9.1.1 Fourier Law and Thermal Conductivity

Conduction is energy transfer by molecular interaction under a temperature gradient, with no bulk motion. Fourier law states the quantitative relation. In one dimension:

q_x = minus k A dT by dx

Here q_x is heat transfer rate in watts, k is thermal conductivity in W per m K, A is area normal to flow, and dT by dx is the temperature gradient. The minus sign enforces flow from hot to cold. Thermal conductivity is a material property. Metals show high k because free electrons carry energy, with copper near 400 and carbon steel near 50 in W per m K. Nonmetals show low k, with furnace brick near 1 and mineral wool insulation near 0.05 to 0.15. Gases show very low k, near 0.026 for air at ambient. Conductivity varies weakly with temperature for solids and more strongly for gases, and the theory below treats k as constant over each layer, with mean-value correction applied in practice.

In three dimensions the vector form reads q vector equals minus k A grad T, and steady conduction with no generation obeys Laplace equation for constant k. Refinery relevance runs from tube-metal temperature drop in fired heaters to heat leak through vessel insulation and soil around buried hot lines.

9.1.2 Slab Resistance Derived by Direct Integration

Consider a plane slab of thickness L, face area A, conductivity k, faces held at T1 and T2 with T1 greater than T2, steady state, one-dimensional, no generation. Fourier law gives:

q = minus k A dT by dx

Separate variables and integrate from x equals 0 to L:

q integral dx from 0 to L equals minus k A integral dT from T1 to T2

Since q is constant at steady state, q L equals k A times T1 minus T2. Hence:

q equals T1 minus T2 divided by L divided by k A

The denominator is the conductive thermal resistance of the slab:

R_slab equals L divided by k A

The electrical analogy is deliberate. Temperature difference plays the role of voltage, heat rate plays the role of current, and resistance blocks flow. Large area or high k gives low resistance. Thick walls give high resistance. This form underlies every furnace wall, exchanger fouling layer, and coke deposit calculation.

9.1.3 Cylindrical Shell Resistance Derived

Consider a long hollow cylinder, inner radius r1, outer radius r2, length L, conductivity k, inner surface at T1 and outer at T2. Area normal to radial flow grows with radius, A(r) equals 2 pi r L. Fourier law in radial form:

q_r equals minus k times 2 pi r L times dT by dr

Separate variables:

q_r dr divided by r equals minus 2 pi k L dT

Integrate from r1 to r2 and T1 to T2 with constant q_r:

q_r ln(r2 divided by r1) equals 2 pi k L times T1 minus T2

Hence:

q_r equals T1 minus T2 divided by ln(r2 divided by r1) divided by 2 pi k L

So:

R_cyl equals ln(r2 divided by r1) divided by 2 pi k L

Resistance grows only logarithmically with radius ratio. Doubling insulation thickness on a large pipe therefore adds steadily less resistance per unit thickness. Small-bore steam tracing lines and large-bore transfer lines thus behave very differently under the same insulation thickness, a fact that leads directly to critical radius.

9.1.4 Spherical Shell Resistance Derived

Consider a hollow sphere, inner radius r1, outer radius r2, conductivity k. Area normal to flow is 4 pi r squared. Fourier law:

q_r equals minus k times 4 pi r squared times dT by dr

Separate:

q_r dr divided by r squared equals minus 4 pi k dT

Integrate from r1 to r2:

q_r times 1 divided by r1 minus 1 divided by r2 equals 4 pi k times T1 minus T2

Hence:

R_sph equals 1 divided by r1 minus 1 divided by r2 divided by 4 pi k

Spherical resistance saturates fastest of the three geometries. Beyond a modest outer radius, added thickness contributes little because area grows as radius squared. Spherical LPG bullets and Horton spheres therefore reach diminishing returns on insulation thickness earlier than cylindrical vessels of comparable size.

9.1.5 Composite Walls, Pipes, and Overall Coefficient

At steady state the same heat rate passes through each layer of a series composite, so temperature drops add and resistances add. For a three-layer slab:

R_total equals L1 divided by k1 A plus L2 divided by k2 A plus L3 divided by k3 A

For a composite cylinder with convection inside and outside, resistances in series give:

q equals Ti minus To divided by R_total

with

R_total equals 1 divided by hi Ai plus ln(r2 divided by r1) divided by 2 pi k1 L plus ln(r3 divided by r2) divided by 2 pi k2 L plus 1 divided by ho Ao

Parallel paths combine by reciprocal addition, exactly as in electrical circuits. Two parallel wall sections with resistances Ra and Rb give 1 divided by R_parallel equals 1 divided by Ra plus 1 divided by Rb.

Overall heat transfer coefficient U packages all resistances referred to a chosen area. Referred to outer area Ao:

1 divided by Uo Ao equals 1 divided by hi Ai plus sum of conduction resistances plus 1 divided by ho Ao

U referred to inner area differs by the area ratio. Consistent area basis is mandatory in exchanger specifications. Fouling, coke, and scale enter as additional series resistances that depress U over operating time.

9.1.6 Critical Radius of Insulation Derived

Adding insulation to a pipe adds conduction resistance but also enlarges the outer surface exposed to ambient convection. The two effects compete. Total resistance for a pipe of inner radius r1, insulation outer radius r, insulation conductivity k_ins, external convection coefficient h_o, per unit length:

R_total equals ln(r divided by r1) divided by 2 pi k_ins plus 1 divided by h_o times 2 pi r

Differentiate with respect to r:

dR_total by dr equals 1 divided by 2 pi k_ins r minus 1 divided by 2 pi h_o r squared

Set the derivative to zero:

1 divided by k_ins r equals 1 divided by h_o r squared

Hence the extremum sits at:

r_c equals k_ins divided by h_o

The second derivative is positive at this point, so total resistance is minimum and heat loss is maximum at r_c. For a sphere, outer convection resistance is 1 divided by h_o times 4 pi r squared, and the same differentiation gives:

r_c sphere equals 2 k_ins divided by h_o

Physical reading: if the bare pipe radius is smaller than r_c, the first layers of insulation increase heat loss because the new surface area effect dominates. Only after crossing r_c does added thickness reduce loss. If the bare radius already exceeds r_c, every added layer helps from the start.

9.1.7 The k-of-Insulation Trap and the Wire-Cooling Case

The trap lies in the numerator. Insulation conductivity k_ins fixes r_c together with h_o. A low-k covering gives a small r_c, so normal pipes exceed it and insulation behaves as expected. A modest-k covering in still air with low h_o gives a large r_c that can exceed the wire radius.

Canonical case: a thin hot copper wire of radius 3 mm in still air with h_o near 10 W per square m K, wrapped with a covering of k_ins near 0.15 W per m K. Then r_c equals 0.15 divided by 10 equals 0.015 m, that is 15 mm. The bare 3 mm radius lies far below r_c. Wrapping the wire therefore raises heat loss steadily until the outer radius reaches 15 mm, and only thicker covering beyond 15 mm starts to insulate. A proposed 12 mm wrap thickness, giving outer radius 15 mm, sits exactly at maximum loss and cools the wire fastest rather than protecting it.

Electrical cables rely on this effect deliberately, using the covering to reject heat. Steam tracing and small-bore hot instrument lines suffer from it accidentally when oversized low-grade insulation is applied in still air. Large furnace ducts, crude lines, and exchangers operate far above r_c, where the area effect is negligible and insulation always reduces loss. Design rule: compute k_ins divided by h_o before specifying insulation on any line below roughly 25 mm radius.

