Response Spectrum Simulator Back
Earthquake Engineering Simulator

Response Spectrum Simulator — SDOF Seismic Design Response

Real-time IBC design response spectrum. Vary period $T$, damping $\zeta$, PGA and site factor to read $S_{DS}$, $S_{D1}$, transition period $T_S$ and current $S_a$.

Parameters
Period T
s
Damping ratio zeta
Peak ground acceleration PGA
g
Site coefficient F_a = F_v

IBC simplified form: $S_{DS} = 2.5 \cdot \mathrm{PGA} \cdot F_a$, $S_{D1} = \mathrm{PGA} \cdot F_v$, $T_S = S_{D1}/S_{DS}$, $T_0 = 0.2 T_S$. Damping correction $\eta = \sqrt{7/(2 + 100\zeta)}$.

Results
Spectral acceleration S_a
Short-period S_DS
1-second S_D1
Transition period T_S
Design response spectrum S_a(T)

x-axis = natural period $T$ in seconds / y-axis = spectral acceleration $S_a$ in g / blue line = design spectrum (linear ramp, plateau, $1/T$ descent) / yellow dot = current $S_a$ at period $T$ / orange dashed = transition period $T_S$ / green dashed = plateau cap $S_{DS}$.

Structure schematic (soil, building, ground motion)

Brown band = soil layer / gray = building (height scales with period $T$) / blue arrow = ground motion PGA / red arrow = roof spectral acceleration $S_a$ / arrow lengths are proportional to current values.

Theory & Key Formulas

IBC simplified design spectrum coefficients:

$$S_{DS} = 2.5 \cdot \mathrm{PGA} \cdot F_a,\qquad S_{D1} = \mathrm{PGA} \cdot F_v$$

Transition and corner periods:

$$T_S = \frac{S_{D1}}{S_{DS}},\qquad T_0 = 0.2\,T_S$$

Piecewise spectral acceleration $S_a(T)$:

$$S_a(T) = \begin{cases} \mathrm{PGA} + (S_{DS} - \mathrm{PGA})\,T/T_0 & (T < T_0) \\ S_{DS} & (T_0 \le T < T_S) \\ S_{D1}/T & (T \ge T_S) \end{cases}$$

Damping correction (eta = 1 at zeta = 0.05):

$$\eta = \sqrt{\frac{7}{2 + 100\,\zeta}}$$

$T$ is the natural period, $\zeta$ the damping ratio, $\mathrm{PGA}$ the peak ground acceleration, and $F_a$, $F_v$ are site amplification factors. $S_a$ is the SDOF maximum spectral acceleration and forms the basis of the design lateral force $V = (S_a/R)\,W$.

What is a Response Spectrum

🙋
A "response spectrum" comes up everywhere in seismic design. What is it, and how is it different from the earthquake record itself?
🎓
They are different. An earthquake record is a time history of ground shaking, while a response spectrum tells you "if this ground motion is fed into a single-degree-of-freedom oscillator, what is the maximum response for each period $T$ and damping $\zeta$." With one curve you can read the maximum response for any building once you know its period. At this tool's defaults (PGA=0.40 g, $F_a=F_v=1.5$, $T=0.50$ s, $\zeta=0.05$) the Results card shows $S_{DS}=1.50$ g, $T_S=0.40$ s and, since the current period is past $T_S$, $S_a = S_{D1}/T = 0.60/0.50 = 1.20$ g.
🙋
If I drag $T$ down to 0.20 s, $S_a$ pins to 1.50 g. Is this the "plateau" region you mentioned?
🎓
Exactly. Between $T_0 = 0.2 T_S = 0.08$ s and $T_S = 0.40$ s the curve sits flat at $S_a = S_{DS} = 1.50$ g, and short-period stiff structures (low-rise RC, masonry) typically fall in this range. Push $T$ further to 1.0 s or 2.0 s and $S_a$ drops as $S_{D1}/T$. Tall flexible buildings (high-rises, long-span bridges) trade smaller acceleration for larger displacement.
🙋
Increasing $\zeta$ from 0.05 to 0.20 lowers the entire spectrum. Is this the effect of base isolation?
🎓
Yes, that's the essence. The damping correction $\eta = \sqrt{7/(2 + 100\zeta)}$ equals 1 at $\zeta=0.05$ and drops to about 0.55 at $\zeta=0.20$, cutting the response by 45%. Base-isolated buildings combine lead dampers and rubber bearings to push $\zeta$ up to 15-30% and simultaneously stretch $T$, getting a double benefit. Sliding $\zeta$ and $T$ together in this tool shows the combined effect.
🙋
Pushing the site coefficient $F_a = F_v$ from 1.5 to 2.5 jumps $S_{DS}$ from 1.50 g to 2.50 g. Soft soils are scary!
🎓
They really are. A soft top 30 m can amplify rock motion by 2-3x at the surface. The 1985 Mexico City and 1989 Loma Prieta earthquakes both showed soft-soil resonance concentrating damage on isolated patches kilometers apart. Modern codes (IBC, Eurocode 8) start by classifying the soil profile because that determines $F_a$ and $F_v$. This tool lumps them into one slider; in real codes the short-period and long-period factors differ.

