Interpolating Property Tables — Interpolate in property tables and charts - Thermodynamics

SECTION 1

Historical Context & Motivation

Long before digital equation-of-state solvers became commonplace, engineers relied on carefully measured and compiled property tables to look up the thermodynamic state of substances such as water, refrigerants, and combustion gases. These tables—cataloging specific volume, internal energy, enthalpy, and entropy at discrete pressures and temperatures—were the backbone of power-plant design, refrigeration engineering, and aerospace propulsion for over a century. Because experimental measurements could only be taken at finite intervals, the values a practicing engineer needed almost never coincided with a tabulated entry, and a reliable method for estimating intermediate values was essential.

1824 • Carnot's Foundation

Sadi Carnot published Réflexions sur la puissance motrice du feu, establishing the theoretical need for precise thermodynamic properties of working fluids in heat-engine analysis.

1915 • Mollier's Steam Tables

Richard Mollier published comprehensive steam tables and the h–s diagram, giving engineers standardized data sets that demanded interpolation for practical use between listed pressure and temperature increments.

1936 • Keenan & Keyes Tables

Joseph Keenan and Frederick Keyes published their landmark Thermodynamic Properties of Steam, which became the de facto reference in American engineering education and codified the interpolation practices still taught today.

1984 • NIST/JANAF Digital Databases

The National Institute of Standards and Technology began releasing digital thermodynamic databases, yet linear interpolation remained the standard pedagogical tool for quick calculations and exam settings.

Present • Modern Software with Table Literacy

Software such as EES, CoolProp, and REFPROP automates property lookups, but understanding interpolation remains essential for validating outputs, handling exam problems, and developing physical intuition about how properties vary.

The central question that interpolation answers is deceptively simple: if a table gives you the enthalpy of water at 200 °C and at 250 °C, what is the enthalpy at 227 °C? Answering this with confidence requires understanding when a linear assumption is acceptable, how to structure the arithmetic, and how to extend the technique to double interpolation when the desired state falls between tabulated values in two independent variables simultaneously.

SECTION 2

Core Principles & Definitions

Interpolation in thermodynamics rests on a small set of foundational ideas that connect the discrete nature of tabulated data to the continuous behavior of real substances. Before applying any formula, you must be comfortable with the structure of a property table—its independent variables (typically pressure and temperature for the superheated region, or pressure and quality for the two-phase region) and its dependent properties (v, u, h, s). The following principles frame every interpolation calculation you will encounter.

Linear Interpolation Assumption

Between two adjacent table entries, the property is assumed to vary linearly with the independent variable. This first-order approximation is accurate when table increments are small relative to the curvature of the actual property surface.

Interpolation Fraction

The dimensionless ratio (x − x₁)/(x₂ − x₁) measures how far the desired point lies between the two bounding entries. This interpolation fraction is applied identically to every dependent property at that pair of bounding states.

Bounding Values

You must identify the two table rows (or columns) that bracket the desired independent variable—one value just below and one just above. Using non-adjacent rows introduces unnecessary error.

Double Interpolation

When neither independent variable matches a table entry, two successive single interpolations are performed—first along one variable (e.g., temperature) at each bounding pressure, then along the second variable (pressure) between the two intermediate results.

Region Identification First

Before interpolating, determine whether the state is compressed liquid, saturated, or superheated. The correct table (saturation table vs. superheated table) must be selected; interpolating across phase boundaries yields nonsensical results.

✦

KEY TAKEAWAY

Think of interpolation the way a GPS estimates your position between two known mile markers. If marker A is at mile 10 and marker B is at mile 20, and your speedometer says you have traveled 54 % of the distance, you must be near mile 15.4. In a property table, the 'mile markers' are the tabulated temperatures or pressures, the 'distance traveled' is the interpolation fraction, and the 'position' is the property value you seek. As long as the road (property curve) is roughly straight between markers, the estimate is excellent.