9.2.1 Governing Equation and the Hyperbolic Solution Form

A fin conducts heat along its length while losing heat laterally by convection. Consider a constant cross-section fin, cross-sectional area Ac, perimeter P, conductivity k, convection coefficient h, ambient T_inf, base temperature Tb. Energy balance on a slice dx gives:

d squared theta by dx squared minus m squared theta equals 0

where theta(x) equals T(x) minus T_inf and:

m equals square root of h P divided by k Ac

General solution is a sum of exponentials, equivalently expressed in hyperbolic functions. With an adiabatic tip boundary condition d theta by dx equals 0 at x equals L, the temperature distribution is:

theta divided by theta_b equals cosh of m times L minus x divided by cosh of m L

Heat rate at the base follows from Fourier law at x equals 0:

q_fin equals square root of h P k Ac times Tb minus T_inf times tanh of m L

The tanh factor carries the saturation physics. For mL beyond about 2.5, tanh mL exceeds 0.986, so added length adds almost no heat transfer. Extra metal becomes dead weight exposed to fouling and corrosion. Refinery fin-fan coolers are sized with this saturation limit in view.

9.2.2 Effectiveness Versus Efficiency

Fin effectiveness and fin efficiency measure two distinct ideas and are never interchangeable.

Effectiveness epsilon_fin equals q_fin divided by q_without_fin, where q_without_fin equals h times Ab times Tb minus T_inf and Ab is the base area the fin occupies. Effectiveness settles the investment issue, whether adding the fin was worthwhile. Values well above unity, often above 5 for gas-side fins, justify the metal. Short fat fins of low conductivity material, or fins in high-h boiling service, can give effectiveness near unity and waste material.

Efficiency eta_fin equals q_fin divided by q_ideal, where q_ideal equals h times A_fin times Tb minus T_inf is the heat transfer if the entire fin surface sat at base temperature. Efficiency settles the quality issue, how close the fin comes to isothermal perfection. Long thin fins show low efficiency because the tip runs cool. Short thick high-k fins show high efficiency but may still show poor effectiveness if area gain is small.

Design tension: raising area through longer or denser fins lowers efficiency while initially raising effectiveness, then effectiveness saturates through tanh while efficiency keeps falling. Fouled air coolers above refinery pipe racks lose effectiveness first, as dust and oily deposits blanket the added area that justified the fins.

9.2.3 Fin Selection Logic

High fin payoff needs low h, high k, and thin closely spaced geometry on the limiting side. Air-cooled condensers fin the air side only. Steam condensers fin the cooling-water side lightly or not at all. Aluminium or copper fins on steel tubes balance conductivity against strength. In fouling service, wider fin spacing preserves effectiveness longer than dense packing with higher clean effectiveness.

9.3.1 Newton Law and the Meaning of h

Convection couples conduction in a thin fluid layer at the wall to bulk advection that sweeps energy away. Newton law of cooling packages the entire mechanism into:

q equals h A times Ts minus T_inf

The coefficient h is not a fluid property. It depends on geometry, velocity, fluid properties, and thermal boundary condition. Typical magnitudes anchor intuition: natural convection in air 5 to 25, forced air 25 to 250, forced water 250 to 15000, boiling and condensation 2500 to 100000, all in W per square m K. The remainder of convection theory is the systematic recovery of h from dimensionless groups.

9.3.2 The Five Dimensionless Groups

Reynolds number Re equals rho V D divided by mu, equivalently V D divided by nu. It measures inertia relative to viscous force. Low Re means laminar ordered flow, high Re means turbulent mixing. Pipe flow transitions near Re 2300 to 10000. All forced-convection correlations use Re as the flow descriptor.

Prandtl number Pr equals mu cp divided by k, equivalently momentum diffusivity divided by thermal diffusivity. It is a fluid property alone. Gases show Pr near 0.7, water near 1 to 7 depending on temperature, oils show Pr in the hundreds to thousands. High Pr means the thermal boundary layer is much thinner than the velocity boundary layer, so convection is efficient once turbulence mixes the near-wall fluid.

Nusselt number Nu equals h D divided by k. It is the dimensionless heat transfer coefficient, the unknown being solved for. Nu of order unity means conduction-dominated transfer across a stagnant layer. Nu of hundreds means vigorous turbulent enhancement.

Grashof number Gr equals g beta delta-T L cubed divided by nu squared. It measures buoyancy relative to viscous restraint and drives natural convection. Beta is the volume expansion coefficient, 1 divided by T absolute for ideal gases. Large temperature difference or large size drives strong natural circulation.

Rayleigh number Ra equals Gr times Pr. It is the control knob for natural convection regime. Below about 1e9 on vertical plates the boundary layer stays laminar, above it transition to turbulence raises h sharply. Tank heating coils in still air, hot vessels in pipe trenches, and cooling of idle furnaces all sit in Grashof and Rayleigh territory.

9.3.3 Dittus-Boelter Correlation and the Heating Exponent

For fully developed turbulent flow inside tubes, the workhorse correlation is Dittus-Boelter:

Nu equals 0.023 times Re to the 0.8 times Pr to the n

Applicable roughly for Re above 10000, Pr between 0.7 and 160, L divided by D above 10, with properties evaluated at bulk mean temperature. The exponent n encodes the thermal boundary condition:

n equals 0.4 when the fluid is being heated, wall hotter than fluid

n equals 0.3 when the fluid is being cooled, wall cooler than fluid

Reasoning: heating a fluid lowers its near-wall viscosity for liquids and reshapes the temperature profile so the thermal resistance of the viscous sublayer falls, giving stronger heat transfer and the larger exponent. Cooling reverses the effect, thickening the resistant sublayer. Memory anchor: hot wall gets the hotter exponent 0.4. Crude preheat trains and lube-oil coolers operate in Dittus-Boelter range on the tube side, with shell-side correlations playing the analogous role outside the tubes.

9.4.1 Stefan-Boltzmann Law and the Kelvin Mandate

Every surface above absolute zero emits thermal radiation across a spectrum set by its temperature. A black surface emits the maximum possible, given by Stefan-Boltzmann law:

Eb equals sigma T to the fourth

with sigma equals 5.67 times 10 to the minus 8 W per square m K4. A real gray surface emits epsilon times Eb, where emissivity epsilon lies between 0 and 1. Temperature must be absolute in kelvin. Because of the fourth power, a modest Celsius-to-kelvin slip produces a large error. Polished aluminium shows epsilon near 0.05, oxidized steel near 0.8, furnace brick near 0.9. Refinery furnace fireboxes transfer the dominant fraction of heat by flame and hot-brick radiation, which is why tube-metal temperature is controlled by radiation even when convection is present.

9.4.2 Wien Displacement and Kirchhoff Reasoning

Wien displacement law locates the spectral peak:

lambda_max T equals 2898 micrometre K

Hotter bodies peak at shorter wavelengths. Flames and furnace interiors peak in the near infrared, while ambient bodies peak in the far infrared. This underlies selective coatings and infrared thermography interpretation.

