, the losses and temperatures are calculated again. The procedure is repeated until the solution is found. Please note that for such calculations, you have to enable iterative calculations in the spreadsheet.

Summary

We have seen that there is no universal rule for translating between the nominal current of an IGBT and the on-state resistance of a MOSFET. Mapping always depends on the design goal and the individual boundary conditions. However, it is possible to find a mapping for specific cases or problems. Two approaches to determine such a mapping were discussed here: simulations using IPOSIM and calculations in spreadsheets. Feel free to contact us if you would like to see further information on any of those approaches.

Excel.png_1849974214.png

Figure 1

1067 Views
2 Comments
Authors