agentleFS
equinorequinor/neqsim-community-skills/skills/process/neqsim-heat-exchanger-fouling-assessment/SKILL.md

neqsim-heat-exchanger-fouling-assessment

Skill4 starsChanged 5 months agoApache-2.0View on GitHub

Educational heat-exchanger monitoring and fouling assessment for plate and shell-and-tube coolers, condensers and cooling-medium loops: operating U-value, flow normalisation without scaling the deposit, fouling vs film resistance, maldistribution rule-out, capacity loss and cleaning interval. USE WHEN: a task asks whether an exchanger or cooling loop has fouled, by how much against design U, what it costs in duty or capacity, or how often to clean.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    When a terminal difference is small, the LMTD is sensitive to small temperature errors.

    line 79
  2. Distinctive:

    No proprietary vendor rating software, data sheets, or company specifications are included.

    line 188

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Heat Exchanger Fouling and Performance Assessment

Use this skill to answer the recurring operational question "has this exchanger degraded, by how much, and what is it costing me?". It reduces one operating snapshot to an overall coefficient, normalises that coefficient to design flow correctly, splits the shortfall into fouling and film resistance, rules out area effects before blaming the deposit, and converts the answer into the operating limit a production engineer can act on.

When to Use

  • A cooler, condenser, or cooling-medium exchanger is suspected of fouling or of limiting capacity.
  • A measured U-value must be compared against a vendor data-sheet design U at different flows.
  • A cleaning interval, or the value of cleaning at all, has to be justified.
  • A "the valve is wide open and it still will not cool" complaint needs an algebraic explanation.

Inputs

  • area: heat-transfer area in m2 (the credited area, see the maldistribution rule-out).
  • design_overall_coefficient: data-sheet overall coefficient in W/(m2 K).
  • hot_mass_flow, hot_specific_heat, hot_inlet_temperature, hot_outlet_temperature.
  • cold_mass_flow, cold_specific_heat, cold_inlet_temperature, cold_outlet_temperature.
  • design_hot_mass_flow, design_cold_mass_flow: data-sheet flows, for normalisation.
  • design_fouling_allowance: fouling resistance already credited in the design U, m2 K/W.
  • wall_resistance: wall and contact resistance in m2 K/W, default 0.
  • hot_film_coefficient, cold_film_coefficient: design films in W/(m2 K); when absent they are split out of the design U using hot_film_resistance_fraction.
  • lmtd_correction_factor: F factor for multi-pass or cross-flow, default 1.0.
  • duty_source: auto, hot, cold, or mean.
  • installed_units: number of parallel units sharing the area, for the maldistribution scan.

Outputs

  • duty_kw, duty_hot_kw, duty_cold_kw, duty_imbalance_fraction, duty_source.
  • lmtd_k, lmtd_reliable, effectiveness, ntu, capacity_ratio.
  • u_measured_w_m2k and u_measured_basis (lmtd or effectiveness-ntu).
  • u_normalised_to_design_flow_w_m2k and u_naive_one_sided_w_m2k, with naive_optimism_percentage_points.
  • performance_ratio: normalised U as a fraction of design U.
  • fouling_resistance_m2k_w, design_fouling_allowance_m2k_w, excess_fouling_resistance_m2k_w.
  • condition: clean, within-allowance, fouled, or heavily-fouled.
  • maldistribution_scan: apparent U crediting N, N-1, N-2 ... units.
  • warnings, assumptions.

Separate entry points return CapacityLimitResult and CleaningIntervalResult.

Engineering Method

1. Duty reduction

Take the duty from the better-instrumented side and close the other side from the energy balance. State which side was measured and which was derived. The model reports both and the imbalance fraction; an imbalance above tolerance is a warning, not a silent average.

2. Flow normalisation, the two-sided form

A film coefficient scales roughly as $G^{0.8}$. A fouling deposit does not: it is a fixed conduction resistance, and pushing more flow through it does not make it thinner. Only the two convective films may be re-evaluated:

$$\frac{1}{U}=\underbrace{\frac{1}{h_1 (G_1/G_{1,d})^{0.8}}+\frac{1}{h_2 (G_2/G_{2,d})^{0.8}}+R_{wall}}{\text{flow dependent}}+\underbrace{R_f}{\text{flow independent}}$$

The common error is the one-sided form

$$U_{scaled}=\left(\frac{G_1}{G_{1,d}}\right)^{0.8} U$$

which multiplies the whole coefficient, fouling included. It is always optimistic, and the error can exceed the measurement uncertainty of the whole study. The model computes both and reports the gap in percentage points of design so the optimistic form cannot be used unnoticed.

3. Fouling resistance against design allowance

$$R_f = \frac{1}{U_{meas}} - R_{films} - R_{wall}$$

A vendor design U usually already contains a fouling allowance, so the measured shortfall is fouling in excess of design margin, not total fouling. Plate exchangers in seawater are typically designed with 0.2 to 0.5 x 10^-4 m2 K/W, far tighter than shell-and-tube, because of the high channel shear.

4. Effectiveness-NTU in the low-approach regime

When a terminal difference is small, the LMTD is sensitive to small temperature errors. Below a 3 K terminal difference the model switches the coefficient basis to effectiveness-NTU and says so:

$$\varepsilon = \frac{Q}{C_{min}(T_{h,in}-T_{c,in})},\qquad NTU = \frac{1}{1-C_r}\ln\frac{\varepsilon-1}{\varepsilon C_r-1},\qquad U=\frac{NTU,C_{min}}{A}$$

5. The maldistribution rule-out

Apparent U is inversely proportional to the credited area. Before attributing everything to fouling, recompute U crediting N-1, N-2 ... effective units. If the shortfall disappears at N-1, the finding is an area or distribution problem (a blocked pass, a bypassing unit, a closed branch), not a film problem, and cleaning will not fix it.