Kirchhoff law states that for a diffuse surface at the same temperature and wavelength, absorptivity equals emissivity, alpha equals epsilon. Reasoning: place the surface in an isothermal enclosure at its own temperature. Steady state demands emitted plus reflected radiation balance incoming radiation, and the only consistent assignment for an opaque diffuse surface is alpha equals epsilon. Consequence: good emitters are good absorbers, and polished shields that emit poorly also absorb poorly, which is exactly why they serve as radiation barriers. The equality strictly applies per wavelength and direction, with the gray diffuse approximation extending it to total hemispherical values in engineering work.

9.4.3 Enclosure Exchange and the Small-in-Large Result

Radiative exchange between gray surfaces depends on emissivities and view factors. The general two-surface enclosure formula contains surface resistances 1 minus epsilon divided by A epsilon at each surface plus a space resistance 1 divided by A F. The gift simplification arises when a small convex object of area A1 and emissivity epsilon1 sits in a large isothermal room at T2. The view factor from object to room is unity, the room surface resistance vanishes because A2 is huge, and exchange reduces to:

q equals epsilon1 A1 sigma times T1 to the fourth minus T2 to the fourth

No view factor algebra is needed. Hot sample tubes, thermocouple beads, and small hot flanges radiating to pipe-rack surroundings follow this form directly.

9.4.4 Radiation Shields and the 1-over-n-plus-1 Proof

A radiation shield is a thin low-emissivity sheet inserted between two radiating surfaces. Consider two large parallel plates at fixed T1 and T2 with equal emissivity epsilon. Without shields, exchange per unit area is:

q0 equals sigma times T1 to the fourth minus T2 to the fourth divided by 2 divided by epsilon minus 1

Each gray surface contributes a surface resistance 1 minus epsilon divided by epsilon, and the space resistance between parallel plates is unity. Insert one thin shield of the same emissivity with both faces active. The heat stream now crosses two gaps in series, each gap carrying the same q and each gap presenting the same combined resistance as the original single gap. Total resistance doubles, so heat flow halves. With n identical shields there are n plus 1 gaps in series, each of equal resistance, so:

q_n equals q0 divided by n plus 1

One shield halves the load, two shields cut it to one third, three to one quarter. Proof rests only on series addition of equal gap resistances under fixed end temperatures. Multilayer vessel insulation and floating shields in furnace peepholes exploit this division directly. Low emissivity strengthens each gap resistance further, compounding the benefit.

Quick Example — One Shield Halves the Radiative Load

Given parallel plates at $T_1 = 600$ K and $T_2 = 300$ K with $\epsilon = 0.8$ and $\sigma = 5.67 \times 10^{-8}$.

Step 1: evaluate $\sigma(T_1^4 - T_2^4) \approx 6889$ W per square m.

Step 2: divide by $(2/\epsilon - 1) = 1.5$ to get $q_0 \approx 4593$ W per square m.

Step 3: apply $q_1 = q_0/2 \approx 2296$ W per square m.

Result: a single identical shield cuts the load from about $4593$ to about $2296$ W per square m.

Trap: end temperatures must stay fixed for the division to hold, and all temperatures must be in kelvin.

9.4.5 Reradiating Surfaces

A reradiating surface is insulated on the back, so net heat flow is zero. It floats to a temperature where absorbed irradiation exactly equals its own emission. Large refractory walls that see flame on one face and tubes elsewhere behave approximately this way over short intervals, redirecting radiation without net gain. The reradiating temperature always lies between the hot and cold surface temperatures, weighted by view factors and emissivities. Recognizing such surfaces simplifies furnace and duct enclosure networks by eliminating one unknown heat rate in exchange for a floating temperature node.

9.5.1 LMTD Derived from End-to-End Balances

Consider a double-pipe exchanger with hot capacity rate Ch equals m_dot_h cp_h and cold capacity rate Cc equals m_dot_c cp_c. Over a differential area dA with local temperature difference delta-T equals Th minus Tc and overall coefficient U:

dQ equals U delta-T dA

Energy balances on the two streams give dTh equals minus dQ divided by Ch and dTc equals plus or minus dQ divided by Cc, with the sign depending on flow direction. For either arrangement, the change in local difference obeys:

d(delta-T) divided by delta-T equals minus U times 1 divided by Ch plus-or-minus 1 divided by Cc times dA

Integrating from end 1 to end 2 over total area A gives:

ln(delta-T2 divided by delta-T1) equals minus U A times 1 divided by Ch plus-or-minus 1 divided by Cc

Eliminating the capacity combination using total heat Q equals Ch times Th_in minus Th_out equals Cc times Tc_out minus Tc_in yields the compact result:

Q equals U A times delta-T1 minus delta-T2 divided by ln(delta-T1 divided by delta-T2)

The second factor is the log mean temperature difference, delta-T_lm. Each delta-T is evaluated at one physical end of the exchanger. When the two end differences are nearly equal, delta-T_lm approaches their arithmetic mean. When one end pinches, the log mean falls below the arithmetic mean and correctly penalizes the pinched design.

9.5.2 Counter Versus Parallel Proved on Temperature Profiles

In parallel flow both streams enter at the same end, so the hot stream cools while the cold stream warms in the same direction, and the local difference decays monotonically toward the exit. The cold outlet can never exceed the hot outlet. In counter flow the streams enter at opposite ends, so a hot inlet meets already-warmed cold fluid and a cold inlet meets already-cooled hot fluid. The local difference stays more uniform along the length, and the cold outlet can approach the hot inlet temperature.

For identical inlet temperatures, identical U, and identical area, the counter-flow exchanger sustains a larger delta-T_lm and therefore transfers more heat. Equivalently, a given duty needs less area in counter flow. Proof follows from the derivation above: the counter-flow capacity combination keeps delta-T1 and delta-T2 both finite and balanced, while parallel flow forces one end difference to collapse. Multi-pass shell-and-tube and cross-flow geometries fall between the two ideals and are handled by a correction factor F multiplying the counter-flow LMTD, with F below unity. Refinery crude-versus-product preheat trains use counter-flow series banks precisely to squeeze maximum heat from product rundown into cold crude before furnace firing.

9.5.3 NTU, Effectiveness, and Fouling

The LMTD method needs outlet temperatures to evaluate delta-T values, forcing iteration when outlets are unknown. The NTU method removes the iteration. Define Cmin as the smaller capacity rate, Cmax as the larger, Cr equals Cmin divided by Cmax, NTU equals U A divided by Cmin, and effectiveness epsilon equals actual Q divided by maximum possible Qmax equals Cmin times Th_in minus Tc_in. Effectiveness for each arrangement is a closed-form function of NTU and Cr, rising steeply at first and flattening toward an asymptote set by Cr. Large NTU buys little extra effectiveness once the asymptote is approached, guiding economic sizing.

Fouling adds a series resistance Rf, usually tabulated as a fouling factor, so that:

1 divided by U_fouled equals 1 divided by U_clean plus Rf

Crude-side asphaltene deposition, cooling-water scaling, and coke films all act through this single added term. A fouled crude exchanger forces the downstream furnace to fire harder for the same coil inlet temperature, which is why plant performance stories trace fuel overconsumption back to a hidden fouling resistor rather than to burner fault.

9.6.1 The Nukiyama Curve Walked Phase by Phase

The boiling curve plots wall heat flux against wall superheat, defined as wall temperature minus saturation temperature at the prevailing pressure. Traversed from low to high superheat at controlled heat flux:

Natural-convection boiling occupies the first segment. No bubbles form. Liquid circulates by buoyancy and heat flux rises gently with superheat.

