Gas Volume
Pressure
Temperature
Heating Value (Natural Gas)
Pipe Volume Calculator
Z-Factor Estimator (Papay)
Petroleum Gravity (°API ↔ SG ↔ Density)
°API = 141.5 / SG − 131.5 · density = SG × 999.016 kg/m³ (water at 60 °F, ASTM D1250 basis).
Viscosity (Dynamic ↔ Kinematic)
ν = μ / ρ · 1 cSt = 1 mm²/s · 1 cP = 1 mPa·s. Water at 20 °C ≈ 1 cP ≈ 1 cSt.
Mass ↔ Volumetric Flow (by density)
Q_vol = ṁ / ρ. For gases, use the density at the stated T & P (or the compositional ρ_std from the Advanced tab).
Gas Property Estimator
SCREENINGScreening estimates, not standards. Z uses Papay (1968) with Standing-Katz pseudo-criticals — the same routine as the Z-Factor Estimator. μ uses Lee, Gonzalez & Eakin (1966), a published correlation fitted to four sweet natural gases over 100–340 °F and 100–8,000 psia (≈2.7 % AAE); it carries no H₂S/CO₂/N₂ correction, so treat sour or high-inert gas as indicative only. c = √(k·Z·R·T/M) is the real-gas form used in API 520 for critical flow; it neglects the (∂Z/∂P)_T and (∂Z/∂T)_P terms. μ_JT is exact thermodynamics wrapped around Papay's ∂Z/∂T and an ideal-gas C_p = kR/(k−1) — positive means cooling on expansion. Gas MW = SG × 28.9647.
Steam Properties (IAPWS-IF97)
REGIONS 1·2·4IAPWS-IF97 (Revised Release, 2007), Regions 1 (compressed liquid), 2 (steam) and 4 (saturation line) — the same industrial formulation behind commercial steam tables. All 259 coefficients are carried verbatim from the Release and re-verified in CI against its own computer-program verification tables (Tables 5, 15, 35, 36) to 9 significant figures. Valid 0–800 °C and up to 100 MPa; Region 3 (dense/near-critical, roughly 350–590 °C above 16.5 MPa) is out of scope — the card says so rather than extrapolating. Saturation enthalpies h_f/h_g are Region 1/2 values at T_sat(P), available up to 16.529 MPa. Btu(IT) conversions: 1 Btu/lb = 2.326 kJ/kg exactly; 1 Btu/(lb·°R) = 4.1868 kJ/(kg·K) exactly.
Pump Suction — NPSHa
SCREENINGNPSHa = (P_surface − P_v)/(ρ·g) + z − h_f with g = 9.80665 m/s² exactly — a first-principles head balance at the impeller centerline, not a standard's method. z is the liquid level above the centerline (negative for a suction lift); h_f is the suction-line friction head (the ΔP card's frictional term is a good source). The water helper uses IAPWS-IF97: Pv = p_sat(T) (Region 4) and ρ = ρ_f(T), the saturated-liquid density (Region 1 on the saturation line), valid 0–350 °C. This card deliberately renders NO verdict: the required margin over NPSHr belongs to the pump vendor's curve and the Hydraulic Institute / API 610 margin guidance, which are licensed documents this application does not reproduce.
Compressor — Head & Power
ISENTROPIC · POLYTROPICFirst-principles screening: H_poly = Z_avg·(R/M)·T₁·(r^m − 1)/m with m = (k−1)/(k·η_p), and H_is with m_s = (k−1)/k — the textbook isentropic and perfect-gas polytropic relations, not a performance-test procedure (ASME PTC 10 is a licensed test code and is not used here). Real-gas correction enters through Z_avg = (Z₁ + Z₂)/2 from the same Papay correlation as the Z-Factor card; Z₂ is evaluated once at the perfect-gas discharge estimate T₂ = T₁·r^m, a deterministic single pass. T₂ is the perfect-gas polytropic path temperature — real-gas and cooling effects will shift the actual value. Gas power = ṁ·H_poly/η_p, the enthalpy rise rate only: add mechanical losses (bearings, seals, gearbox) per the vendor. Unit factors are exact by definition: 1 ft·lbf/lbm = 9.80665 × 0.3048 = 2.98906692 J/kg; 1 hp = 550 ft·lbf/s = 745.69987 W. Typical polytropic efficiencies run ≈ 70–85 % for centrifugal machines — use the vendor's number when you have it.
Gas Heating Value (Compositional) & Flow
1. GAS COMPOSITION
2. Conditions & Operating State
3. Physical & Combustion Properties
4. Mass & Volumetric Flow Conversions
LNG Cargo Estimator
1. Vessel
Photo: —
Source: —
View on MarineTraffic ↗ · VesselFinder ↗ (links by IMO, new tab — nothing fetched or embedded)
Fleet data: IGU World LNG Report 2026, Appendix 3 (data: Rystad Energy) — the full table of all 804 LNG carriers active at end-2025, reproduced with the International Gas Union’s written permission. The 36 featured vessels additionally cite their own primary sources. Please do not re-extract the data from this tool.
2. Cargo
Capacity pre-fills from the vessel and stays editable. Loading limit: the IGC Code Ch. 15 filling limit is 98 % unless a higher limit is approved for the ship; 98.5 % is common practice — use the vessel's certificate. Heel, boil-off rate and voyage days are optional; at zero the card shows loaded quantities only.
3. Cargo quantities — loaded → delivered
Loaded = capacity × loading limit. Delivered = loaded − heel − BOR × days. Mass = V × ρ_liq (ISO 6578 at the LNG temperature set above); standard volume = mass ÷ ρ_std; energy = standard volume × HHV — the custody-transfer basis; LHV is shown for reference. Copy buttons are unit-aware.
Fleet catalogue — the full active fleet, end-2025
Thumbnails: Wikimedia Commons photographs where a CC-licensed one exists (author and licence under each — full list in assets/vessels/CREDITS.json), otherwise an original schematic by containment type. No MarineTraffic / VesselFinder / ShipSpotting imagery is used. The 36 featured vessels carry their own public sources; all other entries: IGU World LNG Report 2026, Appendix 3 (data: Rystad Energy) — reproduced with permission.
Pipe Delta Pressure Calculator (Darcy-Weisbach)
1. SYSTEM & PIPE PARAMETERS
2A. VAPOR PHASE
2B. LIQUID PHASE
2C. FITTINGS & VALVES (OPTIONAL) ΣK = 0.00
Crane TP-410 resistance coefficients, K = n·f_T, where f_T is the fully-turbulent friction factor for the nearest nominal size. Entrance and exit are direct K values with no n. Leave all counts at 0 to exclude fittings entirely.
3. HYDRAULIC OUTPUTS
Awaiting Calc...Line-sizing limits are NORSOK P-001 §6.3.2/§6.4 and Tables 3–4, applied to the frictional ΔP only (static head is excluded — it is not a sizing criterion). Gas: v_max = min(175·ρ^−0.43, 60) m/s. Two-phase: v_max = min(183·ρ_mix^−0.5, 25) m/s non-corrosive, 10 m/s corrosive. Liquid velocity bands are material-dependent; Cu-Ni is the lowest at 3 m/s because of seawater erosion. Pump-circuit limits are ΔP-based. Screening guidance for line sizing — not a substitute for a full hydraulic study, and independent of the API RP 14E erosional check above.
The TWO-PHASE METHOD selector changes the frictional term only: Lockhart-Martinelli uses the Chisholm C constants (20/12/10/5 by the phase-alone flow regimes), Müller-Steinhagen-Heck (1986) interpolates between the all-liquid and all-gas gradients, and Friedel (1979) additionally uses the surface tension input. Static head, fittings ΔP and the velocity / Reynolds / friction-factor readouts always stay on the homogeneous no-slip mixture, which keeps the API RP 14E and NORSOK screens comparable across methods. All are screening-level correlations — see Theory Part X.
