Feature: Power supplies
significantly smaller, lighter and more efficient power adapters and power supply units (PSUs) by allowing faster switching speeds, higher power density and lower energy losses, especially as heat. Te interest in how these devices perform has led to a
corresponding need to accurately simulate various GaN devices to optimise application performance. LTspice includes IC models of ADI’s latest DC-to-DC controllers that are optimised to drive GaN FETs. Tis enables design engineers to determine which GaN FET is best suited to their particular application and try various combinations for best performance. A common frustration users experience is choosing a FET when
product availability and selection from multiple vendors changes faster than the soſtware update in the native component library. Tis leaves the user responsible for library management of custom symbols and devices. It can be a time-consuming distraction from the goal of finding an ideal solution for a particular application. Additionally, collaboration can be hampered when everyone on a design team does not have the component libraries synchronised. Examples of such portable circuits with GaN FET devices are
Using GaN FET models in LTspice simulations
By James R. Staley, Senior Manager, Power Product Applications, Analog Devices
T
he interest in wide bandgap devices such as gallium nitride (GaN) and silicon carbide (SiC) FETs has significantly grown in the industrial power market in recent years. Because of their dramatically reduced charge characteristics, GaN devices offer high power densities at higher
switching frequencies that would otherwise create tremendous thermal losses in MOSFETs operating at the same conditions.
Modern power supplies GaN is widely used in modern power supplies, replacing traditional silicon-based components. Te technology enables
34 May 2026
www.electronicsworld.co.uk
available now on the product landing pages on
analog.com. As shown in Figure 1, an LTC7891 is configured for 12V, 240W operation using a pair of EPC2218 GaN FETs. Tis file can be downloaded, to run directly without any device library changes required. Te feature that makes it self-contained and portable is that the GaN FETs used in the simulation are referenced as a sub-circuit model. Te symbols used are standard NMOS type symbols, which any LTspice installation always has, and each of them has been configured to point to the .sub directive statement with the same model name. If the designer wishes to evaluate a different model, the process
is simple and leaves the LTspice file every bit as portable for evaluation by any other team member as when it was downloaded. If additional devices are placed, the process is the same.
Working with LT Spice To start, a model library file (typically provided by the vendor) is needed to extract the model data. For easy reference, this web location is included on all example circuits provided on the power product landing page. Figure 2 shows an example listing of device models from the vendor of the EPC2218A. To demonstrate the process, the EPC2218A device is selected as
the example. Te downloads from most vendors usually come with several files. Also included are the symbol file and example file. We are interested in the library file; see Figure 3. If this is opened directly, a default library installer opens, which
is not the preferred method. Te goal is not to have to add more component symbols and devices to a local library that must then be managed. Instead, we will use the data contained in the library files directly. Tis can be opened with any basic text editor tool (Notepad, etc.), getting the data without performing a library install.
Te library file has a long list of subcircuit text models, all of which begin with .subckt [model name] and end with .ends. Using
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44