Nucleate boiling follows once cavities on the wall activate. Bubbles nucleate, grow, and detach, stirring the microlayer violently. Heat flux climbs steeply for small additional superheat. This is the efficient operating regime for reboilers, kettles, and steam generators.

Critical heat flux marks the peak. Vapour generation becomes so intense that liquid can no longer rewet the wall continuously. The peak value is called CHF, also departure from nucleate boiling. Any further demand for heat flux cannot be sustained by liquid contact.

Transition boiling lies past the peak. Patches of vapour blanket alternate with wetted spots. Raising wall superheat now lowers heat flux because the insulating blanket spreads faster than temperature driving force rises. The curve slopes downward. Operation here is unstable under heat-flux control.

Film boiling closes the curve. A stable continuous vapour film covers the wall. Heat crosses the film by conduction and radiation, and flux rises again slowly with superheat. Wall temperatures run far above saturation.

Under heat-flux control, crossing CHF jumps the operating point horizontally to the film-boiling branch at drastically higher wall temperature, destroying tubes or heater elements. This burnout mechanism is the central safety constraint on boiler and reboiler heat flux. Under temperature control the transition branch can be traced, but process heaters are rarely temperature-controlled at the metal.

9.6.2 CHF Significance and Post-CHF Behaviour

CHF depends on pressure, subcooling, flow velocity, and surface condition. It peaks near intermediate pressures and falls as the critical pressure is approached because latent heat vanishes. Flow and subcooling raise CHF by sweeping vapour away and condensing blankets. Burnout margin, defined as CHF divided by operating flux, is a primary thermal design parameter for steam generators and column reboilers. Post-CHF wall temperatures can exceed metallurgical limits within seconds in high-flux furnaces, so alarms and circulation safeguards guard the margin.

9.6.3 Dropwise Versus Filmwise Condensation

Condensation heat transfer is governed by the liquid film that forms on the cold wall. In filmwise condensation the condensate wets the wall as a continuous film and heat must conduct through that film. Since liquid conductivity is modest, the film is the dominant resistance. In dropwise condensation the condensate beads into discrete droplets that roll off, leaving large areas of nearly bare metal exposed to vapour. Measured coefficients for dropwise condensation exceed filmwise values by an order of magnitude or more.

Promoters that prevent wetting sustain dropwise mode, but contamination and oxidation usually revert the surface to filmwise over time. Noncondensable gases degrade both modes severely by accumulating at the vapour-liquid interface and blanketing diffusion of vapour to the wall. Condenser venting and air-pump duty therefore protect heat transfer directly. Refinery overhead condensers and steam surface condensers live or die on film management and gas removal rather than on raw area.

9.7.1 Zeroth, First, and Second Laws Stated

Zeroth law: if two bodies are each in thermal equilibrium with a third body, they are in thermal equilibrium with each other. Temperature is the common property that marks equilibrium, and thermometry is thereby legitimized.

First law: energy is conserved. For a closed system undergoing a process, heat added to the system minus work done by the system equals the change in internal energy. In differential form dU equals delta-Q minus delta-W with the work-done-by convention. Over a cycle the net heat equals the net work because internal energy returns to its start value.

Second law, Kelvin-Planck statement: no cyclic device can convert heat from a single reservoir entirely into work. Some heat must be rejected. Perpetual motion of the second kind is impossible.

Second law, Clausius statement: no cyclic device can transfer heat from a cold reservoir to a hot reservoir without work input. Refrigeration demands work. The two statements are logically equivalent, each implying the other, and both encode the directionality of natural processes. Carnot efficiency is their quantitative child.

9.7.2 Steady Flow Energy Equation Term by Term

For a control volume at steady state with one inlet and one exit, mass and energy conservation give:

m_dot times h1 plus V1 squared divided by 2 plus g z1 plus Q_dot equals m_dot times h2 plus V2 squared divided by 2 plus g z2 plus W_dot_cv

Each term carries distinct physics. Enthalpy h equals u plus p v combines internal energy with flow work needed to push fluid across the boundary. Kinetic term V squared divided by 2 dominates nozzles and diffusers and is negligible in large ducts and vessels. Potential term g z is negligible in gas paths and short vertical runs but retained for tall columns and penstocks. Q_dot is heat added to the volume. W_dot_cv is shaft work done by the volume.

Device reductions follow by crossing out dead terms. Throttling valves and capillary tubes: no shaft work, no appreciable heat, negligible kinetic change, so h1 equals h2, throttling is isenthalpic. Nozzles: no shaft work, near-adiabatic, potential change negligible, so enthalpy drop converts to kinetic gain, h1 plus V1 squared divided by 2 equals h2 plus V2 squared divided by 2. Compressors and pumps: work input raises enthalpy, heat loss usually neglected in first estimates. Turbines: enthalpy drop delivers shaft work. Heat exchangers: no shaft work, kinetic and potential changes negligible, so heat lost by one stream equals heat gained by the other through enthalpy changes.

9.7.3 Entropy, Irreversibility, and the TdS Equations

Entropy S measures irreversibility and the degradation of energy quality. Clausius inequality states that cyclic integral of delta-Q divided by T is less than or equal to zero, with equality only for reversible cycles. For a process, dS is greater than or equal to delta-Q divided by T, equality only when internally reversible. Isolated-system entropy never decreases. Every friction, throttling, mixing, unrestrained expansion, and finite-temperature-difference heat transfer generates entropy.

Combining the first and second laws for a simple compressible substance gives the TdS equations per unit mass:

T ds equals du plus p dv

T ds equals dh minus v dp

These relations connect entropy change to measurable property changes along any path, reversible or not, because they relate state properties only. For ideal gases with constant specific heats, integration gives ds equals cv ln(T2 divided by T1) plus R ln(v2 divided by v1) and equivalently ds equals cp ln(T2 divided by T1) minus R ln(p2 divided by p1). Constant-volume heating therefore gives ds equals cv ln(T2 divided by T1), constant-pressure heating gives ds equals cp ln(T2 divided by T1). On T-s diagrams, reversible heat equals the area under the path, so steeper constant-volume lines and flatter constant-pressure lines can be read directly.

9.7.4 Polytropic Work Derived

A polytropic path obeys p V to the n equals constant, with index n selecting the family: n equals 0 isobaric, n equals 1 isothermal, n equals k isentropic for ideal gases, n approaching infinity isochoric. Boundary work for a closed-system displacement is integral of p dV from state 1 to 2.

For n not equal to 1, p equals C divided by V to the n, so:

W equals C integral V to the minus n dV equals p1 V1 minus p2 V2 divided by n minus 1

For n equals 1, p V equals constant C equals p1 V1, so:

W equals C ln(V2 divided by V1) equals p1 V1 ln(V2 divided by V1)

Sign follows the work-done-by convention: expansion work is positive. Steady-flow shaft work for the same polytropic compression carries an extra factor n, discussed under compressors. The polytropic family unifies textbook processes into one formula with one exceptional logarithmic case.

9.7.5 Maxwell Relations and Clapeyron Uses

Maxwell relations are exact cross-derivatives from the Helmholtz and Gibbs potentials, converting unmeasurable entropy derivatives into measurable p-v-T slopes. The four central forms are: partial T by partial v at constant s equals minus partial p by partial s at constant v, partial T by partial p at constant s equals partial v by partial s at constant p, partial p by partial T at constant v equals partial s by partial v at constant T, and minus partial v by partial T at constant p equals partial s by partial p at constant T. They prove that cp minus cv, Joule-Thomson coefficient, and entropy changes can all be extracted from equation-of-state data.