Flow Regime
Awaiting Visualization...Visualizes the two-phase flow pattern from the inputs of the Pipe Delta Pressure Calculator above. Inclination θ = asin(Δz / L) is taken from PIPE LENGTH and ELEVATION CHANGE. θ ≥ +30° uses the vertical-flow map (Hewitt & Roberts type); θ < +30° uses the horizontal-flow map (Baker type). The map is rendered server-side with Python seaborn. Boundaries are simplified — indicative only.
3D FLOW ANIMATION
Conceptual animation — phase distribution, layer depths, speeds and inclination are schematic, scaled from jG, jL, the no-slip holdup λL and θ. Drag to orbit, scroll to zoom. Not a CFD result.
Gas Turbine Selection
Selecting a model pre-fills the estimator below with its ISO rating and efficiency (simple-cycle GT, or GTCC plant values in Combined cycle mode). All figures are indicative vendor catalogue data at ISO conditions (15 °C, 101.325 kPa, 60% RH), natural gas. You can overwrite any input.
Fuel Gas Rate Estimator
| Period | Fuel volume | Fuel mass | Fuel energy in | Power sent out |
|---|---|---|---|---|
| Hourly (continuous) | — | — | — | — |
| Daily | — | — | — | — |
| Monthly (730 h) | — | — | — | — |
| Yearly (8,760 h) | — | — | — | — |
Screening estimate: Q_fuel = P / η (LHV basis); V = 3600·Q / LHV; ṁ = V·ρ_std. Daily/monthly/yearly totals apply the availability factor; the hourly row is the instantaneous rate at load. Efficiency and heat-rate definitions per ISO 3977-2 / ASME PTC 22 convention (HR = 3600/η). Actual fuel consumption varies with site conditions, degradation, part load and fuel composition — verify against vendor performance data before commercial use.
Gas Turbine Catalogue
Indicative ISO ratings from public vendor catalogues (natural gas). Click a card to load the model into the estimator.
Thumbnails are original schematic illustrations (not vendor imagery). Data sources: MHI GTCC brochure METP-11GT01E1-E-0; GE Vernova fact sheets GEA35750/GEA35768 and product pages; Siemens Energy product pages. Ratings are net or gross per each vendor's published convention and change with catalogue revisions — always confirm against the current vendor datasheet.
PRV Sizing Calculator (API 520 Part I)
Gas / Vapor Inputs (§5.6)
Steam Inputs (§5.7)
Liquid Inputs (§5.8 Certified)
Two-Phase Inputs — Omega Method (§5.10 / Annex C.2.2)
Sizing Results
Awaiting Calc...How To Use — Operations Manual v3.9
A step-by-step visual guide covering every panel of the O&G Engineering Converter. Use the wireframe diagrams and annotations to master each feature.
1. Application Header & Tab Navigation
The dark header bar contains the main navigation tabs. Click any tab to switch context; the active tab shows an amber underline. The Advanced tab has a second layer: a segmented strip directly under the bar with three sub-tabs — Gas Quality & LNG Cargo (heating value, density, flow conversions and the LNG Cargo Estimator), Hydraulics (Pipe ΔP and Flow Regime) and GT Fuel (turbine selection, fuel-gas estimator and catalogue). The strip remembers your last choice, and a share link opens the exact sub-tab.
2. Mobile Navigation & Sharing Custom Modules
On a phone or narrow window, the nine-tab bar is replaced by a single button showing the current section. Tap it for a dropdown listing every tab; choosing one switches section and closes the menu. Tapping outside or pressing Esc also closes it. At tablet width and above the familiar tab bar returns unchanged.
Share links now carry your custom modules. Any converter cards you built in the General tab travel with the link — title, unit labels and conversion factor — and are added to the recipient's browser when they open it, with a brief notice of how many arrived. They persist, so the recipient can keep using them; each can be removed with its own TERMINATE button.
Share links encode state after the # in the URL, so nothing is ever sent to a server. They can get long — a full session with several custom modules easily exceeds 2,000 characters, and some chat and mail clients silently truncate at about that length. When a copied link passes that threshold the confirmation says so; the link is still valid, but check that it pasted whole.
3. Unit-Aware Copy
Every copy button in the application copies the bare number on a plain click — exactly as before, so a value pasted into a spreadsheet cell stays a number. To copy the value with its unit (for a report, an email, a datasheet), hold Ctrl (Windows) or ⌘ (Mac) — Shift works too — while clicking. On a phone or tablet, press and hold the copy button for ≈ 0.5 s. A toast confirms what landed on the clipboard, e.g. 95.18714 kJ/kg.
The unit is read live from the control next to the value — change a unit selector and the next copy carries the new unit, including the VOL/MOL flow outputs, which follow whichever selector is currently visible. Dimensionless results (Z-factor, SG, Wobbe Index, the API 526 orifice letter) never gain a suffix, and the PSV orifice area keeps embedding its unit in the displayed text as it always has. Custom-module cards get the same treatment with their own unit labels.
4. General Tab — Standard Converter Blocks
Pre-built converter cards for the most common O&G unit pairs. Each card has two input fields; editing either side auto-converts the other.
5. Custom Modules — Configure & Save
Create unlimited custom unit converters using a linear multiplier (Unit1 × Factor = Unit2). Saved to browser Local Storage — persists across sessions.
6. Basic Eng Tab — Various Calculators
Nine independent engineering cards, each with its own inputs and its own result — nothing here feeds anything else. A jump-link strip sits directly under the tab bar so you can reach any card in one click. The four cards that need more than a paragraph get their own sections below (7–10).
7. Basic Eng — Gas Property Estimator
Enter gas gravity (air = 1.0), pressure, temperature and the isentropic exponent k (default 1.3, matching the PRV card). The card returns four values at once:
Papay (1968) with Standing-Katz pseudo-criticals. The same routine the Z-Factor Estimator uses, so both cards always agree.
Lee, Gonzalez & Eakin (1966). Density comes from ρ = PM/(ZRT), so the viscosity inherits Z.
√(k·Z·R·T/M). Useful for flare and relief headers, where a Mach limit usually governs before pressure drop does.
Temperature change per unit pressure drop across a restriction. Positive means the gas cools as it expands — the normal case for natural gas, and the reason valve outlets can reach hydrate or brittle-fracture territory.
Read the SCREENING label seriously. These are not standards. Papay and Lee-Gonzalez-Eakin are published correlations; the sonic-velocity expression is an engineering approximation (the form API 520 uses for critical flow); and the Joule-Thomson coefficient has no single governing document at all. The card warns — it never clamps — when you leave the Papay envelope or the correlation's 100–340 °F / 100–8,000 psia experimental basis. Lee-Gonzalez-Eakin carries no H₂S/CO₂/N₂ correction, so treat sour or high-inert gas as indicative only.
8. Basic Eng — Steam Properties (IAPWS-IF97)
Enter absolute pressure and temperature and the card classifies the state on the IAPWS-IF97 industrial formulation — the same one behind commercial steam tables — then returns the full property set: density, enthalpy, entropy, heat capacity and sonic velocity, plus the saturation line at your pressure (Tsat, hf, hg, hfg). Everything runs in your browser; nothing is sent to a server.
Pressure above the saturation pressure at your temperature: boiler feedwater, condensate, BFW pump suction. The state box shows how many kelvin of subcooling you have in hand.
Pressure below saturation: superheater outlets, turbine inlets, relief flows. The state box shows the degrees of superheat — the number API 520's KSH table wants.
Dense/near-critical water (roughly 350–590 °C above 16.5 MPa) uses a different, iterative formulation. The card says "not covered" rather than extrapolating into it.
PRV steam mode gets the same physics. In the Safety tab, entering a relieving pressure in steam mode now shows the IF97 saturation temperature right under the inputs, and an optional advisory-only steam temperature tells you whether your steam is superheated and KSH needs to come from API 520 Table 9 — or saturated, where KSH stays 1.0. The advisory never touches the sizing calculation itself.
Valid from 0 to 800 °C and up to 100 MPa (Regions 1 and 2). All 259 coefficients are transcribed verbatim from the IAPWS Release and re-verified in CI against the Release's own verification tables to 9 significant figures — see Theory Part II for the equations and the verification contract.