SECTION 3

Visual Explanation of Linear Interpolation

The diagram below illustrates the geometric basis of linear interpolation. Two known data points from a property table are plotted on a coordinate system where the horizontal axis represents the independent variable (here, temperature) and the vertical axis represents the dependent property (here, specific enthalpy). The straight line connecting the two points defines the interpolated estimate at any intermediate temperature. The actual property curve, shown as a dashed arc, deviates slightly from the straight line; this deviation is the interpolation error, which shrinks as the table increment decreases.

The cyan line represents the linear interpolation between the two known points (T₁, h₁) and (T₂, h₂). The pink point marks the interpolated enthalpy at the desired temperature T. The small gap between the pink point and the dashed actual-property curve indicates the interpolation error, which is negligible for closely spaced table entries.

Notice that the interpolation fraction (T − T₁)/(T₂ − T₁), highlighted in amber on the diagram, acts as a weighting factor. When the desired temperature is close to T₁, the fraction approaches zero and the interpolated enthalpy approaches h₁; when it is close to T₂, the fraction approaches unity and h approaches h₂. This symmetric behavior ensures that the estimate smoothly transitions between the bounding values.

SECTION 4

Mathematical Framework

The mathematics of single and double linear interpolation are straightforward, but careful notation prevents sign errors and mix-ups between rows. We present the formulas in order of increasing complexity.

Single Linear Interpolation

SINGLE INTERPOLATION

y = y₁ + (y₂ − y₁) × [(x − x₁) / (x₂ − x₁)]

x = desired independent variable (e.g., T or P); x₁, x₂ = bounding table values (x₁ < x < x₂); y₁, y₂ = property values at x₁ and x₂ respectively; y = interpolated property.

An equivalent and sometimes more intuitive form introduces the interpolation fraction f explicitly.

INTERPOLATION FRACTION

f = (x − x₁) / (x₂ − x₁) → y = y₁ + f × (y₂ − y₁) = (1 − f) y₁ + f y₂

The weighted-average form on the right makes it clear that f = 0 yields y₁ and f = 1 yields y₂, confirming consistency at the endpoints.

Double Linear Interpolation

When the desired state lies between tabulated values of both independent variables—say temperature T and pressure P—double interpolation is required. The procedure involves three applications of the single-interpolation formula.

DOUBLE INTERPOLATION — STEP 1

y_A = y(P₁, T₁) + f_T × [y(P₁, T₂) − y(P₁, T₁)]

Interpolate along temperature at the lower bounding pressure P₁. f_T = (T − T₁)/(T₂ − T₁).

DOUBLE INTERPOLATION — STEP 2

y_B = y(P₂, T₁) + f_T × [y(P₂, T₂) − y(P₂, T₁)]

Repeat the temperature interpolation at the upper bounding pressure P₂ using the same f_T.

DOUBLE INTERPOLATION — STEP 3

y = y_A + f_P × (y_B − y_A)

Interpolate between the two intermediate results along pressure. f_P = (P − P₁)/(P₂ − P₁). The final value y is the doubly interpolated property.

💡

Order Independence

The result of double interpolation is independent of the order in which you perform the two single interpolations. You may interpolate first along pressure at each bounding temperature and then along temperature, or vice versa; the final answer is identical because bilinear interpolation is commutative.

SECTION 5

Detailed Breakdown — Double Interpolation on the Property Surface

Double interpolation can be visualized as locating a point on a two-dimensional property surface defined by two independent variables. The four nearest tabulated states form the corners of a rectangle in (P, T) space, and the desired state lies somewhere inside that rectangle. The following diagram shows this geometry.

The four corners of the rectangle are the tabulated property values at the bounding pressures and temperatures. The amber arrow shows the temperature interpolation fraction f_T, while the red arrow shows the pressure interpolation fraction f_P. The pink point is the desired doubly interpolated state.

In the diagram, Step 1 uses the temperature interpolation fraction f_T to find the property along the bottom edge (at P₁) and along the top edge (at P₂). Step 2 uses the pressure interpolation fraction f_P to interpolate between those two intermediate values, yielding the final estimate at the interior pink point. This bilinear interpolation assumes the property surface is planar over the small rectangle, which is an excellent approximation for typical table increments.

