Oil & gas

Shell-and-tube heat exchanger

Fouling costs a little every day. The twin tells you when cleaning pays.

A crude preheat exchanger saves furnace fuel by moving heat from a hot product into the crude. Deposits build on the tubes and the duty falls a little each day. The twin works out U from the four temperatures and two flows, prices the lost duty, and says when a cleaning pays for itself.

TEMA AES, crude preheat, 8 MW, U clean 450 W/m²·K, 420 m²

Built and run to TEMA, API 660

Sketch of a TEMA AES shell-and-tube heat exchanger: front channel with pass partition, tube bundle, segmental baffles in the shell, floating head and shell cover, with crude on the tube side, hot product on the shell side, and numbered temperature, flow and pressure-drop points.12345678crude incrude outhot inhot outfloating headchannel
Duty, clean
8.0 MW
U at 18 months
331 W/m²·K
Best cleaning interval
365 days
Payback at 18 months
170 days

Anatomy and nameplate

Sketch of a TEMA AES shell-and-tube heat exchanger: front channel with pass partition, tube bundle, segmental baffles in the shell, floating head and shell cover, with crude on the tube side, hot product on the shell side, and numbered temperature, flow and pressure-drop points.12345678crude incrude outhot inhot outfloating headchannel
  1. 1Crude inlet temperature. Cold-side start of the heat balance.
  2. 2Crude outlet temperature. Falls as the tubes foul. The furnace makes up the rest.
  3. 3Hot inlet temperature. Hot-side start of the heat balance.
  4. 4Hot outlet temperature. Rises as less heat leaves the hot stream.
  5. 5Crude flow. With the crude temperatures, gives the duty.
  6. 6Hot product flow. Checks the duty from the other side.
  7. 7Tube-side pressure drop. Deposits narrow the bore and the drop climbs.
  8. 8Shell-side pressure drop. Bypass and maldistribution show here.
Shell-and-tube heat exchanger nameplate
Thermal
Duty, clean8.0 MW
U clean450 W/m²·K
Area420 m²
Crude, tube side150 → 200 °C
Hot product, shell side245 → 205 °C
LMTD and F49.9 K, F = 0.85
NTU and effectiveness1.18, 53%
Streams
Crude69.6 kg/s, about 44,500 bbl/d
Hot product74.1 kg/s
Tube-side pressure drop, clean70 kPa
Mechanical
TypeTEMA AES: removable channel and cover, one-pass shell, floating head
PassesOne shell pass, two tube passes
Tubes1 in OD, 19.86 mm bore, carbon steel
BafflesSingle segmental
Fouling modelR_f settles at 9 × 10⁻⁴ m²·K/W, time constant 250 days

From the well to market

Wells lift the fluid. The separator splits it. Oil goes to the tank battery and gas to the compressor station, and both leave by pipeline for processing.

  • Well fluid
  • Crude oil
  • Gas
  • Has its own twin page

The physics

The equations behind every curve on this page.

Duty

Q = U·A·F·LMTD

8.0 MW clean, with LMTD 49.9 K and F 0.85 for one shell pass and two tube passes.

Fouling resistance

R_f = 1/U_dirty − 1/U_clean

Each deposit adds a layer of resistance on top of the clean tube.

Effectiveness

Q = ε·C_min·(T_hot,in − T_cold,in)

Gives the duty a fouled bundle can make with today’s inlet temperatures.

Cleaning payback

Σ (fuel saved) = C_clean

Days until the fuel a cleaning saves adds up to what it costs.

Fouling and cleaning

U falls as deposits build. The lost duty is fuel the furnace burns instead. Cleaning costs a lot at once and puts U back.

Overall heat transfer coefficient over time

Clean U 450 W/m²·K, 420 m²

Overall heat transfer coefficient over time25030035040045002004006008001,000Clean at day 365331 at 18 monthsDays since the bundle was last cleanedU (W/m²·K)
  • Never cleaned
  • Cleaned every 365 days
Fouling at 18 months
8.0 × 10⁻⁴ m²·K/W
U after fouling settles
320 W/m²·K

Deposits add a layer of resistance on top of the clean tube. U falls fastest at first, then levels off. Cleaning puts it back.

R_f = 1/U_dirty − 1/U_clean R_f(t) = R_∞·(1 − e^(−t/τ))

What fouling costs against what cleaning costs

Lost duty made up in the furnace at 85% efficiency, fuel at $4.00/MMBtu

What fouling costs against what cleaning costs010020030040050060002004006008001,0001,2001,40002004006008001,000One cleaning $45kLowest $331/day at 365 daysDays since cleaning, or cleaning interval (days)Extra fuel since cleaning ($ thousand)Cost per day over the cycle ($)
  • Extra fuel since cleaning
  • Cleaning plus fuel, per day of cycle (right axis)
Fuel cost at 18 months
$376 a day
Clean at 18 months, pays back in
170 days

Each day of fouling costs a little. Cleaning costs a lot at once. The best interval is where the two, spread over the cycle, cost least per day.

cost per day = (C_clean + Σ fuel_extra) / T fuel_extra = (Q_clean − Q) / η_furnace

Pressure drop

Deposits that block heat also narrow the tubes. Reading both tells fouling from a flow problem.