9. Basic Eng — Pump Suction NPSHa
Five inputs build the classic suction head balance: surface pressure (absolute — 101.325 kPa for an open tank), vapor pressure of the liquid at pumping temperature, density, static head z (liquid level above the impeller centerline — enter it negative for a suction lift), and suction friction h_f (the ΔP card's frictional term is a good source). The result is NPSHa in metres or feet, with the three terms shown separately so the balance can be audited at a glance.
Pumping water? Type the temperature and press FILL Pv & ρ FROM IF97 (WATER) — the card writes the saturated-water vapor pressure and density from the same IAPWS-IF97 functions as the Steam Properties card, then recalculates. The two fields stay ordinary inputs: overwrite them freely for any other liquid.
No verdict, on purpose. The card reports the physical number and stops. How much NPSHa you need above the pump's NPSHr — and the margin on top of that — comes from the vendor's tested curve and the Hydraulic Institute / API 610 guidance, which are licensed documents this application does not reproduce. A big positive number is not automatically safe; hot water near saturation erodes the margin quickly, which is exactly what the two warnings watch for.
10. Basic Eng — Compressor Head & Power
Six inputs describe the duty: gas SG (air = 1), suction pressure and temperature, discharge pressure (both pressures absolute), k = Cp/Cv, and the polytropic efficiency η_p in percent — vendors quote it; 70–85 % is typical for centrifugal machines. Add a mass flow if you want power: without it the head results still appear and the power box stays blank.
The card returns the polytropic head (what the vendor's curve is drawn in), the isentropic head (the ideal-path benchmark), the discharge temperature along the polytropic path, and the gas power ṁ·H_p/η_p — the enthalpy rise rate, before bearings, seals and gearbox. The Z box shows the real-gas correction at work: Z₁ at suction, Z₂ at the estimated discharge state, and the Z_avg the heads actually use, all from the same Papay correlation as the Z-Factor card. The path line reads back the pressure ratio, the polytropic exponent n, and the isentropic efficiency implied by your η_p.
Screening, not a performance test. These are the textbook isentropic and perfect-gas polytropic relations with a one-pass Papay Z correction — first principles, deliberately not a test-code procedure (ASME PTC 10 is a licensed standard and is not reproduced here). The discharge temperature is the perfect-gas path value; real-gas behaviour and any cooling shift the actual number. For a machine selection or an acceptance test, use the vendor's tools and the vendor's efficiency.
11. Advanced › Gas Quality & LNG Cargo (Section 1) — Gas Composition Input
This one block feeds everything else on the Advanced tab. Heating value, specific gravity, Wobbe index, molecular weight, MCP, the LNG liquid density and every mass ↔ volume flow conversion further down the page are all derived from the fourteen numbers you enter here — nothing on the tab has an independent input. Two settings decide how those numbers are read: the Input Mode selector in the card header, and the Reference Composition pull-down at the top of the block.
Source: GIIGNL Information Paper No.1 'Basic Properties of LNG', Table 1 — Examples of LNG composition (data: 2018 GIIGNL Annual Report). Mole %. The table's lumped C4+ column is split 50/50 iC4/nC4 — an assumption … This row is published as summing to 100.01% and is reproduced verbatim rather than rescaled; the calculator normalizes it.
Nine source-cited entries. Choosing one fills all fourteen boxes at once and sets the Input Mode to the basis that entry is reported on. The badge tells you what kind of number you just loaded — PUBLISHED, REFERENCE or ASSUMED — and the citation line underneath states exactly where it came from. Section 12 covers the three tiers and the cross-check chips in full.
The single most common mistake on this card. A gas analysis reported on one basis and entered on the other shifts the heating value by roughly 0.06 MJ/Nm³ — small enough to look plausible, large enough to fail a cross-check. Set it to match your certificate. In Vol % mode the panel below the conditions shows the calculated mole fractions, and in Mol % mode it shows the calculated volume fractions, so you can always see both.
CH₄, C₂H₆, C₃H₈, iC₄, nC₄, iC₅, nC₅, nC₆+, C₂H₄, C₃H₆, H₂, CO₂, N₂, O₂. Type into any of them; results update on every keystroke, no button to press. Leave anything your analysis does not report at 0 — a blank box is read as zero. Heavier hydrocarbons are lumped into nC₆+ and evaluated as n-hexane. Editing any box by hand drops the Reference Composition selector back to Custom and removes the citation, so a source is never left attached to numbers it no longer describes.
Orange when the sum is within 0.01 of 100 %, red otherwise. Red is a warning, not a block: the fractions are always normalised to 100 % before any calculation runs, so an analysis totalling 99.7 % still gives a correct answer for its own proportions. A separate amber banner appears if you go more than 0.5 % off, or if any entry is negative.
12. LNG Reference Compositions
The Advanced tab's Gas Composition block opens with a Reference Composition pull-down. Choosing an entry fills all fourteen component boxes at once and switches the Input Mode to the basis that entry is reported on — so you can go from a blank screen to a full LNG property set in one click. Every entry states where its numbers came from, directly under the selector.
Entries carry one of three badges, and the badge is the important part:
- PUBLISHED — five LNG origins transcribed from GIIGNL Information Paper No. 1, Basic Properties of LNG, Table 1 (data: 2018 GIIGNL Annual Report): Australia NWS, Malaysia Bintulu, Nigeria, Qatar and Trinidad. Reported as mole %.
- REFERENCE — this calculator's own JIS K 2301 worked example, reported as volume %.
- ASSUMED — three indicative archetypes (lean U.S. Gulf Coast pipeline feed, rich associated gas, nitrogen-rich feed). These are engineering assumptions, not project data. Each states the numbers it assumed so you can audit or overwrite them.
Two details worth knowing. GIIGNL reports butanes as a single lumped C4+ column, so the split into iC₄/nC₄ is an assumption — 50/50 here; sweeping it across the full range moves HHV and Wobbe index by no more than 0.03 MJ/Nm³, which CI re-measures on every build. And the published Qatar row sums to 100.01 %, which is reproduced verbatim rather than quietly rescaled; the calculator normalizes it before use, exactly as it does for any off-100 total.
Editing any component drops the selector back to Custom and hides the citation — a source is never left attached to numbers it no longer describes. The selection travels in share links and is restored with the rest of your inputs. For real work, always substitute the cargo certificate or contract specification: cargo compositions vary shipment to shipment, and none of these entries is a custody-transfer document.
13. Advanced › Gas Quality & LNG Cargo (Section 2) — Operating Conditions
Override standard conditions (0°C, 0 MPaG) to compute actual volumetric flows at operating pressure and temperature. LNG liquid temp is only used for liquid density calculation.
14. Advanced › Gas Quality & LNG Cargo (Section 3) — Physical & Combustion Properties
All outputs computed per JIS K 2301 (2011) using Z-corrected mole fractions. Use the HHV / LHV toggle inside the Section 3 header row to switch between Higher (総発熱量) and Lower (真発熱量) Heating Value modes. Results update automatically. Note: Wobbe Index always uses HHV per JIS K 2301 §7 regardless of mode.
15. Advanced › Gas Quality & LNG Cargo (Section 4) — Mass & Volumetric Flow Conversions
Two independent sub-panels: (A) Given Mass → Calculated Output and (B) Given Volume → Calculated Output. Each output panel has a toggle button (VOL/MOL for panel A; MASS/MOL for panel B) to switch between the standard flow output and Molar Flow. Available units: Mass: kg/s, kg/h, ton/h, ton/d, lb/h, lb/d. Molar: kmol/h or mol/s.