Step	Action	Formula
1a	Interpolate T at P₁	$y_A$ = y(P₁,T₁) + $f_T$ × [y(P₁,T₂) − y(P₁,T₁)]
1b	Interpolate T at P₂	$y_B$ = y(P₂,T₁) + $f_T$ × [y(P₂,T₂) − y(P₂,T₁)]
2	Interpolate P between results	y = $y_A$ + $f_P$ × ( $y_B$ − $y_A$ )

Summary of the three-formula procedure for double interpolation.

SECTION 6

Worked Example — Single Interpolation in Superheated Steam Tables

Suppose we need the specific enthalpy of superheated steam at P = 1.0 MPa and T = 427 °C. The superheated steam table at 1.0 MPa lists the following entries:

T (°C)	h (kJ/kg)
400	3264.5
500	3479.1

Excerpt from superheated steam table at P = 1.0 MPa (Çengel & Boles, Table A-6).

Find h at T = 427 °C and P = 1.0 MPa

1
Step 1 — Identify bounding values

The desired temperature 427 °C lies between the table entries T₁ = 400 °C and T₂ = 500 °C. The corresponding enthalpies are h₁ = 3264.5 kJ/kg and h₂ = 3479.1 kJ/kg.
2
Step 2 — Compute the interpolation fraction

f = (T − T₁) / (T₂ − T₁) = (427 − 400) / (500 − 400) = 27 / 100 = 0.27.

f = 0.27
3
Step 3 — Apply the interpolation formula

h = h₁ + f × (h₂ − h₁) = 3264.5 + 0.27 × (3479.1 − 3264.5) = 3264.5 + 0.27 × 214.6 = 3264.5 + 57.94.
4
Step 4 — State the result

The specific enthalpy at 1.0 MPa and 427 °C is approximately 3322.4 kJ/kg. This value falls 27 % of the way between h₁ and h₂, consistent with the temperature being 27 % of the way between 400 °C and 500 °C.

h ≈ 3322.4 kJ/kg

⚠

Sanity Check

Always verify that your interpolated value lies between h₁ and h₂. If it does not, you have made an arithmetic or sign error. Additionally, the interpolated value should be closer to the bounding value whose independent variable is nearer to the desired value.

SECTION 7

Strengths, Limitations, and Common Pitfalls

Strengths	Limitations
Simple arithmetic — no computer or calculator program needed beyond basic operations.	Accuracy degrades when the property curve is highly nonlinear between table entries (e.g., near the critical point).
Works identically for every property (v, u, h, s) once the fraction is computed.	Cannot cross phase boundaries — interpolating between a saturated and a superheated entry is physically meaningless.
Double interpolation extends the method naturally to two independent variables.	For substances with steep property gradients (e.g., near saturation), finer table increments are required for acceptable accuracy.
Provides physical intuition about how properties change with state variables.	Quadratic or spline interpolation may be warranted for highest precision, especially in research applications.

Comparison of strengths and limitations of linear interpolation in property tables.

Common Pitfalls

Wrong table region: Using the superheated table when the state is actually a compressed liquid (or vice versa) produces wildly incorrect values. Always compare the given temperature with the saturation temperature at the given pressure first.
Arithmetic sign errors: Forgetting which value is y₁ and which is y₂, or inverting the order in the denominator, flips the interpolation fraction and sends the answer in the wrong direction.
Extrapolation beyond table bounds: If the desired state falls outside the range of the table, the linear formula still produces a number, but it is an extrapolation, not an interpolation, and its accuracy is uncontrolled.
Unit inconsistency: Mixing kPa and MPa (or °C and K) for the independent variable while computing the interpolation fraction is a frequent source of error.

✦

KEY TAKEAWAY

Linear interpolation in property tables is the thermodynamic equivalent of 'connecting the dots.' It is reliable when the dots are close together and the underlying curve is gentle, but it should never be used to draw a line between dots that belong to different pictures—that is, across phase boundaries. Treat region identification as the prerequisite gatekeeping step before any interpolation.

SECTION 8

Connection to Advanced Interpolation Methods

Linear interpolation is the simplest member of a family of polynomial interpolation methods. In advanced courses and research settings, higher-order techniques provide greater accuracy at the cost of increased complexity. The table below compares the linear method you have learned with two more sophisticated alternatives.