Tube-side pressure drop as deposits build

19.86 mm tube bore, deposit conductivity 0.35 W/m·K

Tube-side pressure drop as deposits build6065707580859002004006008001,000Clean 70 kPa80.1 kPa at 18 monthsDays since cleaningTube-side pressure drop (kPa)
  • Tube-side Δp at design flow

The same deposit that blocks heat also narrows the bore. Pressure drop and U moving together point to fouling. U falling with flat pressure drop points to bypass or a flow problem instead.

Δp ∝ d^(−4.75) d = d_clean − 2·R_f·k_deposit

Failure modes

What breaks, what it looks like, what the twin does about it.

Failure modeThe physicsSignals that show itHow the twin catches it
FoulingAsphaltenes, salts and corrosion products build up on the tube walls and add thermal resistance.Crude outlet temperature down at the same flows and inlet temperatures. Tube-side pressure drop up.Works out U and R_f from the heat balance every hour and prices the lost duty in furnace fuel.
Tube leakA cracked or corroded tube lets the higher-pressure stream into the other side.Heat balance no longer closes. Lab results on the low-pressure stream change.Compares the duty from each side. A gap that grows points to a leak, not fouling.
Baffle bypassWorn baffles or a loose bundle let shell-side fluid go round the tubes.U down with shell-side pressure drop down, not up.Reads U and both pressure drops together to tell bypass from fouling.
Flow maldistributionUneven flow across the passes leaves part of the bundle doing little work.U down with flat pressure drop. Outlet temperature swings with flow.Checks the duty against flow over time. Fouling does not change with flow, maldistribution does.

What the twin does not do

  • We say when cleaning pays. Planning the outage and isolating the bundle stay with your operations team.
  • We never change furnace firing or bypass valves. TwinEdge is never in the safety loop.

How we catch it early

A physics twin and a PINN, working on one shell-and-tube heat exchanger.

The twin is a set of equations for this machine, tuned to its nameplate. Give it the flow, speed and conditions and it returns what every reading should be. It always gives the same answer for the same inputs.

A physics-informed neural network, or PINN, sits on top. It is trained on the plant’s own data and it is penalized whenever it breaks a physics rule. So it can only learn what the equations leave out: wear, fouling, friction. It cannot invent an answer that the machine could never produce.

  1. Step 1

    Measure

    Flow, pressure, current, temperature, vibration. Whatever the asset already has.

  2. Step 2

    Physics twin

    The equations of this machine, set to its nameplate. It says what the readings should be right now.

  3. Step 3

    Residual

    Measured minus expected. Demand, weather and speed are already taken out.

  4. Step 4

    PINN

    A small neural network that learns the cause of the gap, under the physics rules.

  5. Step 5

    Finding

    A named cause, a date, and the kWh or hours at stake. An engineer reviews it.

loss = (measured − predicted)² + λ · (physics residual)²

The second term is the rulebook. It is zero when the network obeys the equations.

Why this beats black-box AI

A black-box model learns patterns from history. It needs examples of each failure, and it goes quiet when the pump runs at a point it has not seen. The twin starts from physics, so it works on day one. The PINN adds what physics cannot know about this particular machine.

That is also what makes the savings countable. The baseline is a calculation, not a guess, so the gap between baseline and meter is a number a customer can check.

What the twin checks on this asset

  • U and R_f from the heat balance, every hour
  • Duty from the crude side against duty from the hot side
  • Tube and shell pressure drop against U
  • Cost of fouling to date against the cost of a cleaning
Fixed alarmBlack-box AITwin + PINN
Needs failure history to learnNoYes, lotsNo
Says what is wrongNoRarelyYes, by name
Works at a flow it has never seenNoOften failsYes, the equations still hold
Puts a number on the savingNoNoYes, in kWh and dollars
Lead time on slow wearLateVariesWeeks to months
Every number can be tracedYesNoYes, to an equation

What the twin puts a number on

Energy and cost, from the equations.

Duty lost at 18 months

976 kW

Crude leaves 6.1 °C cooler. The furnace makes up the rest.

Extra furnace fuel

$376 a day

At $4.00/MMBtu and 85% furnace efficiency

Best cleaning interval

365 days

$331 a day over the cycle, with $45,000 per cleaning

Clean at 18 months

Pays back in 170 days

Fuel saved against the cost of one cleaning

What we do for you

We watch it. We catch it. We prove it.

Our reliability engineers run the twin on your shell-and-tube heat exchanger, review every finding, and help your crew fix it. After the repair we compare the meter to the baseline and write the result in the ledger.