100 ton/h → 122.056 kNm³/h | 100 kNm³/h → 81.930 ton/h (matches JIS K 2301 / Excel exactly)
16. Advanced › Gas Quality & LNG Cargo — LNG Cargo Estimator
Sits directly under the compositional card and answers the procurement question its outputs make possible: how much LNG is on this ship? Pick a vessel from a curated catalogue (or type a tank capacity), set the loading limit, and the card returns the cargo as liquid volume, mass, standard gas volume and energy — TBtu on the HHV basis used in custody transfer, with LHV alongside. It never recomputes anything itself: ρliq (ISO 6578 at the LNG temperature set in Section 13), ρstd, HHV and LHV all arrive live from the composition above, so changing a preset, a component or the LNG temperature re-prices the cargo instantly.
The list now holds the full active fleet — all 804 LNG carriers at end-2025 — reproduced from Appendix 3 of the IGU World LNG Report 2026 (data: Rystad Energy) with the International Gas Union’s written permission, with attribution and a link to the report on every fleet entry. 36 featured vessels — every class, all four containment systems, all nine propulsion families, from LNG Aquarius (1977) to 2025 deliveries — additionally carry their own public primary source (an owner’s fleet page, a shipbuilder release, a class-society register or Wikipedia) and a photograph or schematic; where a featured row’s primary source and the report disagree on capacity, the primary source’s figure is shown. Every vessel links out to MarineTraffic and VesselFinder by IMO. Filter with the pills, or expand the catalogue for the grid. And you no longer have to scroll 804 rows: type a name, IMO number, owner, builder, propulsion or year into the Find box and both the pull-down and the catalogue narrow as you type; Enter picks the ship as soon as only one is left.
Loaded = capacity × loading limit. The IGC Code Ch. 15 filling limit is 98 % unless a higher limit is approved for the ship — many carriers load to 98.5 %, so use the vessel's certificate; the card warns above 98.5 %. Delivered = loaded − heel retained − boil-off rate × voyage days (a linear screening model: BOG is taken as a constant fraction of the loaded volume per day; real boil-off depends on tank pressure, ambient and whether it is burnt as fuel). The heel defaults to 3,000 m³ (a typical figure for a large carrier), so the delivered column is shown from the start; set heel and days to zero and only loaded quantities are shown.
Mass = volume × ρliq (Klosek-McKinley at the LNG temperature you set — colder LNG is denser, so the same tank holds more). Standard volume = mass ÷ ρstd, shown in kNm³ or MMNm³ (0 °C) and MMscf or Bscf (60 °F, via the mandated 37.3258). Energy = standard volume × HHV, in TBtu (10⁶ MMBtu; 1 MMBtu = 1.055056 GJ), MMBtu, GJ, TJ, MWh or GWh; the LHV figure is for reference. The energy density chip (MMBtu per tonne) and the expansion ratio (Nm³ of gas per m³ of liquid, at 0 °C — the ≈600 often quoted is at 15 °C) are the two numbers traders and terminal operators sanity-check first.
If the GT Fuel tab holds a plant, a green line shows how many such cargoes would fuel it for a year (from that tab's yearly fuel energy — see Section 20). Vessel, capacity, limits and unit selections travel in Share links and are restored. Thumbnails are Wikimedia Commons photographs (author and licence shown under each; full list in assets/vessels/CREDITS.json) or an original schematic by containment type — no MarineTraffic / VesselFinder / ShipSpotting imagery is used. Every copy button is unit-aware.
17. Advanced › Hydraulics — Pipe Delta Pressure Calculator (Darcy-Weisbach)
Computes the pressure drop along a pipe for single-phase (vapor or liquid) and two-phase flow. The iterative Colebrook–White friction factor and the full Darcy–Weisbach solution run server-side (Python / Vercel API). Since v3.7 the two-phase frictional term is computed by a selectable correlation — HEM (Homogeneous Equilibrium Model, default), Lockhart–Martinelli, Müller-Steinhagen–Heck or Friedel — chosen with the TWO-PHASE METHOD selector; static head and fittings always stay on the homogeneous (no-slip) basis. Enter the pipe geometry and per-phase properties, then click CALCULATE PRESSURE DROP. The phase regime — Single-Phase Vapor, Single-Phase Liquid, or Two-Phase with the active method — is auto-detected from which mass flows are non-zero and shown as a badge.
18. ΔP Card — Fittings, Valves & Line Sizing
Two additions sit either side of the CALCULATE button. Both are optional and neither changes an existing calculation unless you use it.
2C. Fittings & valves — expand the block and enter how many of each fitting the run contains. The card looks up Crane TP-410 resistance coefficients (K = n·fT, where fT is the fully-turbulent friction factor for the nearest nominal size — shown beneath the grid so you can audit it) and sums them into ΣK. The pressure drop gains ΣK·ρv²/2, reported as its own output alongside the equivalent straight length Leq. Every count starts at 0, so an untouched block leaves the result byte-identical to v2.7.
Line sizing — choose the service (or leave it on Auto, which follows the detected phase) and the card compares your result with NORSOK P-001 criteria, showing vmax or the ΔP limit, the ratio, and a WITHIN LIMIT / NEAR LIMIT / EXCEEDS badge. Liquid services also flag velocities below the 1 m/s minimum, where solids can settle out.
Choosing a service whose phase does not match the calculated one shows PHASE MISMATCH rather than a verdict. The reported velocity for two-phase flow is the homogeneous mixture velocity, and judging it against a single-phase criterion would give a confident but wrong answer. Crane TP-410 is an industry technical paper, not an ISO or API standard; gradual reducers are deliberately omitted because their coefficients could not be verified against a primary source.
19. Advanced › Hydraulics — Flow Regime Visualizer
Located directly beneath the Pipe Delta Pressure Calculator and driven by the same inputs. Click VISUALIZE FLOW REGIME to classify the two-phase flow pattern on a regime map rendered server-side (Python seaborn / Vercel API), followed by a conceptual 3D animation (Three.js) of how the flow would appear inside the pipe. Layer depths, film thicknesses and bubble loading scale with the no-slip holdup λL, and speeds and inclination follow jG, jL and θ. Since v3.7 the animation has a mode strip: View (Exterior / Cutaway / Inside the pipe), a Regime override that previews any of the eight patterns under the calculated conditions (flagged PREVIEW), playback Speed (0.25×–2×) and pause — and the canvas can be orbited by dragging and zoomed by scrolling.
Example output for the default inputs — the seaborn regime map (top) and a frame of the Three.js 3D animation (bottom):
20. Advanced › GT Fuel — Turbine Selection & Fuel Gas Estimator
| Period | Volume | Mass |
| Hourly | 91.07 kNm³ | 74.6 t |
| Monthly ×92% | 61.16 MMNm³ | 50,109 t |
| Yearly ×92% | 733.93 MMNm³ | 601,308 t |
The GT Fuel tab holds three panels. In Gas Turbine Selection, pick a vendor and model from the pull-down (31 machines: MHI including FT-series aeroderivatives, GE Vernova, Siemens Energy) — or scroll to the Gas Turbine Catalogue below and click Use this model on any card. The selection pre-fills the estimator with the machine's ISO rating and %-LHV efficiency and shows its spec chips (rating, η, heat rate, TIT and exhaust where published). The Cycle selector switches between simple-cycle GT and GTCC plant basis: in Combined cycle mode the published plant output and plant efficiency are used — the fuel still burns only in the gas turbine, so fuel = plant output ÷ plant efficiency. Choose Manual input as vendor to type everything yourself.
The Fuel Gas Rate Estimator needs five numbers: power output (MW/kW), efficiency (%-LHV), fuel heating value (MJ/Nm³ or Btu/scf; if you set the basis to HHV, an HHV/LHV ratio field appears — default 1.108, the reference-composition value), standard gas density, and availability. The dashed Import button pulls LHV/HHV and ρ_std from the last Advanced-tab compositional run; the Send button pushes the computed mass flow and density into the Basic Eng Mass↔Vol card and jumps there. Outputs update live: fuel energy input (MW-th), heat rate in kJ/kWh and Btu/kWh, volumetric flow (kNm³/h · Nm³/h · scf/h · MMSCFD) and mass flow (t/h · kg/h · kg/s), each with a unit-aware copy button, plus a totals table — the hourly row is the instantaneous rate, while daily, monthly (730 h) and yearly (8,760 h) rows are multiplied by the availability factor.