6. Capacity limit

For a fixed-rate loop holding a supply set point:

$$Q_{max}=\frac{\varepsilon C_{min}(T_{set}-T_{cold,in})}{1-\varepsilon C_{min}/C_h}$$

which goes to zero as the cold inlet approaches the set point. This is the algebraic statement of "it is the temperature difference that runs out, not the valve".

7. Cleaning interval

Two fouling resistances and the time between them give a rate, hence the time to a limit. Frequency cannot fix a clean that is not effective: with a residual fraction left behind, the achievable cycle shortens, and once the residual exceeds the limit no interval recovers the duty and the cleaning method itself must change.

Python Usage Pattern

from heat_exchanger_fouling_assessment import HeatExchangerFoulingModel

model = HeatExchangerFoulingModel()
result = model.evaluate(
    area=420.0,
    design_overall_coefficient=1250.0,
    hot_mass_flow=180.0,
    hot_specific_heat=3.6,
    hot_inlet_temperature=338.15,
    hot_outlet_temperature=318.15,
    cold_mass_flow=260.0,
    cold_specific_heat=4.0,
    cold_inlet_temperature=288.15,
    cold_outlet_temperature=300.65,
    design_hot_mass_flow=220.0,
    design_cold_mass_flow=300.0,
    design_fouling_allowance=2.0e-4,
    installed_units=3,
)

print(result.condition, result.performance_ratio)
print(result.naive_optimism_percentage_points)

limit = model.capacity_limit(
    effectiveness=result.effectiveness,
    hot_capacity_rate=180.0 * 3.6,
    cold_capacity_rate=260.0 * 4.0,
    set_point_temperature=318.15,
    cold_inlet_temperature=298.15,
    required_duty=result.duty_kw,
)

cycle = model.cleaning_interval(
    fouling_resistance_start=2.0e-5,
    fouling_resistance_end=result.fouling_resistance_m2k_w,
    elapsed_days=180.0,
    fouling_resistance_limit=6.0e-4,
    cleaning_residual_fraction=0.4,
)

Related NeqSim Functionality

For validated, design-grade calculations, redirect to NeqSim:

  • neqsim.process.equipment.heatexchanger.HeatExchanger — rigorous two-stream exchanger with UA, duty, and outlet conditions on a real fluid.
  • neqsim.process.equipment.heatexchanger.Cooler and Heater — single-stream duty with rigorous enthalpy.
  • neqsim.process.mechanicaldesign.heatexchanger.ThermalDesignCalculator and BellDelawareMethod — thermal-hydraulic sizing and shell-side film coefficients.
  • neqsim.process.mechanicaldesign.heatexchanger.HeatExchangerDesignFeasibilityReport — TEMA/ASME feasibility verdict and cost.
  • neqsim.process.equipment.heatexchanger.heatintegration.PinchAnalysis — where the exchanger sits in the plant heat recovery.

Related community skills:

  • neqsim-control-authority-screening — the other half of the "wide open and it still will not cool" complaint: a degraded exchanger consumes the cooling valve's margin until the loop has no authority left. Run both before blaming an external disturbance.
  • neqsim-root-cause-analysis — the framework that consumes a fouling verdict as hypothesis evidence.

A NeqSim flash gives the specific heats and the property basis; take the cooling-medium heat capacity from the fluid model rather than a constant when the medium is water-rich or glycol-rich.

Validation Checklist

  • Both side duties are reported and the imbalance is stated, with the measured side named.
  • The design U basis is quoted from a data sheet, with its fouling allowance and its design flows.
  • Flow normalisation uses the two-sided form; the one-sided value, if quoted anywhere, is flagged as optimistic.
  • The coefficient basis (LMTD or effectiveness-NTU) is stated, with the terminal differences.
  • The maldistribution scan is run before fouling is asserted.
  • The capacity statement names the constraint (temperature difference, flow, or area).
  • Cleaning recommendations state the assumed cleaning effectiveness.
  • Qualified human review is completed before operational or investment decisions.

Common Mistakes

SymptomCauseFix
Performance looks better than it isOne-sided flow normalisation scaling the whole UUse the two-sided form; hold the fouling resistance out of the scaling
Fouling looks worse than it isDesign U compared without its own fouling allowanceReport fouling in excess of allowance, not total
U swings wildly between snapshotsLMTD near a close approachSwitch to effectiveness-NTU and check instrument accuracy
Everything blamed on foulingFull installed area credited when a unit is bypassingRun the N-1, N-2 credited-area scan first
Cleaning more often does not helpCleaning leaves most of the depositFix cleaning effectiveness before interval
Duty short but valve saturatedTemperature approach exhaustedUse the capacity-limit form; the constraint is not the valve

Limitations

  • Screening reduction of a steady snapshot; no transient, no tube-by-tube, no vendor rating method.
  • Counter-current terminal differences; multi-pass and cross-flow need an externally supplied F factor.
  • Constant specific heats; no phase change, so condensers and reboilers need a latent-duty treatment.
  • The film split from a design U is an assumption unless the data sheet gives both films.
  • No proprietary vendor rating software, data sheets, or company specifications are included.

References

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.