Clapeyron equation governs phase boundaries:

dp by dT_sat equals h_fg divided by T_sat times v_fg

It gives the slope of the saturation dome on p-T diagrams from latent heat and volume change. Integrated with ideal-gas approximation for vapour volume, it yields Clausius-Clapeyron form used for vapour pressure extrapolation. Practical uses include saturation temperature shift with pressure in vacuum condensers, flash calculations for condensate letdown, and consistency checks on steam tables.

9.8.1 Ideal Gas Relations and cp-minus-cv Derived in Idea

Ideal gas equation per unit mass is p v equals R T, with R the specific gas constant, or p V equals m R T for mass m. Internal energy of an ideal gas depends on temperature alone, u equals u(T), a consequence of absent intermolecular forces. Define cv equals du by dT at constant volume and cp equals dh by dT at constant pressure with h equals u plus p v equals u plus R T. Differentiating h with respect to T gives:

cp equals du by dT plus R equals cv plus R

Hence cp minus cv equals R. With ratio k equals cp divided by cv, the pair inverts to cv equals R divided by k minus 1 and cp equals k R divided by k minus 1. Air takes k near 1.4, superheated steam near 1.3, monatomic gases near 1.66. The derivation idea to retain: enthalpy of an ideal gas exceeds internal energy by exactly R T per unit mass because flow work p v equals R T, and differentiating that gap gives the specific-heat gap.

9.8.2 Pure Substance Diagrams and Dryness Fraction

A pure substance passes through distinct states as heat is added at constant pressure: compressed liquid, saturated liquid on the bubble line, liquid-vapour mixture under the dome, saturated vapour on the dew line, then superheated vapour. The dome closes at the critical point. On p-v diagrams work is the area under the path. On T-s diagrams reversible heat is the area under the path. Reading cycles on both diagrams simultaneously reveals work and heat at a glance.

Dryness fraction x equals mass of vapour divided by total mass, defined only inside the dome between 0 and 1. Mixture properties weight saturated values:

v equals vf plus x v_fg

u equals uf plus x u_fg

h equals hf plus x h_fg

s equals sf plus x s_fg

Outside the dome x does not exist, and applying mixture formulas to superheated steam is a categorical error. Superheated and compressed-liquid states use direct table or chart values.

9.8.3 Triple and Critical Anchors for Water

Water triple point sits at 0.01 C and 0.611 kPa, where ice, liquid, and vapour coexist. It anchors fixed-point thermometry and the kelvin scale definition lineage. Water critical point sits near 374 C and 22.09 MPa. Beyond it, liquid and vapour merge into a supercritical fluid with no phase boundary and no latent heat. Supercritical steam generators exploit the continuous density change and single-phase heat addition to reach high mean temperatures of heat supply. Vacuum condensers exploit the opposite end, lowering saturation temperature to widen the cycle while respecting cooling-water limits.

9.9.1 Otto Cycle Description and Efficiency Derived

Otto cycle models spark-ignition engines with air as working fluid, constant specific heats, and heat addition at constant volume. Processes: 1 to 2 isentropic compression, 2 to 3 constant-volume heat addition, 3 to 4 isentropic expansion, 4 to 1 constant-volume heat rejection. On p-v the compression and expansion branches are steep isentropes joined by vertical constant-volume lines. On T-s the isentropes are vertical lines joined by constant-volume curves that rise more steeply than constant-pressure lines because cv is smaller than cp.

Heat added equals cv times T3 minus T2. Heat rejected equals cv times T4 minus T1. Efficiency equals 1 minus heat rejected divided by heat added:

eta_Otto equals 1 minus T4 minus T1 divided by T3 minus T2

Isentropic relations give T2 divided by T1 equals r to the k minus 1 and T3 divided by T4 equals r to the k minus 1, where r equals V1 divided by V2 is the compression ratio. Hence T4 minus T1 divided by T3 minus T2 equals 1 divided by r to the k minus 1, and:

eta_Otto equals 1 minus 1 divided by r to the k minus 1

Efficiency depends only on compression ratio and k. Higher r raises the temperature at which heat is added on average and lowers the temperature at which heat is rejected on average, widening the net area.

Quick Example — Otto Efficiency at Compression Ratio 8

Given air with $k = 1.4$ and $r = 8$.

Step 1: use $\eta_{Otto} = 1 - 1/r^{k-1}$.

Step 2: evaluate $r^{k-1} = 8^{0.4} \approx 2.297$.

Step 3: compute $\eta_{Otto} = 1 - 1/2.297 \approx 0.565$.

Result: efficiency is about $56.5$ percent.

Trap: the ratio $r$ is a volume ratio $V_1/V_2$, never a pressure ratio, and $k$ stays dimensionless.

9.9.2 Diesel Cycle Description and Cut-Off Penalty Derived

Diesel cycle replaces constant-volume heat addition with constant-pressure addition, modelling fuel burning into air already compressed beyond spark limits. Processes: 1 to 2 isentropic compression, 2 to 3 constant-pressure heating, 3 to 4 isentropic expansion, 4 to 1 constant-volume rejection. Cut-off ratio rc equals V3 divided by V2 measures how long heat addition persists along the constant-pressure branch.

Heat added equals cp times T3 minus T2. Heat rejected equals cv times T4 minus T1. Efficiency:

eta_Diesel equals 1 minus T4 minus T1 divided by k times T3 minus T2

Isentropic and constant-pressure relations give T2 equals T1 r to the k minus 1, T3 equals T2 rc, and T4 equals T3 times rc divided by r to the k minus 1. Substitution yields:

eta_Diesel equals 1 minus 1 divided by r to the k minus 1 times rc to the k minus 1 divided by k times rc minus 1

The bracket exceeds unity and grows with rc. At fixed compression ratio, Diesel efficiency therefore sits below Otto efficiency, and admitting more fuel at longer cut-off lowers efficiency because late heat arrives with diminished expansion potential.

9.9.3 Dual Cycle as the Mixed Intermediate

Dual cycle adds part of the heat at constant volume and the remainder at constant pressure. Real high-speed diesel combustion follows this mixed pattern, with a premixed spike followed by diffusion burning. Efficiency lies between Otto and Diesel for comparable inputs, rising when a larger fraction of heat is added at constant volume. The Dual formula carries both a pressure ratio for the constant-volume leg and a cut-off ratio for the constant-pressure leg, collapsing to Otto when cut-off is unity and toward Diesel when the constant-volume share vanishes.

9.9.4 Brayton Cycle Description and Pressure-Ratio Formula

Brayton cycle is the steady-flow cousin for gas turbines. Processes: 1 to 2 isentropic compression in the compressor, 2 to 3 constant-pressure heat addition in the combustor, 3 to 4 isentropic expansion in the turbine, 4 to 1 constant-pressure rejection, closed-loop equivalent of exhaust and intake. On T-s the compression and expansion are vertical isentropes joined by flatter constant-pressure curves.

Heat added equals cp times T3 minus T2, heat rejected equals cp times T4 minus T1. Isentropic temperature-pressure relations with pressure ratio rp equals p2 divided by p1 give T2 divided by T1 equals rp to the k minus 1 divided by k and T3 divided by T4 equals the same group. Efficiency reduces to:

eta_Brayton equals 1 minus 1 divided by rp to the k minus 1 divided by k

Efficiency rises with pressure ratio while net specific work peaks and then falls, because compressor work grows faster than turbine work at high ratios. Back work ratio, compressor work divided by turbine work, is large in gas turbines, often 40 to 60 percent, which is the signature Brayton weakness exploited by intercooling, reheating, and regeneration variants. Refinery flare-gas power and captive gas turbines operate on this cycle.