Feature	Linear Interpolation	Quadratic / Lagrange	Cubic Spline
Data points required	2 (bounding pair)	3 (bounding + one neighbor)	All table entries in the range
Accuracy near critical point	Moderate to low	Improved	High
Smoothness of result	C⁰ (continuous but kinked)	C⁰ (piecewise)	C² (smooth second derivative)
Hand-calculation feasibility	Easily done by hand	Feasible with care	Requires software
Typical use case	Exams, quick engineering estimates	Intermediate calculations	EES, CoolProp, REFPROP backends

Comparison of interpolation methods used with thermodynamic property tables.

For the vast majority of undergraduate thermodynamics problems, linear interpolation provides sufficient accuracy because standard steam tables and refrigerant tables have small enough increments to keep the interpolation error below the precision of the tabulated data itself (typically four to five significant figures). However, if you proceed to courses in advanced thermodynamics or computational fluid dynamics, you will encounter cubic-spline-based property libraries where the underlying mathematics is automated but the conceptual foundations remain those you have learned here.

ℹ

Looking Ahead

Modern equation-of-state packages (e.g., IAPWS-IF97 for steam) replace tables entirely with piecewise analytical functions. Understanding linear interpolation provides the conceptual baseline for appreciating why these functions are constructed to be continuous and differentiable—properties that simple linear interpolation lacks across segment boundaries.

SECTION 9

Practice Problems

Work through the following five problems in order. They progress from conceptual understanding to critical analysis, mirroring the depth expected on a college-level thermodynamics exam. Use the standard property tables from your textbook (e.g., Çengel & Boles or Borgnakke & Sonntag) for tabulated data.

Question 1: CONCEPTUAL

Why is it essential to verify the thermodynamic region (compressed liquid, saturated mixture, or superheated vapor) before performing interpolation in a property table? What error would result if you mistakenly interpolated between a saturated-liquid entry and a superheated-vapor entry at the same pressure?

Show answer

Region identification is essential because different tables describe fundamentally different phase states, and the property relationships change discontinuously across phase boundaries. If you interpolated between a saturated-liquid enthalpy (

h_f

) and a superheated-vapor enthalpy at the same pressure, you would obtain a value that lies on a fictitious straight line passing through the two-phase dome—ignoring the latent heat of vaporization h_fg that physically separates the two states. The resulting number would not correspond to any real equilibrium state and could be off by hundreds of kJ/kg, leading to grossly incorrect energy and mass balances.

Question 2: BASIC CALCULATION

Using the superheated steam table at P = 0.4 MPa, the following data are available: at T = 200 °C, h = 2860.5 kJ/kg; at T = 250 °C, h = 2964.2 kJ/kg. Find the specific enthalpy at T = 218 °C and P = 0.4 MPa.

Show answer

First compute the interpolation fraction: f = (218 − 200) / (250 − 200) = 18/50 = 0.36. Then apply the formula: h = 2860.5 + 0.36 × (2964.2 − 2860.5) = 2860.5 + 0.36 × 103.7 = 2860.5 + 37.33 = 2897.8 kJ/kg. The result lies between the two bounding values and is closer to h at 200 °C, which is consistent with 218 °C being closer to 200 °C than to 250 °C.

Question 3: INTERMEDIATE

Consider the following superheated steam table excerpt. At P = 0.6 MPa: h(300 °C) = 3062.3 kJ/kg, h(350 °C) = 3166.8 kJ/kg. At P = 0.8 MPa: h(300 °C) = 3057.6 kJ/kg, h(350 °C) = 3162.2 kJ/kg. Determine the specific enthalpy at P = 0.7 MPa and T = 330 °C using double interpolation.