Screening, not a guarantee. Catalogue figures are indicative vendor data at ISO conditions; real fuel burn moves with ambient conditions, degradation, part load and fuel composition. Every dataset entry carries its source, and CI re-verifies that each machine's published heat rate equals 3600/η within rounding.
The consumption-totals table now has a unit select on each column header — Fuel volume in Nm³ (auto kNm³/MMNm³) or scf (auto MMscf/Bscf), a new Fuel energy in column (Q × hours) and Power sent out each in MWh/GWh, GJ/TJ/PJ or MMBtu/TBtu — and the fuel-energy tile switches between MW-th, MMBtu/h and GJ/h. Fuel-rate selects gain kNm³/d, MMscf/h, t/d and lb/h. Defaults are unchanged. Gas density accepts lb/scf: because ρstd is mass per standard volume, the factor is 0.45359237 × 37.3258 = 16.9307 kg/Nm³ per lb/scf — not the 16.0185 that converts an actual-condition lb/ft³ elsewhere in the app; 0.8193 kg/Nm³ ↔ 0.04839 lb/scf. The yearly fuel energy in TBtu (28.00 for the M701JAC example) is what the LNG Cargo card in Section 16 compares its cargoes against.
21. Safety Tab — PRV Sizing (API 520 Part I)
Sizes Pressure Relief Valves (PRVs) per API Standard 520 Part I §5.6–§5.10 and Annex C. The card opens pre-filled with a complete gas/vapor case, so a single click on CALCULATE ORIFICE SIZE returns a result (orifice H) before you type anything — then replace the figures with your own. Select the sizing mode and unit system, enter the process inputs, and read the required effective discharge area and the recommended API 526 orifice letter (D–T). Every input with a physical unit has its own unit drop-down — mass flow W (lb/h · kg/h · t/h · kg/s · lb/s), temperature T (°R · K · °F · °C), flow rate Q (gal/min · L/min · m³/h · bbl/h), specific volume vo/v9 (ft³/lb · m³/kg · L/kg) and viscosity µ (cP · mPa·s · Pa·s) — and the card converts each to the unit system's basis before sizing. If a saved session left a panel incomplete, it falls back to the default case on reload, so the one-click example is always there. Every pressure field (P1, P2, Ps, Po, Pa) has its own unit drop-down (psi · kPa · bar · MPa · Pa · atm · kg/cm²), an Abs/Gauge toggle and a ⇩ button that imports the figure entered in the General tab's Pressure card; the card converts to the absolute basis API 520 expects (gauge for the §5.8/§5.9 liquid modes) before sizing, so units and bases may be mixed freely. The USC/SI switch sets the output area unit and re-expresses every field sitting on a canonical unit (psi↔kPa, lb/h↔kg/h, °R↔K, gal/min↔L/min, ft³/lb↔m³/kg) in the other system, converting the figures — so the default case gives orifice H in USC and SI alike. ↺ LOAD EXAMPLE in the card header restores the shipped case in all modes at any time, expressed in whichever unit system is selected (kg/h · K · kPa under SI).
22. Report Tab — Bug Reports & Feature Requests
Select report type, enter your name and description, then click GENERATE EMAIL TO DEVELOPER. Your default email client will open with a pre-filled message addressed to the developer.
Appendix — Release Notes
What changed in each release, newest first. Everything listed here is documented properly in the numbered sections above — this appendix is the history, not the manual. Click a release to expand it.
★ New in Version 3.9
★ New in Version 3.8
★ New in Version 3.7
★ New in Version 3.6
★ New in Version 3.5
★ New in Version 3.4
★ New in Version 3.3
★ New in Version 3.2
★ New in Version 3.1
★ New in Version 3.0
★ New in Version 2.8
★ New in Version 2.5
★ New in Version 2.4
Theory & Specifications Manual
Complete technical reference for all formulas, standards, coefficients and algorithmic logic implemented in v3.9. Primary reference: JIS K 2301:2011 and ISO 6578:1991.
Part I — Real-Gas Properties (Basic Eng Tab)
1.1 Z-Factor: Papay Method (Standing-Katz Pseudo-Criticals)
1.2 Gas Viscosity — Lee, Gonzalez & Eakin (1966)
A correlation fitted to four natural gases over 100–340 °F and 100–8,000 psia, with a reported average absolute error near 2.7 %. Units are not negotiable: T in °R, M in lb/lbmol, ρ in g/cm³, μ in cP.
These are the original unrounded coefficients. A rounded variant (9.4 / 0.02 / 209 / 19) circulates widely and shifts μ by about −2.1 % — the same order as the correlation's own error, so it destroys reproducibility for no benefit. Two further variants on wiki sites are simply corrupt (X = 3.488, and 0.001·M in place of 0.01·M). The density term inherits Papay's Z, so the viscosity is only as good as the Z estimate beneath it.
1.3 Sonic Velocity
The speed of a pressure wave in a real gas, in the form API 520 Part I uses for critical-flow PRV sizing:
This is an engineering approximation, not an identity. The exact definition is c² = (∂P/∂ρ)s; the expression above neglects the (∂Z/∂P)T and (∂Z/∂T)P contributions. It is defensible because it is the form the sizing standard itself adopts, but it should not be treated as rigorous acoustics.
1.4 Joule-Thomson Coefficient
The temperature change per unit pressure drop in an isenthalpic expansion — across a control valve, an orifice or a choke. Starting from the exact definition and substituting V = ZRT/P:
The Z terms cancel exactly in that substitution, so no approximation enters at this step. Everything uncertain lives in two places: Papay's Z, and the ideal-gas heat capacity. ∂Z/∂T is evaluated by central difference on the same Papay expression, with a fixed step of 0.5 °R — small enough to track the derivative, large enough to stay clear of floating-point cancellation. A positive coefficient means the gas cools on expansion, which is the normal natural-gas case and the reason valve outlets can drop into hydrate-formation or low-temperature-embrittlement territory.
Part II — Steam Properties: IAPWS-IF97
2.1 What IF97 is, and which slice is implemented
The IAPWS Industrial Formulation 1997 (Revised Release, 2007) is the international standard property formulation for water and steam — the equations behind essentially every commercial steam table and heat-balance program. It divides the p–T plane into five regions. This application implements the three that cover everyday plant conditions: Region 1 (compressed liquid), Region 2 (superheated and supercritical-pressure steam up to 800 °C), and Region 4 (the saturation line), plus the B23 boundary equation that separates Region 2 from Region 3. Region 3 (dense/near-critical) is a Helmholtz formulation requiring density iteration, and Region 5 (> 800 °C flue-gas territory) has no business in a process tool — both are declared out of scope rather than approximated. R = 0.461526 kJ/(kg·K) throughout.
2.2 Regions 1 and 2 — dimensionless Gibbs energy
Both regions express the specific Gibbs free energy in dimensionless form γ = g/(RT) and obtain every property from its partial derivatives — one consistent equation per region, no auxiliary correlations:
Region 1 is valid 273.15–623.15 K up to 100 MPa; Region 2 covers 273.15–1073.15 K up to 100 MPa on the steam side of the saturation line and B23. Classification: below 623.15 K the input pressure is compared with psat(T) — above it is liquid, below it is steam; from 623.15 to 863.15 K the quadratic B23 equation p = n₁ + n₂T + n₃T² decides between Region 2 and the unsupported Region 3.
2.3 Region 4 — the saturation line, exactly
The saturation equation is an implicit quadratic in both β = (psat/1 MPa)1/4 and ϑ ≈ Tsat, which is why it solves in closed form in both directions — no iteration, no lookup error:
Valid from the triple point (611.657 Pa) to the critical point (647.096 K, 22.064 MPa). Saturated-liquid and saturated-vapor enthalpies are obtained as hf = hR1(p, Tsat) and hg = hR2(p, Tsat) — the industrial formulation is constructed so Regions 1 and 2 meet the saturation line consistently. Above psat(623.15 K) = 16.529 MPa the liquid side of the line belongs to Region 3, so hf/hg are shown only up to that pressure.