9.9.5 Rankine Cycle Description and Enthalpy Efficiency

Rankine cycle is the vapour standard for steam plant. Processes: pump compression of liquid from condenser pressure to boiler pressure, near isentropic; constant-pressure heating, evaporation, and superheating in boiler and superheater; near-isentropic expansion in the turbine; constant-pressure condensation back to liquid. Pump work is tiny because liquid specific volume is tiny, so net work is dominated by turbine output minus a small pump debit.

Efficiency in enthalpy terms:

eta_Rankine equals h3 minus h4 minus h2 minus h1 divided by h3 minus h2

with state 1 pump inlet, state 2 pump exit, state 3 turbine inlet, state 4 turbine exit. On T-s the pump rise is a short steep segment, heating follows the compressed-liquid line into the dome, evaporation runs horizontal at saturation, superheat climbs toward turbine inlet, expansion drops toward the dome, and condensation returns horizontal. Superheat shifts turbine exhaust toward drier states and raises mean temperature of heat addition, the twin Rankine gains.

9.9.6 Comparison Table

Cycle	Heat addition	Efficiency expression	Efficiency rises when	Refinery embodiment
Otto	constant volume	1 minus 1 divided by r to the k minus 1	compression ratio rises	petrol standby generators
Diesel	constant pressure	Otto value penalized by cut-off bracket	compression ratio rises, cut-off falls	diesel gensets and fire pumps
Dual	mixed volume plus pressure	between Otto and Diesel	constant-volume heat share rises	high-speed diesel engines
Brayton	constant pressure in flow	1 minus 1 divided by rp to the k minus 1 divided by k	pressure ratio rises toward optimum	gas turbines and flare-gas power
Rankine	constant pressure in flow with phase change	net enthalpy work divided by heat added	superheat, reheat, regeneration added	captive steam and cogeneration plant

9.9.7 The Same-r Versus Same-Peak Reversal Proved

Two comparison constraints give opposite winners, and both results follow from the cut-off bracket plus knock-limited compression.

Same compression ratio: Otto wins. Proof: at equal r the Diesel bracket exceeds unity, so eta_Diesel equals eta_Otto minus a strictly positive penalty. On T-s, constant-volume addition reaches a higher peak temperature and pressure than constant-pressure addition from the same compressed state, enclosing more net area for the same heat rejected. The longer the cut-off, the larger the penalty.

Same peak pressure and temperature: Diesel wins. Proof sketch on T-s: cap both cycles at the same maximum p and T. Otto must compress less to respect the peak under constant-volume heat addition, while Diesel compresses more and adds heat along constant pressure beneath the cap. Diesel therefore rejects less heat at the low-temperature leg and encloses more net area. Physical enabler: Diesel engines admit fuel only after air compression, so no premixed end gas exists to autoignite, permitting compression ratios of 14 to 22 far above spark-ignition knock limits near 8 to 12. The winner flips with the constraint, summarized as same-r favours Otto and same-peak favours Diesel.

9.10.1 Four-Stroke Versus Two-Stroke Mechanics

Four-stroke operation sequences intake, compression, power, and exhaust over two crankshaft revolutions. Intake draws charge through the inlet valve, compression seals both valves and raises pressure, the power stroke expands burning gas against the piston, and exhaust expels products through the exhaust valve. One power stroke per two revolutions gives clean gas exchange, effective part-load control by throttling or fueling, and superior fuel economy.

Two-stroke operation completes the cycle each revolution using ports and crankcase compression instead of poppet-valve strokes. The descending piston compresses the crankcase charge while expanding burning gas, then uncovers transfer and exhaust ports so fresh charge scavenges exhaust. Firing frequency doubles for the same speed and size, raising specific power, but scavenging is imperfect, fresh charge dilutes with exhaust, and specific fuel consumption and hydrocarbon emissions suffer. Large marine and stationary diesels use uniflow scavenging with blowers to manage the compromise. Refinery emergency machines favour four-stroke diesels for reliability and economy.

9.10.2 The SI Versus CI Knock Reversal Explained

Knock names two opposite failures, and fuel chemistry cures one while aggravating the other.

SI knock is end-gas autoignition ahead of the flame front. After spark initiation, the flame traverses the chamber while the unburned end gas is compressed and heated. If the end gas self-ignites, pressure spikes, metallic ringing, and overheating follow. Cure: high-octane fuel that resists self-ignition, plus compact chambers, cool intake, retarded timing, and moderate compression.

CI knock is excessive premixed burning after a long ignition delay. Fuel sprays into hot compressed air, and if the delay is long, a large mass of fuel accumulates and then burns almost simultaneously, producing a harsh pressure rise. Cure: high-cetane fuel that ignites readily with short delay, plus high compression temperature, good atomization, and pilot injection.

Chemistry therefore reverses: branched paraffins, aromatics, and alcohols resist autoignition and raise octane while lengthening diesel delay and lowering cetane. Straight-chain paraffins ignite readily and raise cetane while inviting spark-engine knock and lowering octane. Iso-octane defines 100 on the octane scale, n-heptane defines 0. N-hexadecane, cetane proper, defines 100 on the cetane scale. Refinery blending segregates these molecular families into petrol and diesel pools accordingly.

9.10.3 Octane, Cetane, Supercharging, and Combustion Chambers

Octane rating measures resistance to SI knock under standardized test conditions, reported as research and motor variants. Cetane number measures readiness to autoignite under diesel test conditions. High-performance SI chambers use pent-roof or hemispherical shapes with central plugs to shorten flame travel. CI chambers use direct-injection bowls or divided prechambers to manage mixing and delay.

Supercharging raises inlet density with a compressor, admitting more air and fuel per stroke and raising power from a fixed displacement. Intercooling the compressed charge compounds the density gain and tempers knock. In SI engines supercharging pushes the end gas toward knock and demands octane margin or retarded timing. In CI engines supercharging shortens delay, eases cold starting, and smooths combustion. Turbochargers recover exhaust energy for the compressor, while superchargers draw shaft work directly.

9.10.4 Morse Test and Heat Balance Meaning

Morse test determines indicated power without an indicator diagram. Hold engine speed constant on a multi-cylinder engine, measure brake power with all cylinders firing, then cut each cylinder in turn while holding speed, recording brake power each time. The drop on cutting cylinder i equals its indicated power under the assumption that friction and pumping losses of the remaining cylinders stay nearly constant. Summing over cylinders gives total indicated power, and mechanical efficiency equals brake power divided by indicated power. The method isolates combustion performance from drivetrain and friction losses.

Heat balance sheet audits fuel energy into brake power, cooling-water loss, exhaust loss, lubricating-oil loss, and radiation plus unaccounted remainder. Percentages sum to 100. A complete balance reveals whether poor economy comes from combustion, heat rejection, or mechanical loss. Large refinery diesel installations log the balance alongside fuel assay data to schedule overhaul and injector service.

