A unified LTspice AC model for current-mode DC-DC converters
intro
When power supply designers want to get a general idea of the feedback loop of a power supply, they use loop gain and phase bod diagrams. Knowing that the loop response can be predicted helps narrow the selection of feedback loop compensation components. The precise way to generate the gain and phase diagram is to connect the power supply on the test bench and use a network analyzer; However, in the early stages of design, most designers choose to use computer simulations to quickly determine the approximate component selection range and to more intuitively understand the response of the loop to parameter changes.
This paper mainly studies the feedback control model suitable for current mode control power supply. Current mode control is common in switched mode DC-DC converters and controllers, and offers several advantages over voltage mode control: better line noise suppression, automatic overcurrent protection, easier parallel operation, and improved dynamic response.
Designers can already use a large number of current mode power averaging models. Some models are accurate to half the switching frequency and can match increasing converter bandwidths, but only for limited topologies such as buck, boost, and buck – boost topologies (non-4-switch buck – boost). Unfortunately, this applies to SEPIC and? The accuracy of the 3-port or 4-port average model in topologies such as uk is less than half of the switching frequency.
This article introduces LTspice® simulation models that are accurate to half the switching frequency (even relatively high frequency) and suitable for a variety of topologies, including:
decompression
boost
Buck – Boost
SEPIC
? uk
Forward type
flyback
This paper presents a simulation of piecewise linear system (SIMPLIS) results to determine the validity of the new model, and illustrates the specific application of the model. In some examples, test results are used to validate the model.
Current mode control models: a brief overview
In this section, we will reiterate some points about current mode control models. For a more complete understanding of current mode models, please refer to the publications mentioned in the Resources section at the end of this article.
The purpose of the current loop is to allow the inductive current to follow the path of the control signal. In the current loop, the average inductance current information is fed back to a modulator with a detection gain. The modulator gain Fm can be calculated geometrically, assuming that the constant inductor current slopes up and the external compensation current also slopes up.
In order to extend the validity of the average model shown in Figure 1 to the high frequency range, the researchers propose several improved average models based on the results of discrete time analysis and sample data analysis. In R. B. Ridley’s model, the sample-holding effect can be represented equivalent by the He(s) function, which can be inserted into the inductance current feedback path of the continuous average model. Because the model is evolved from the discrete-time model, it can accurately predict the subharmonic oscillation.