2.4 The verification contract — how 259 coefficients stay honest
The coefficients live in index.html as a single JSON literal, and the test suite extracts that exact literal — not a copy — re-implements the equations independently in Python, and reproduces the Release's own computer-program verification tables: Table 5 (Region 1), Table 15 (Region 2), Tables 35/36 (saturation) and the B23 check point, every value to 9 significant figures. A single mistyped digit anywhere in the 259 coefficients fails CI. The browser implementation is then verified by hand against the same tables. This two-sided check is the same discipline the JIS K 2301 chain uses, applied to a much larger constant set.
2.5 Unit conversions and the PRV advisory
Imperial outputs use the International Table definitions, which are exact by construction: 1 Btu/lb = 2.326 kJ/kg and 1 Btu/(lb·°R) = 4.1868 kJ/(kg·K); density uses 1 lb/ft³ = 16.0184634 kg/m³, the same factor as the RP 14E erosional-velocity constant. In the Safety tab, the PRV steam mode shows Tsat(P1) from Region 4 and — if the optional advisory temperature is filled in — the degrees of superheat, which is the entry key into API 520's Table 9 for KSH. The advisory is display-only: the sizing payload sent to the API is unchanged, and KSH remains a manual input precisely because Table 9 is API-copyrighted material this application will not reproduce.
Reference: Revised Release on the IAPWS Industrial Formulation 1997 for the Thermodynamic Properties of Water and Steam, IAPWS, August 2007 — freely available at iapws.org. Region/table numbers above refer to that document.
Part III — Pump Suction: NPSHa
3.1 The head balance
Net Positive Suction Head available is the absolute head at the impeller centerline in excess of the liquid's vapor head — the budget that keeps the liquid liquid as it accelerates into the eye. It is plain hydrostatics, not the property of any standard:
Each term has a physical reading: the pressure head is what the vessel pushes with, beyond what the vapor pressure claws back; z is geometry (negative when the pump sits above the liquid — a suction lift); h_f is what the suction line wastes. The card shows all three so a low result can be traced to its cause. Two warnings mark the physical cliff edges: Psurface ≤ Pv (the vessel itself is boiling) and NPSHa ≤ 0 (the liquid flashes before the impeller — cavitation is certain, operation impossible).
3.2 The water helper — IF97 again
For water, the two hardest inputs come free from the formulation already in this application: Pv = psat(T) from Region 4 (Eq. 30, exact closed form) and ρ = ρf(T), the saturated-liquid density, from Region 1 evaluated on the saturation line. Saturated-liquid density is the right choice for suction screening: the liquid at the free surface of a vessel at its bubble point is saturated, and for subcooled or pressurized cases the difference is far below the accuracy any NPSH calculation needs (water is nearly incompressible). Values are written into the fields at 8 significant figures — visible, auditable, and overridable — rather than hidden inside the calculation. Valid 0–350 °C; the helper refuses beyond, like every IF97 surface in this app.
3.3 Why there is no verdict badge
Other cards here judge their result against a published criterion (RP 14E, NORSOK P-001). NPSH cannot be judged that way from the suction side alone: the requirement, NPSHr, is a tested property of the specific pump at the specific flow — it exists on the vendor's curve, nowhere else — and the margin to apply on top of it (commonly 0.6–1.5 m or a 1.1–1.35 ratio, rising steeply for hot water and hydrocarbons near their bubble point) is codified in Hydraulic Institute 9.6.1 and API 610, both licensed documents. Reprinting half-remembered margin rules would manufacture false authority; this application cites what it can verify and stops at the physics.
Part IV — Compressor Head & Power
4.1 The two ideal paths
Compression work per unit mass is ∫v·dP along the machine's thermodynamic path. For a perfect gas on an isentropic path (P·vk constant) the integral closes to the textbook result; a real machine follows a steeper path, modeled as polytropic (P·vn constant), where the polytropic efficiency ηp is defined as the differential efficiency held constant along the path — which fixes the exponent:
Head is path work only — it is what the impeller must impart per kilogram, which is why vendor curves are drawn in polytropic head: unlike the isentropic value it is (to first order) independent of what the gas happens to be. Both formulas are first-principles results found in any thermodynamics text and in the free Turbomachinery Laboratory tutorials; no test-code material is used or needed for them.
4.2 Real-gas correction — Papay again, one pass
The perfect-gas integral acquires a real-gas correction through Z: this card uses Z_avg = (Z₁ + Z₂)/2, the standard screening treatment, with both values from the same Papay/Standing-Katz correlation as the Z-Factor and Gas Property cards (validity 0 < Pr ≤ 15, 1.05 ≤ Tr ≤ 3.0 — the card warns per state when either end is outside). Z₂ needs a discharge temperature before the head is known, so it is evaluated once at the perfect-gas estimate T₂ = T₁·rm — a deterministic single pass, deliberately not an iteration loop: the residual error in Z_avg from that estimate is second-order and far inside the correlation's own tolerance. The displayed T₂ is that same perfect-gas path value. The full real-gas polytropic machinery (Schultz-type corrections, test-code analysis per ASME PTC 10) belongs to vendor software and acceptance tests; it is licensed material and this card neither uses nor reproduces it.
4.3 Efficiencies, power, and units
The two heads imply each other's efficiency: actual work is w = H_poly/ηp, so the isentropic efficiency the same machine exhibits is η_s = η_p · H_is/H_poly — always below ηp in compression, because the polytropic path reheats the gas and makes each successive pressure slice more expensive. Gas power is W = ṁ·H_poly/η_p, the enthalpy rise rate of the gas; bearings, seals and gearbox losses add on top, per the vendor. Unit factors are exact by definition: 1 ft·lbf/lbm = 9.80665 × 0.3048 = 2.98906692 J/kg (display literal 334.55256 per kJ/kg), and 1 hp = 550 ft·lbf/s = 745.69987 W (display literal 1.3410221 per kW).
Part V — Gas Compositional Analysis (JIS K 2301:2011)
5.1 Component Physical Constants (JIS K 2301 Table)
All component constants are sourced directly from JIS K 2301:2011. The compressibility factor Zᵢ and √bᵢ are key to Z-correction; Hᵢ is gross calorific value at 0°C, 101.325 kPa.
| Component | MW (g/mol) | Zᵢ | HHV (MJ/Nm³) | LHV (MJ/Nm³) | √bᵢ | Sᵢ (Air=1) | Flame Spd (cm/s) |
|---|---|---|---|---|---|---|---|
| CH₄ | 16.043 | 0.9976 | 39.84 | 35.818 | 0.049 | 0.554 | 35.9 |
| C₂H₆ | 30.07 | 0.9900 | 69.79 | 63.76 | 0.100 | 1.039 | 41 |
| C₃H₈ | 44.097 | 0.9789 | 99.22 | 91.18 | 0.1453 | 1.523 | 41 |
| iC₄H₁₀ | 58.123 | 0.9580 | 128.23 | 118.18 | 0.2049 | 2.008 | 38 |
| nC₄H₁₀ | 58.123 | 0.9572 | 128.66 | 118.61 | 0.2069 | 2.008 | 38 |
| N₂ | 28.0135 | 0.9995 | 0 | 0 | 0.0224 | 0.968 | — |
| CO₂ | 44.01 | 0.9933 | 0 | 0 | 0.0819 | 1.520 | — |
| H₂ | 2.016 | 1.0006 | 12.788 | 10.777 | 0 | 0.0696 | 282 |
5.2 Volumetric → Mole Fraction Conversion (JIS K 2301 §7)
5.3 Sample Gas Compressibility Factor (Z)
5.4 Gross Calorific Value / HHV
5.5 Specific Gravity & Wobbe Index
5.6 Maximum Combustion Point (MCP)
5.7 Lower Heating Value (LHV) — 真発熱量 (JIS K 2301 Table 30)
Part VI — LNG Reference Compositions
6.1 Provenance and the three tiers
The preset dataset separates what is published from what is assumed, because the two carry entirely different weight in an engineering argument. Five entries are transcribed from GIIGNL Information Paper No. 1, Basic Properties of LNG, Table 1 — Examples of LNG composition, whose data is credited to the 2018 GIIGNL Annual Report. One entry is this application's own JIS K 2301 reference case. Three are engineering archetypes carrying no external citation at all, and are badged ASSUMED so they can never be mistaken for measured data. The tier boundary is enforced in CI: an assumed entry is forbidden from declaring a published calorific value, because a cross-check against an invented number would dress up a guess as a citation.
