The Peak Time Formula and Symbol Definitions
In control theory and circuit analysis, the standard form of the peak time equation relies on the damped natural frequency. The primary formula is:
tp = π / ωd
Because the damped natural frequency (ωd) is derived from the system's undamped characteristics, the expanded formula is:
tp = π / (ωn √(1 - ζ2))
| Symbol | Parameter | Standard Unit | Definition & Constraints |
|---|---|---|---|
| tp | Peak Time | Seconds (s) | Time from the step input application to the first maximum peak of the response. |
| π | Pi | Dimensionless | Mathematical constant (≈ 3.14159). Must be in radians, not degrees. |
| ωd | Damped Natural Frequency | Radians per second (rad/s) | The actual frequency of oscillation during the transient decay. |
| ωn | Undamped Natural Frequency | Radians per second (rad/s) | The theoretical oscillation frequency if damping were zero (ζ = 0). |
| ζ | Damping Ratio | Dimensionless | Ratio of actual damping to critical damping. For peak time to exist, 0 < ζ < 1. |
Assumptions, Boundary Conditions, and Unit Traps
The peak time formula is not a universal law; it is a specific solution to a specific differential equation. Before plugging numbers into a calculator, verify your system meets the baseline assumptions.
When the Formula Applies
- System Order: Must be a linear time-invariant (LTI) second-order system (e.g., a series or parallel RLC circuit, or a mass-spring-damper mechanical system).
- Damping Region: The system must be underdamped (0 < ζ < 1). If ζ ≥ 1, the system does not overshoot, and peak time is mathematically undefined.
- Input Type: Derived assuming a unit step input with zero initial conditions (e.g., a sudden DC voltage applied to an uncharged RLC network).
Unit Mistakes That Break the Math
Additionally, ensure your calculator is set to radians when evaluating the π numerator. Using 180 (degrees) instead of 3.14159 (radians) will yield a peak time 57 times larger than reality.
Rearranged Forms: Solving for Every Variable
On the bench, you rarely know all variables upfront. Often, you measure the peak time on an oscilloscope and need to back-calculate the parasitic damping or the required inductance. Here are the algebraic rearrangements for every variable in the expanded formula:
- Solve for Damped Frequency (ωd):
ωd = π / tp - Solve for Undamped Frequency (ωn):
ωn = π / (tp √(1 - ζ2)) - Solve for Damping Ratio (ζ):
ζ = √( 1 - (π / (tp ωn))2 )
Reference: For a deeper mathematical derivation of these roots from the characteristic equation, see the Step Response analysis on Wikipedia or standard dynamics coursework like MIT OCW Engineering Dynamics.
Worked Examples with Unit Tracking
Let's apply the formula to two distinct scenarios: a high-speed power electronics snubber and an electromechanical servo loop. Notice how tracking units through the intermediate steps prevents magnitude errors.
Problem 1: SiC MOSFET Gate Driver RLC Snubber
Given: A series RLC snubber circuit designed to dampen ringing on a 2026-spec 1200V SiC MOSFET drain node. The components are $L = 10 \text{ mH}$, $C = 1 \text{ \mu F}$, and $R = 50 \text{ \Omega}$. Find the peak time ($t_p$) of the transient voltage ring.
- Calculate Undamped Natural Frequency (ωn):
ωn = 1 / √(L × C)
ωn = 1 / √(0.01 H × 0.000001 F) = 1 / √(10-8) = 10,000 rad/s - Calculate Damping Ratio (ζ):
ζ = (R / 2) × √(C / L)
ζ = (50 / 2) × √(0.000001 / 0.01) = 25 × √(10-4) = 25 × 0.01 = 0.25 (Underdamped, so peak time exists). - Calculate Damped Natural Frequency (ωd):
ωd = ωn × √(1 - ζ2)
ωd = 10,000 × √(1 - 0.0625) = 10,000 × √(0.9375) ≈ 9,682.46 rad/s - Calculate Peak Time (tp):
tp = π / ωd
tp = 3.14159 / 9682.46 ≈ 0.0003245 seconds = 324.5 μs
Problem 2: DC Servo Motor Position Control
Given: A closed-loop servo system has a characterized undamped natural frequency of $\omega_n = 15 \text{ rad/s}$ and a damping ratio of $\zeta = 0.4$. Find the peak time.
- Calculate Damped Natural Frequency (ωd):
ωd = 15 × √(1 - 0.42)
ωd = 15 × √(1 - 0.16) = 15 × √(0.84) ≈ 13.748 rad/s - Calculate Peak Time (tp):
tp = π / 13.748
tp = 3.14159 / 13.748 ≈ 0.2285 seconds (or 228.5 ms)
Realistic Magnitudes in Practical Circuits
What does a realistic answer look like? If your calculator spits out a number, you need a mental benchmark to know if you dropped a decimal point. The magnitude of $t_p$ is entirely dependent on the physical domain of the energy storage elements (inductors/capacitors vs. mass/springs).
| Application Domain | Typical Components / Mechanics | Expected tp Range |
|---|---|---|
| High-Frequency Snubbers / Gate Drivers | Parasitic PCB inductance (nH), ceramic caps (pF/nF) | 1 ns to 500 ns |
| Power Supply Output Filters (Buck/Boost) | Power inductors (μH), electrolytic/tantalum caps (μF) | 1 μs to 500 μs |
| Audio Crossover Networks | Audio inductors (mH), film capacitors (μF) | 10 μs to 5 ms |
| Electromechanical Servos & Robotics | Rotor inertia, shaft compliance, gearbox backlash | 10 ms to 2.0 s |
Frequently Asked Questions
How do I calculate peak time if the system is critically damped?
You don't. Peak time specifically measures the moment of maximum overshoot. By definition, critically damped (ζ = 1) and overdamped (ζ > 1) systems approach their steady-state value asymptotically without ever crossing it. Therefore, there is no overshoot, and $t_p$ is mathematically undefined (or conceptually infinite). For these systems, engineers use rise time and settling time instead.
What is the exact difference between peak time and rise time?
Rise time ($t_r$) is the duration it takes for the signal to transition from a lower threshold to an upper threshold—typically 10% to 90% of the final steady-state value for overdamped systems, or 0% to 100% for underdamped systems. Peak time ($t_p$) is the time elapsed from the step input until the signal hits its absolute first maximum peak, which is above the 100% steady-state line. Rise time tells you how fast the system reacts; peak time tells you when the maximum stress (voltage spike or mechanical torque) occurs.
Can I use the peak time formula for AC sinusoidal steady-state analysis?
No. The peak time formula is strictly a transient time-domain metric derived from the Laplace transform of a step response. If you are analyzing an RLC circuit driven by a continuous AC sine wave (like a 60Hz mains filter or an RF tank circuit), you must use phasor analysis, complex impedance ($Z = R + jX$), and calculate the steady-state amplitude and phase angle. Transient peak time does not apply to continuous-state AC resonance.
Why does my oscilloscope measurement of peak time not match the formula?
The formula assumes ideal, lumped components. In reality, parasitics dominate. If your math predicts $t_p = 150 \text{ ns}$ but your scope shows $210 \text{ ns}$, the discrepancy is likely due to Equivalent Series Inductance (ESL) in your capacitors, parasitic trace inductance on the PCB, and the input capacitance of your oscilloscope probe loading the node. To reconcile the math with the bench, you must extract the actual $\omega_d$ from the scope's ring frequency ($f_d = 1 / T_{period}$) and back-calculate the true parasitic ζ using the logarithmic decrement method. For a foundational review of how these physical components interact, refer to the RLC Circuit topology guide.