Show answer

Step 1: Compute interpolation fractions.

f_T

= (330 − 300)/(350 − 300) = 30/50 = 0.60.

f_P

= (0.7 − 0.6)/(0.8 − 0.6) = 0.1/0.2 = 0.50. Step 2: Interpolate along T at P₁ = 0.6 MPa:

h_A

= 3062.3 + 0.60 × (3166.8 − 3062.3) = 3062.3 + 0.60 × 104.5 = 3062.3 + 62.7 = 3125.0 kJ/kg. Step 3: Interpolate along T at P₂ = 0.8 MPa:

h_B

= 3057.6 + 0.60 × (3162.2 − 3057.6) = 3057.6 + 0.60 × 104.6 = 3057.6 + 62.76 = 3120.4 kJ/kg. Step 4: Interpolate along P: h = 3125.0 + 0.50 × (3120.4 − 3125.0) = 3125.0 + 0.50 × (−4.6) = 3125.0 − 2.3 = 3122.7 kJ/kg. The answer h ≈ 3122.7 kJ/kg lies within the rectangle defined by the four corner values, confirming consistency.

Question 4: APPLIED

A turbine operates with superheated steam entering at 1.4 MPa and 380 °C. Your steam table at 1.4 MPa provides: h(350 °C) = 3150.2 kJ/kg, s(350 °C) = 6.9553 kJ/(kg·K), h(400 °C) = 3258.7 kJ/kg, s(400 °C) = 7.1435 kJ/(kg·K). Determine both the specific enthalpy and specific entropy at the turbine inlet using interpolation.

Show answer

The interpolation fraction is the same for both properties because it depends only on the independent variable: f = (380 − 350)/(400 − 350) = 30/50 = 0.60. For enthalpy: h = 3150.2 + 0.60 × (3258.7 − 3150.2) = 3150.2 + 0.60 × 108.5 = 3150.2 + 65.1 = 3215.3 kJ/kg. For entropy: s = 6.9553 + 0.60 × (7.1435 − 6.9553) = 6.9553 + 0.60 × 0.1882 = 6.9553 + 0.1129 = 7.0682 kJ/(kg·K). These inlet properties would then be used in an isentropic or polytropic turbine analysis. Note that using the same interpolation fraction for all properties at a single state is a key efficiency technique.

Question 5: CRITICAL THINKING

Near the critical point of water (

P_c

≈ 22.064 MPa,

T_c

≈ 373.95 °C), property tables have much finer increments than in the low-pressure superheated region. Explain from a mathematical and physical perspective why linear interpolation can produce significantly larger errors near the critical point. Propose a strategy, still based on tabulated data, that could improve accuracy without resorting to software.

Show answer

Near the critical point, thermodynamic properties such as specific volume and specific heat exhibit extremely steep gradients and strong curvature—

c_p

diverges, and (∂v/∂T)_P becomes very large. The linear interpolation assumption treats the property as varying along a straight line between table entries, so when the actual curve has large second derivatives (d²y/dx²), the chord between adjacent points can deviate significantly from the true function. Physically, this region is characterized by density fluctuations and the convergence of liquid and vapor properties, producing a highly nonlinear property surface. One tabular strategy to improve accuracy is to use a three-point Lagrange (quadratic) interpolation: given three adjacent table entries (x₁, y₁), (x₂, y₂), (x₃, y₃), fit a second-degree polynomial that passes through all three and evaluate at the desired x. This captures the curvature that linear interpolation misses, reducing the error from O(Δx²) to O(Δx³), at the modest cost of a more involved hand calculation.

SUMMARY

Lesson Summary

Summary

Linear interpolation is the essential technique for extracting thermodynamic property values that fall between the discrete entries in property tables. The method hinges on computing a dimensionless interpolation fraction f = (x − x₁)/(x₂ − x₁) from the independent variable and applying it via the formula y = y₁ + f × (y₂ − y₁). When both independent variables (e.g., pressure and temperature) fall between tabulated values, double interpolation extends the procedure through three successive applications of the single-interpolation equation.

Before interpolating, always perform region identification to ensure you are using the correct table (saturation vs. superheated vs. compressed liquid). The method is accurate when table increments are small relative to the curvature of the property surface, and it serves as the conceptual foundation for the cubic-spline and equation-of-state methods used by modern property software. Mastery of this skill ensures that you can solve any table-based thermodynamics problem quickly and confidently, whether on an exam or in engineering practice.