Among the factors determining the maximum frequency of MCUs, the selection of the transistor threshold voltage—specifically the use of HVT, SVT, and LVT—is a core trade-off at the chip design level.

1. Basic Concept: What is Threshold Voltage?

Threshold voltage is the gate voltage required to switch a transistor (MOSFET) from the “off” state to the “on” state. It is a fundamental physical characteristic of the transistor.

• When V_gs < Vth: The transistor is off, and the current is minimal (leakage current).

• When V_gs > Vth: The transistor is on, and current can flow significantly.

2. Definitions of HVT, SVT, and LVT

In chip design libraries, engineers can choose transistors with different threshold voltages for different circuit modules:

1. HVT – High Threshold Voltage Transistor

   Characteristics: Requires a higher gate voltage to turn on.

   Advantages: Very low static leakage current. When the transistor is off, it consumes almost no power.

   Disadvantages: Slow switching speed. It takes more time to charge to reach a higher turn-on voltage, resulting in increased delay.

2. LVT – Low Threshold Voltage Transistor

   Characteristics: Requires only a very low gate voltage to turn on.

   Advantages: Extremely fast switching speed. The delay is minimal, significantly increasing the operating frequency of the circuit.

   Disadvantages: Very high static leakage current. Even when the transistor is off, there is considerable current leaking from the source to the drain, leading to a sharp increase in standby power consumption.

3. SVT – Standard Threshold Voltage Transistor

   Characteristics: Achieves a balance between HVT and LVT.

   Advantages: Provides a good compromise between speed and power consumption.

   Disadvantages: Not as fast as LVT, nor as power-efficient as HVT.

3. How HVT/LVT Affects the Maximum Frequency of MCUs

• The critical path is the slowest path in the internal logic chain of the chip, and its delay determines the maximum clock frequency at which the chip can operate stably.

• Path delay is primarily the sum of the logic gate delays (such as inverters, NAND gates) that make up that path.

• Logic gate delay directly depends on the switching speed of the transistors.

Therefore, choosing LVT transistors can significantly reduce the delay of logic gates, thereby shortening the total delay of the critical path, allowing for higher clock frequencies (i.e., increasing the maximum frequency).

Conversely, if a path is entirely composed of HVT transistors, its delay will be large, becoming a bottleneck that limits the increase in maximum frequency.

4. Trade-offs: Performance vs. Power Consumption and Reliability

Since LVT can increase speed, why not use LVT for the entire chip? This is the core art of chip design—trade-offs.

Characteristic	HVT (High Threshold)	SVT (Standard Threshold)	LVT (Low Threshold)
Speed	Slow	Medium	Fast
Static Power Consumption (Leakage)	Very Low	Medium	Very High
Impact on Maximum Frequency	Limits	Balances	Increases

The cost of using all LVT:

1. Power Consumption Disaster: Even when the MCU is in sleep mode and the clock is stopped, the huge leakage current can quickly deplete the battery. Static power consumption can be comparable to or even exceed dynamic power consumption.

2. Reliability Challenges: High leakage current can cause the chip to heat up, potentially leading to thermal runaway issues. Additionally, the reliability of the transistors may decrease.

5. Advanced Design Strategy: Multi-Threshold Voltage Technology

To address this contradiction, modern high-performance MCUs commonly adopt multi-threshold voltage technology. The design philosophy is to use LVT on critical paths for speed and HVT on non-critical paths to suppress leakage.

Specific Practices:

1. Identify Critical Paths: Use static timing analysis tools to find the few critical paths that limit frequency.

2. LVT Optimization: Replace standard transistors on these critical paths with LVT transistors to directly “speed up” and effectively increase the maximum frequency.

3. HVT as a Safety Net: Most non-critical logic on the chip (such as state machines, control logic, etc.) still uses HVT transistors. Since these paths have a large delay margin, using slower HVT will not affect the overall frequency but can significantly reduce the static power consumption of the entire chip.

For example: An MCU with a Cortex-M7 core, where the complex arithmetic logic unit and floating-point unit may be critical paths, will use a lot of LVT or even ULVT transistors. In contrast, peripheral controllers, GPIOs, and other parts that are not speed-sensitive will use HVT transistors.

• HVT/LVT are key design-level factors determining the maximum frequency of MCUs. LVT directly increases speed by reducing transistor delay.

• Do not blindly use LVT, as the enormous static power consumption it brings is unacceptable.

• Modern MCUs achieve fine optimization through multi-threshold voltage technology: Using LVT for the “fast lane” (critical paths) and HVT for the “regular lane” (non-critical paths) to control overall power consumption within a reasonable range while meeting target frequencies.

Therefore, when you see two MCUs using the same process and core but with different maximum frequencies, the differences in their internal HVT/LVT strategies are likely one of the reasons.

Previous Articles:

What is Nano Electrical Measurement?

Why is the sample size in HTOL testing 77 pieces? (Revised)

How to interpret the MTTF value after HTOL life testing?

Chip Decap Process and Steps

Are VSS and VSSA internally connected or separated in MCUs? Neither!

What is the typical yield of the MCUs you use?

Introduction to Wire Bonding in MCUs

Analysis of a single bit failure issue in MCU SRAMWhy do different MCUs have different system ESD levels in the same environment?Is your MCU using Spi Flash or eFlash, and how to choose?Dynamic Voltage Stress Test (DVST)Early Failure Screening of ChipsChip Failure Analysis: Principles, Functions, and Differences between EMMI and OBIRCHCan X-RAY cause data loss in MCU Flash?High-frequency signals are not necessarily high-speed signals, and high-speed signals are not necessarily high-frequency signals; please distinguish!Chip ESD HBM levels and their related failure ratesDetailed Explanation: How to Improve the Accuracy of ADCs in MCUsDifferences between System-Level ESD and Chip ESDAnalysis of Causes and Countermeasures for Chip Early FailuresWhen the MCU frequency soars, is a 0.1μF decoupling capacitor still sufficient?The Rise of Mid-Range MCUs: Domestic Chips Rising, the “Invisible Champion” in the Era of IoTESD Case 1: Electrostatic Discharge Issues Caused by Internal Wiring in an Industrial ProductAnalysis and Solutions for MCU POR and PDR Dead Zone IssuesAnalysis of MCU Power Architecture: Exploring Multi-Voltage Domain Design with Several Mainstream MCUs2024 Domestic MCU Chip Company Sales RankingMust-Read for Engineers! Basics of MCU Electrical Overstress (EOS)