9.11.1 Stage Work and Isothermal Versus Adiabatic Ideals

Steady-flow polytropic shaft work per stage for mass flow m_dot, gas constant R, inlet temperature T1, pressure ratio rp, and index n is:

W_dot equals n divided by n minus 1 times m_dot R T1 times rp to the n minus 1 divided by n minus 1

Derivation integrates v dp along p V to the n equals constant, converting boundary work by the factor n. Isothermal compression with n equals 1 is the minimum-work limit:

W_dot_iso equals m_dot R T1 ln(rp)

Achieved only with perfect interstage and intrastage cooling. Adiabatic shaft work with n equals k is higher because the gas heats as it compresses and resists further compression. Real polytropic paths fall between, with cooled reciprocating stages nearer isothermal and fast uncooled rotary stages nearer adiabatic. Cooling during compression always pays in work, quite apart from material limits.

9.11.2 Volumetric Efficiency with Clearance Derived

Clearance volume Vc remains at discharge pressure when the piston reaches top dead centre, then re-expands to suction pressure before fresh charge enters. Let swept volume be Vs, clearance ratio C equals Vc divided by Vs, suction state 1 and discharge state 2 with pressure ratio rp. Re-expansion follows p V to the n equals constant, so the re-expanded clearance volume at suction pressure is Vc times rp to the 1 divided by n.

Effective suction volume equals Vs plus Vc minus Vc times rp to the 1 divided by n. Dividing by Vs gives volumetric efficiency:

eta_vol equals 1 plus C minus C times rp to the 1 divided by n

Clearance always depresses breathing, and the penalty steepens with pressure ratio because the trapped mass grows denser and its re-expansion steals a longer fraction of the stroke. High-ratio single-stage duty therefore breathes poorly and discharges hot. The remedy is multistaging with modest ratio per stage.

Quick Example — Clearance Rise From 5 Percent to 10 Percent

Given $r_p = 4$, $n = 1.3$, and clearance $C = 0.05$ versus $C = 0.10$.

Step 1: evaluate $r_p^{1/n} = 4^{1/1.3} \approx 2.905$.

Step 2: use $\eta_{vol} = 1 + C - C r_p^{1/n}$ to get $\eta_{vol} \approx 0.905$ at $C = 0.05$.

Step 3: repeat at $C = 0.10$ to get $\eta_{vol} \approx 0.810$.

Result: doubling clearance lowers volumetric efficiency by about $9.5$ percentage points.

Trap: clearance penalizes breathing through re-expansion of trapped gas, not through leakage past rings.

9.11.3 Multistage Intercooling Logic

Splitting overall pressure ratio across z stages with intercoolers returning gas to T1 divides the exponent and the temperature rise. With perfect intercooling, stage work is minimized when pressure ratios are equal, each equal to overall ratio to the 1 divided by z. Benefits compound: total shaft work approaches the isothermal ideal, discharge temperature per stage stays within lubrication and seal limits, moisture condenses in intercoolers and aftercoolers where drains remove it, and volumetric efficiency per stage stays high because per-stage rp stays low.

Free air delivery standardizes capacity at intake reference conditions so machines are compared on equal mass terms. Instrument-air and hydrogen-recycle compressors in refineries run multistage intercooled configurations, and a rising interstage or discharge temperature reading points first to intercooler fouling or valve leakage rather than to ambient drift.

9.12.1 Carnot COP Pair Derived

Consider a reversible refrigerator between cold reservoir TL and hot reservoir TH. By the second law, the ratio of heat rejected to heat absorbed equals the ratio of absolute temperatures, QH divided by QL equals TH divided by TL. Work input equals QH minus QL. Coefficient of performance for refrigeration is useful effect divided by work:

COP_R equals QL divided by QH minus QL equals TL divided by TH minus TL

The same machine operated as a heat pump delivers QH as the useful effect:

COP_HP equals QH divided by QH minus QL equals TH divided by TH minus TL

Hence COP_HP equals COP_R plus 1. Both use kelvin. No real cycle operating between the same reservoirs can exceed these values. The pair sets the ceiling against which every vapour compression claim is judged.

Quick Example — Carnot COP Between 260 K and 300 K

Given $T_L = 260$ K and $T_H = 300$ K.

Step 1: use $COP_R = T_L/(T_H - T_L)$.

Step 2: substitute $COP_R = 260/40 = 6.5$.

Step 3: use $COP_{HP} = COP_R + 1 = 7.5$.

Result: the refrigeration ceiling is $6.5$ and the heat pump ceiling is $7.5$.

Trap: always convert reservoir temperatures to kelvin before forming the ratio.

9.12.2 VCRS Components in Flow Order

Vapour compression refrigeration circulates refrigerant through four devices. Evaporator: low-pressure liquid-vapour mixture boils at TL, absorbing the refrigeration load at near-constant temperature. Compressor: near-isentropic compression raises superheated vapour to condenser pressure, consuming shaft work. Condenser: desuperheating, condensation, and often subcooling reject heat to ambient or cooling water. Throttle valve: isenthalpic expansion drops pressure back to evaporator level, flashing part of the liquid. Flow order is evaporator to compressor to condenser to valve and back to evaporator.

On p-h diagrams compression runs steep upward, condensation runs leftward at high pressure, throttling drops vertically at constant h, and evaporation runs rightward at low pressure. The enclosed area scales with work and effect, making subcooling and superheat shifts visible as horizontal displacements.

9.12.3 Subcooling Gain and Superheat Effects

Subcooling the condenser outlet liquid below saturation before throttling lowers the throttled enthalpy entering the evaporator while compressor work stays nearly unchanged. Refrigerating effect h1 minus h4 grows, so COP equals effect divided by work rises. This is the cheapest COP improvement available, recovered through larger condensers, dedicated subcoolers, or cooler cooling media. LPG subcooling and propane refrigeration economizers in gas fractionation apply the same principle at process scale.

Mild suction superheat protects compressors from liquid slugging and wet compression damage at a small COP cost, since superheated suction raises specific volume and discharge temperature slightly. Large superheat wastes work without payload and overheats discharge valves and oil. Flash gas formed in throttling is unavoidable loss from isenthalpic expansion, managed with flash chambers, accumulators, and multi-stage economized circuits in large plants.

9.12.4 Refrigerant Properties Table

Refrigerant	Nature and safety	Thermodynamic and service notes
NH3 R717	natural, toxic, pungent leak self-warning, attacks copper so steel construction	high latent heat, high COP, cold-storage and process standard
R134a HFC	nonflammable, zero ozone depletion, elevated global warming potential	medium pressures, automotive and chiller legacy, phased down under climate rules
CO2 R744	natural, nontoxic, very high operating pressure, critical point near 31 C	transcritical operation in warm climates, cheap and safe, needs robust hardware
HC R290 propane and R600a isobutane	flammable, tiny charge limits, strict leak control	excellent properties, domestic and light commercial standard

Selection balances latent heat, pressure level, oil miscibility, materials compatibility, toxicity, flammability, ozone depletion, and global warming potential. Natural refrigerants dominate new refinery-adjacent specifications wherever charge control permits.

9.12.5 Psychrometry Terms and Comfort Envelope

Air-conditioning load analysis uses moist-air properties. Dry-bulb temperature DBT is the ordinary thermometer reading. Wet-bulb temperature WBT is the reading of a thermometer with a wetted wick in moving air, depressed by evaporation toward the adiabatic-saturation state. Dew-point temperature DPT is the temperature at which cooling at constant moisture condenses the first droplet. Relative humidity RH equals vapour partial pressure divided by saturation pressure at DBT, equivalently moisture content divided by saturated moisture at the same DBT.