Frequently Asked Questions

It follows a simplified IBC (International Building Code) / ASCE 7 Type-I shape. Eurocode 8 uses a four-region spectrum with linear ramp for $T < T_B$, plateau between $T_B$ and $T_C$, $1/T$ descent between $T_C$ and $T_D$, and $1/T^2$ descent for $T > T_D$. For simplicity the tool truncates the long-period region to a $1/T$ tail. The Japanese Building Standard Law uses a similar shape but specifies the seismic coefficient $C_0$ and zone factor $Z$ separately. Use this tool for concept building; consult the appropriate code for design.
Physically the spectrum splits into equal-acceleration, equal-velocity and equal-displacement regions. At short periods the structure moves with the ground, so the maximum response acceleration approaches PGA and rises from PGA to $S_{DS}$. At intermediate periods the inertia of the structure resonates with typical earthquake periods and the response saturates at the plateau value $S_{DS}$. At long periods the structure is essentially stationary while the ground moves below it, so the maximum velocity and displacement become constant and the acceleration falls as $S_a \propto 1/T$. A further $1/T^2$ region exists at very long periods but is omitted here for simplicity.
For an SDOF-equivalent structure the design seismic coefficient is $C_s = S_a / R$ where $R$ is the response modification factor (3 to 8 depending on ductility), and the base shear is $V = C_s \cdot W$ where $W$ is the seismic weight. For multi-degree-of-freedom buildings each modal $S_a$ is read using its modal period and combined by SRSS or CQC. This tool only gives the SDOF $S_a$, so apply $R$, the importance factor and modal combination separately. As an example, with $S_a = 1.20$ g, $R = 6$ for a moment frame and $W = 10000$ kN we get $V = (1.20/6) \cdot 10000 = 2000$ kN.
The elastic response spectrum from a single record is jagged, with sharp peaks and valleys reflecting that earthquake's frequency content. A design spectrum is a smooth envelope obtained by statistical processing (mean plus one standard deviation, for instance) of many records, codified for design purposes. This tool computes the latter, representing the upper bound to consider in design, not the response to a specific record. To plot a true elastic response spectrum you must run a time-history analysis with Newmark-beta or a similar integration scheme.

Real-World Applications

Code-based building design: The Japanese Building Standard Law, US IBC/ASCE 7, European Eurocode 8 and Chinese GB 50011 all use response spectrum analysis as the core design method. The standard workflow is to estimate the natural period from $T = 0.1N$ (with $N$ the number of stories), read $S_a$ from the design spectrum, and compute the base shear $V = (S_a/R) W$. Sweeping $T$ from 0.1 s (single-story) to 3.0 s (30-story) in this tool quickly shows how response changes from low- to high-rise buildings.

Performance evaluation of base-isolated and damped buildings: Lead-rubber bearings, tuned mass dampers and viscous dampers reduce response by stretching $T$ and increasing $\zeta$. Setting $T = 3.0$ s and $\zeta = 0.20$ in the tool drops $S_a$ far below the default 1.20 g, illustrating how isolation can cut seismic response to one third or one fifth. Engineers verify this with full time-history analyses in real projects.

Nuclear and LNG facility seismic review: Seismic review of nuclear power plants and LNG storage tanks generates many elastic response spectra from site-specific design earthquakes and compares them to standard envelopes (US NRC RG 1.60, Japan's JEAG 4601). The $S_a$ at the equipment period drives the response analysis. After Fukushima the design basis earthquake levels were strengthened sharply (for example $S_s = 1209$ gal at Kashiwazaki-Kariwa), and spectrum-based seismic margin assessments became standard.

Bridge and transportation infrastructure: Japan's Road Bridge Specifications, AASHTO and Eurocode 8-2 all use spectrum methods for bridges. Pier periods range from 0.3 to 2 s, well within this tool's range. Long-period viaducts and cable bridges are sensitive to the long-period content of ground motion ($T > 3$ s), and near active faults additional directivity pulses must be considered separately.

Common Misconceptions and Pitfalls

The most common pitfall is to assume that "the $S_a$ from the spectrum is the actual peak acceleration of the building." Spectrum analysis gives the SDOF maximum response, but multi-story buildings require modal combination (SRSS or CQC) of many modes. Codes then apply response modification factor $R$, importance factor $I$, overstrength $\Omega_0$ and so on, so the final design force may be a fraction or a multiple of $S_a$. Treat this tool as an introduction to the concept and rely on the relevant code for actual design.

Next is the belief that "longer $T$ always means smaller response." Yes, $S_a$ falls as $1/T$ at long periods, but the spectral displacement $S_d = (T/2\pi)^2 S_a \propto T$ grows linearly with $T$. Base-isolated buildings cut acceleration to one third but in exchange suffer 30 to 60 cm of displacement, requiring careful design of expansion joints, piping and cable trays. This tool only displays $S_a$, but in the long-period region you must always remember the displacement trade-off.

The last pitfall is to think that "an elastic spectrum predicts the inelastic response." Spectrum methods assume elastic behavior; post-yield response is captured only indirectly through the response modification factor $R$. Short-period structures roughly obey the equal-energy rule (displacement of $\sqrt{2\mu - 1}$ times the elastic value, with $\mu$ the ductility ratio) while long-period structures obey the equal-displacement rule. To estimate cracking, hinge formation and strength degradation, run nonlinear time-history analysis (NLTHA) or nonlinear pushover analysis. Use this tool as the entry point to spectrum methods.