6.2 Fraction basis — why these load as mole %
GIIGNL Table 1 does not state its basis in the column headers. It can be recovered from the table itself: running each composition through the JIS K 2301 chain on a mole basis reproduces the published gross calorific values to within 0.02 MJ/Nm³, whereas a volume basis is systematically 0.06 MJ/Nm³ high across all five rows — a bias far larger than the scatter, and in one direction. The table is therefore mole %, and each preset carries its own basis field which is applied to the Input Mode selector on load. The application's own reference case remains volume %, as JIS K 2301 defines it.
6.3 The lumped C4+ column
GIIGNL reports butanes as one lumped C4+ column, while JIS K 2301 needs iC₄ and nC₄ separately — the two differ slightly in gross calorific value (128.23 vs 128.66 MJ/Nm³), compressibility and √b. The split is therefore an assumption, taken here as 50/50. Its consequence is bounded rather than asserted: sweeping the butane pool across the entire range, from all-iC₄ to all-nC₄, moves HHV and Wobbe index by no more than 0.03 MJ/Nm³ (worst case: Bintulu's Wobbe index). That is inside the ±0.05 MJ/Nm³ band the cross-check chips use, so the split alone can never flip a preset's own verification. CI re-measures this sweep on every build and fails if the spread grows, and separately asserts that the figure quoted in the on-screen citation is the same figure the suite enforces.
6.4 Cross-check as a verification vector
Because GIIGNL publishes a gross calorific value and Wobbe index beside each composition, every published preset doubles as an external verification vector for the JIS K 2301 implementation — five reference points that were not derived from this application and cannot drift with it. The comparison always uses the gross value hhv_mix, never the toggled display value, so switching the panel between HHV and LHV cannot disturb a comparison against a published figure that never moved.
Australia NWS 45.32 / 45.32, WI 56.52 / 56.53 · Malaysia Bintulu 43.69 / 43.67, WI 55.58 / 55.59 · Nigeria 43.41 / 43.41, WI 55.49 / 55.50 · Qatar 43.43 / 43.43, WI 55.38 / 55.40 · Trinidad 41.05 / 41.05, WI 54.23 / 54.23. Largest deviation across all ten comparisons: 0.02 MJ/Nm³.
6.5 Fidelity to the source
Published rows are reproduced exactly as printed, including the Qatar row's total of 100.01 %. Silently rescaling it to 100 would make the on-screen numbers disagree with the cited table and break the traceability the citation exists to provide; instead the existing normalization path handles it, as it does for any composition that does not sum to 100. For the same reason a manual edit of any component immediately clears the attribution — the source line describes the dataset, and the moment the numbers stop being the dataset, the citation must go.
Scope. These are representative compositions for screening and teaching. Real cargo compositions vary shipment to shipment and are established by the custody-transfer analysis; nothing in this dataset is a substitute for a cargo certificate or a contract specification.
Part VII — Standard Gas Density & Flow Conversions
7.1 Standard Gas Density ρ_std
7.2 Mass ↔ Volume Flow Conversion
Part VIII — LNG Liquid Density (ISO 6578:1991 Klosek-McKinley)
8.1 Klosek-McKinley Method Overview
The ISO 6578 Klosek-McKinley method computes LNG liquid density by summing the molar volumes of pure-component liquids, then applying correction factors k₁ and k₂ for CH₄ and N₂ interactions.
8.2 Molar Volume Table (ISO 6578 Table B.2) — Selected Values [m³/kmol]
| Temp (K) | CH₄ | C₂H₆ | C₃H₈ | iC₄ | nC₄ | N₂ |
|---|---|---|---|---|---|---|
| 108 K | 0.037489 | 0.047513 | 0.062047 | 0.077842 | 0.076392 | 0.043571 |
| 110 K | 0.037739 | 0.047683 | 0.062220 | 0.078040 | 0.076582 | 0.044887 |
| 111.15 K (-162°C) | 0.037886 | 0.047776 | 0.062319 | 0.078153 | 0.076687 | 0.045660 |
| 112 K | 0.037995 | 0.047845 | 0.062392 | 0.078236 | 0.076765 | 0.046231 |
| 114 K | 0.038262 | 0.048014 | 0.062574 | 0.078438 | 0.076957 | 0.047602 |
| 116 K | 0.038539 | 0.048192 | 0.062765 | 0.078648 | 0.077157 | 0.049003 |
| 120 K | 0.039130 | 0.048576 | 0.063174 | 0.079094 | 0.077584 | 0.051906 |
Interpolated values computed by linear interpolation between adjacent temperature rows. Highlighted row shows the exact interpolated values at -162°C (111.15 K) from the Excel reference.
8.3 k₁, k₂ Correction Factors (ISO 6578 Table C)
| MW_mix (g/mol) | k₁ (×10⁻³) | k₂ (×10⁻³) |
|---|---|---|
| 16 | −0.0086 | −0.0204 |
| 17 | +0.204 | +0.374 |
| 18.305 (ref) | +0.465 | +0.732 |
| 19 | +0.580 | +0.878 |
| 20 | +0.776 | +1.058 |
Part IX — LNG Cargo Quantities
9.1 The chain
A cargo is bought and sold as energy, but a ship is filled by volume. The card converts one into the other with nothing but the quantities the compositional suite already produces:
Custody transfer is on the gross (HHV) basis — the GIIGNL LNG Custody Transfer Handbook's energy formula multiplies the volume unloaded, the density and the gross calorific value — so HHV is the headline figure and LHV is shown for reference. The card owns none of the physics: ρliq, ρstd, HHV and LHV are read from the composition card through one bridge object, so a change to any JIS or ISO 6578 rule propagates here automatically and the two cards can never disagree.
9.2 Units, and the one factor that is easy to get wrong
1 TBtu = 10⁶ MMBtu; 1 MMBtu = 1.055056 GJ (the same factor the General-tab energy module has carried since v1.3), so 1 TBtu = 1.055056 PJ = 293.07 GWh. Standard gas volume is offered on both bases the industry uses — Nm³ at 0 °C and scf at 60 °F — related by the mandated 37.3258 scf/Nm³ (Part VII), which is why the "expansion ratio" the card shows (Nm³ of gas per m³ of liquid) reads about 562 rather than the ≈600 quoted at 15 °C: same cargo, different reference temperature. Energy density in MMBtu per tonne is composition-driven and sits between ≈48 (lean, high-N₂) and ≈52 (rich Australian or Qatari) — a quick plausibility check on any cargo figure.
The GT Fuel tab's gas-density input now accepts lb/scf (v3.5). A standard density is mass per standard volume, and this application's standard bases are 0 °C for Nm³ and 60 °F for scf, so the conversion is 0.45359237 kg/lb × 37.3258 scf/Nm³ = 16.9307 kg/Nm³ per lb/scf — not the 16.0185 that converts an actual-condition lb/ft³ (same physical volume) on the Mass↔Vol and ΔP cards. Using the wrong one reads 5.7 % high: 0.8193 kg/Nm³ is 0.04839 lb/scf, and 16.0185 would call it 0.05115. The factor is derived and named in the code, and a test pins it.