Dehumidification requires cooling below DPT so moisture condenses, often followed by reheat to supply comfort temperature without clamminess. Sensible cooling alone follows constant moisture lines, humidification follows near-constant enthalpy paths in steam injection or adiabatic paths in evaporative pads. Exam comfort band centres near 22 to 26 C DBT with 40 to 60 percent RH, the envelope in which control-room and occupied-space comfort specifications are framed.

9.13.1 Superheat, Reheat, and Regeneration Effects

Change	Efficiency	Net work per unit mass	Turbine exhaust dryness	Limiting cost
Superheat to higher T at same p	rises through higher mean temperature of heat addition	rises	rises, safer blades	superheater metallurgy and size
Reheat between turbine stages	rises modestly	rises clearly through added enclosed area	rises	extra piping and pressure drops
Regeneration by bled-steam feed heating	rises clearly through reduced external heat per unit work	falls per kg reaching condenser	nearly unchanged	heaters, pumps, control complexity

Superheat is the first upgrade on any Rankine plant because it improves efficiency, work, and blade life together. Reheat is standard on large reheat-regenerative units where the added loop pays through fuel savings. Regeneration sacrifices some mass flow through the low-pressure turbine to preheat feedwater internally, cutting boiler heat demand per unit power. Refinery captive cogeneration headers implement extraction and back-pressure variants of the same ideas to serve process steam and power jointly.

9.13.2 Condenser Vacuum Logic and Cooling Tower Vocabulary

Lowering condenser pressure lowers saturation temperature at heat rejection, widening the cycle and raising efficiency for the same turbine inlet. Vacuum is sustained by condensing the exhaust steam, venting noncondensables with air pumps or ejectors, and supplying cool circulating water. Limits come from cooling-water temperature, air inleakage, condensate subcooling loss, and low-pressure blade moisture.

Cooling towers serve the condenser through two numbers. Range equals hot-water-in minus cold-water-out and measures the imposed heat load. Approach equals cold-water-out minus ambient wet-bulb and measures how close the tower pushes toward the atmospheric limit. Towers cannot cool below the wet bulb. Large range with small approach demands tall fill and high air flow. Induced-draft fin-fan and hyperbolic tower banks around refinery power blocks are sized on this pair.

9.13.3 Combined Cycle and Nozzle Choking

Combined cycle stacks a Brayton topping plant over a Rankine bottoming plant that recovers gas-turbine exhaust in a heat recovery steam generator. Fuel energy first produces gas-turbine work, then exhaust heat produces steam-turbine work without extra fuel. Modern combined efficiencies break 60 percent on natural gas, far above either cycle alone. Supplementary firing, duct burners, and steam injection tune the balance between power and process steam for cogeneration refineries.

Steam nozzles convert enthalpy to kinetic energy for impulse and reaction blading. Convergent-divergent passages accelerate flow to Mach 1 at the throat when the pressure ratio exceeds the critical value, about 0.546 for superheated steam with k near 1.3. Beyond choking, lowering back pressure cannot pass more mass. The jet then expands supersonically in the diverging section if back pressure permits, or shocks and separates if it does not. Turbine governing, safety valves, and flare-steam injectors all operate against this choking limit.

• Conduction resistances follow by integrating Fourier law, giving L divided by k A for slabs, log ratio over 2 pi k L for cylinders, and radial reciprocal difference over 4 pi k for spheres, with series addition for composites and the area-referred U for exchangers.
• Critical radius equals k_ins divided by h_o for cylinders and twice that for spheres, marking maximum loss. Thin wires below r_c cool faster when wrapped, while large refinery lines sit above r_c where insulation always helps.
• Fins obey the hyperbolic tanh law with m from h P over k Ac. Effectiveness judges whether the fin was worth adding, efficiency judges how close it runs to isothermal, and mL beyond about 2.5 wastes metal.
• Convection coefficient h is recovered from Re for flow regime, Pr for fluid diffusive character, Nu for the unknown being solved for, and Gr with Ra for buoyancy drive. Dittus-Boelter uses exponent 0.4 for heating and 0.3 for cooling because heating thins the resistant near-wall layer.
• Radiation uses sigma T to the fourth in kelvin, Wien peak at 2898 over T, Kirchhoff alpha equals epsilon reasoning, small-in-large exchange without view factors, series-proof shield division by n plus 1, and floating reradiating nodes.
• Exchanger duty equals U A times LMTD, with the log mean derived from differential end balances. Counter flow sustains larger LMTD and permits cold outlet beyond hot outlet, while NTU predicts effectiveness without iteration and fouling enters as one added series resistor.
• Boiling traverses natural convection, nucleate, CHF peak, transition with falling flux, and film regimes. CHF crossing under heat-flux control jumps to burnout. Dropwise condensation beats filmwise by an order of magnitude because the film itself insulates, and noncondensables blanket both.
• Zeroth law founds temperature, first law conserves energy, second law forbids complete conversion and uphill transfer without work. SFEE reduces per device to throttling isenthalpic, nozzle enthalpy-to-speed, and compressor-turbine enthalpy-to-work forms. TdS pair plus polytropic work integral unify property and path calculations, with Maxwell and Clapeyron extracting entropy and dome slopes from p-v-T data.
• Ideal gases obey cp minus cv equals R because flow work p v equals R T differentiates to the gap. Pure water anchors at triple 0.01 C with 0.611 kPa and critical near 374 C with 22.09 MPa, with dryness weighting mixture properties only inside the dome.
• Otto, Diesel, Dual, Brayton, and Rankine efficiencies derive from heat-added over heat-rejected ratios reduced by isentropic relations to compression-ratio, cut-off, pressure-ratio, and enthalpy forms. Same compression ratio favours Otto through the cut-off penalty, same peak pressure and temperature favours Diesel through higher survivable compression free of knock.
• Four-stroke engines trade firing frequency for clean scavenging, two-stroke engines trade economy for power density. SI knock is premature end-gas ignition cured by high-octane resistant chemistry, CI knock is delayed pileup cured by high-cetane ready chemistry. Supercharging densifies charge with opposite knock consequences, Morse test isolates indicated power by cylinder cutout, and heat balance audits fuel energy to closure.
• Compressor shaft work follows the polytropic v-dp integral with isothermal log form as minimum. Clearance gives eta_vol equals 1 plus C minus C times rp to the 1 over n, penalizing high ratios. Equal-ratio multistaging with intercooling approaches isothermal work while capping temperature and preserving breathing.
• Carnot COP pair TL over TH minus TL and TH over TH minus TL ceilings every refrigerator and heat pump. VCRS sequences evaporator, compressor, condenser, and valve, with subcooling raising effect at fixed work and controlled superheat guarding against slugging. Refrigerant choice trades latent heat against pressure, materials, toxicity, flammability, and climate metrics, while psychrometry DBT, WBT, DPT, and RH frame comfort near 22 to 26 C and 40 to 60 percent RH.
• Steam plant gains come from superheat, reheat, and regeneration with distinct efficiency, work, dryness, and cost signatures. Condenser vacuum widens the cycle within cooling-water and air-pump limits, towers live on range versus approach against the wet-bulb floor, combined Brayton plus Rankine plant exceeds 60 percent by stacking ceilings, and nozzles choke at Mach 1 at the throat beyond the critical pressure ratio.