9.3 Loading limit, heel and boil-off
The IGC Code (Ch. 15) sets the tank filling limit at 98 % at the reference temperature unless the Administration approves a higher limit for the specific ship and tank; many modern carriers are approved to 98.5 %, so the card defaults to 98.5 % and warns above it — the vessel's certificate governs. Heel is the liquid deliberately left on board so the tanks stay cold for the ballast voyage (typically a few thousand m³ on a large carrier). Boil-off is modelled as a constant fraction of the loaded volume per day (0.10 %/day is representative of a modern membrane ship; steam-era Moss carriers were nearer 0.15 %) multiplied by the voyage days — a linear screening approximation. Real boil-off varies with tank pressure, ambient conditions and cargo composition, and on most ships it is burnt as fuel or reliquefied rather than lost; treat "delivered" as an order-of-magnitude view, and use the discharge port's custody-transfer figures for anything commercial.
9.4 The vessel dataset — provenance by design
The catalogue holds the full active fleet — 804 vessels at end-2025 — of which 36 are featured entries chosen for coverage: every vessel type in the active fleet (Q-Max, Q-Flex, conventional, Arc7 icebreaker, FSRU, FSU, mid-scale, small-scale), all four containment systems (membrane, Moss spherical, SPB self-supporting prismatic, Type C), all nine propulsion families (steam, steam reheat, STaGE, DFDE, TFDE, SSD, X-DF, ME-GI, ME-GA) and a dozen shipyards. Each entry's particulars are confirmed against a public primary page — the owner's or operator's fleet list, a shipbuilder delivery release, a classification-society register or, as a last resort, an English Wikipedia article — and that page is the row's cited source in the app. The remaining 768 entries are reproduced from the report’s Appendix 3 table with the International Gas Union’s written permission (granted 24 August 2026), each carrying the attribution and a link to the report; where a featured vessel’s primary source and the report disagree on capacity, the featured row keeps its primary source’s figure. Nothing is scraped from AIS platforms; the "View on MarineTraffic / VesselFinder" links are ordinary hyperlinks by IMO number, opened in a new tab.
Photographs come from Wikimedia Commons only, each under CC BY-SA or public domain, downloaded once as Commons-rendered thumbnails and committed with a credits file that the card renders under every image; ShareAlike attaches to the photograph, not to the surrounding MIT-licensed application (Creative Commons' own "collection" interpretation). Where no free-licensed photograph of a ship exists, an original schematic silhouette by containment type stands in. Spotter-site photographs (MarineTraffic, VesselFinder, ShipSpotting) are photographer-owned with no licence to third parties and are not used or hot-linked.
9.5 Worked example (Vector 13)
Vloaded = 174,000 × 0.985 = 171,390 m³ · m = 171,390 × 467.3168 / 1000 = 80,093 t · Vstd = 80,093,422 / 0.83108 = 96,372,923 Nm³ = 96,373 kNm³ = 96.373 MMNm³ = 3,597 MMscf · EHHV = 96,372,923 × 45.32 = 4,367,621 GJ = 4,368 TJ = 4.140 TBtu (÷ 1.055056 ÷ 10⁶) · ELHV = 3.739 TBtu · energy density = 4,139,705 MMBtu / 80,093 t = 51.69 MMBtu/t · expansion = 96,372,923 / 171,390 = 562 Nm³/m³.
Delivered with 3,000 m³ heel and 0.10 %/day × 15 days (2,571 m³ boil-off): 165,819 m³ → 77,490 t → 4.005 TBtu. GT cross-link: the M701JAC example burns 28.00 TBtu/yr, so ≈ 6.8 such cargoes a year.
Part X — Pipe Hydraulics (Advanced › Hydraulics)
10.1 Pressure Drop: Darcy-Weisbach + Colebrook-White Friction Factor
10.2 Fitting Losses — Crane TP-410
Crane expresses each valve and fitting as a resistance coefficient K, a fixed number of velocity heads:
Note that fT and f are different quantities. fT characterises the fitting and depends only on nominal size; f characterises the flow and depends on Reynolds number and roughness. The application applies the K method and reports Leq as derived information, because a fitting's loss is a property of its geometry while the equivalent length that produces that loss depends on the pipe's operating point. The alternative L/D shortcut (Leq = n·D) implicitly assumes f = fT — that the line is in fully-rough flow — which is false for most process piping.
10.3 Line-Sizing Criteria — NORSOK P-001
Screening limits for process piping, applied to the frictional pressure drop only. Static head is a real pressure change but not a sizing criterion — a vertical line is not undersized because it is tall.
| Service | Velocity | Frictional ΔP / 100 m |
|---|---|---|
| Gas / vapour | min(175·ρ−0.43, 60) m/s | 0.001–0.11 bar (<35 barg) · 0.11–0.27 (35–138) · P/500 (>138) |
| Liquid — carbon steel | 1–6 m/s | — |
| Liquid — stainless / titanium | 1–7 m/s | — |
| Liquid — copper-nickel | 1–3 m/s | — |
| Liquid — GRP | 1–6 m/s | — |
| Pump suction — sub-cooled / boiling | — | ≤ 0.25 / ≤ 0.05 bar |
| Pump discharge | — | ≤ 0.9 bar |
| Two-phase — non-corrosive / corrosive | min(183·ρm−0.5, 25 / 10) m/s | — |
Copper-nickel carries the lowest velocity band of the four liquid materials — 3 m/s, half the carbon-steel figure — because of its susceptibility to erosion-corrosion in seawater. The minimum of 1 m/s matters as much as the maximum: below it, solids and water can settle out of the stream.
Only NORSOK P-001 criteria are implemented. Velocity and ΔP tables attributed to other sources circulate widely but could not be verified against primary text, and this application does not cite a standard it has not checked. These limits screen a line size; they do not replace a hydraulic study, a surge analysis, or a vibration assessment.
10.4 Erosional Velocity Limit (API RP 14E)
10.5 Two-Phase Flow Regime Maps (Flow Regime Card)
10.6 Selectable Two-Phase Frictional Correlations (v3.7)
The TWO-PHASE METHOD selector on the ΔP card replaces the frictional term only. The static head stays ρm·g·Δz on the no-slip density, fittings stay ΣK·ρm·v²/2, and the reported velocity, Reynolds number and friction factor remain the homogeneous-mixture values — so the API RP 14E and NORSOK screens stay comparable across methods. All phase-alone and whole-flow friction factors reuse the Colebrook-White function at the actual pipe roughness (an engineering adaptation; the original correlations were fitted with smooth-pipe Blasius-type factors).
Part XI — Gas Turbine Fuel Estimation
11.1 Governing relations
A gas turbine converting fuel energy to electrical output at thermal efficiency η absorbs fuel energy Q_fuel = P / η. Vendor catalogue efficiencies are quoted on a lower-heating-value basis at ISO conditions (15 °C, 101.325 kPa, 60 % RH), per the ISO 3977-2 / ASME PTC 22 convention, so the heat rate follows directly as HR = 3600/η. The volumetric fuel demand at standard conditions and the mass flow are then closure arithmetic:
In GTCC plant mode the same relation runs on the published plant output and plant efficiency: all fuel is fired in the gas turbine (the steam cycle recovers exhaust heat), so plant fuel = plant output ÷ plant η. This is why a >64 % LHV combined-cycle plant burns ~31 % less gas per MWh than its 44 % simple-cycle turbine alone.
11.2 Heating-value basis and unit conversions
Turbine efficiencies are LHV-based, but commercial gas is often specified in HHV. When the estimator's basis selector is set to HHV, the input is divided by the HHV/LHV ratio (default 1.108, the app's reference composition 44.59/40.25; the Import button recomputes it from the live Advanced-tab composition). Heating-value units interconvert with the same constants as the General-tab converter: 1 Nm³ = 37.3258 scf (0 °C ↔ 60 °F standard-volume basis) and 1 Btu = 1,055.056 J, giving Btu/scf = MJ/Nm³ ÷ (0.001055056 × 37.3258) = MJ/Nm³ × 25.393. Standard-volume flows convert likewise: MMSCFD = Nm³/h × 37.3258 × 24 / 10⁶.
11.3 Dataset, consistency and scope
The embedded catalogue holds 31 machines with published ISO ratings: MHI heavy-duty, H-series industrial and FT-series aeroderivative units transcribed from the MHI GTCC brochure (METP-11GT01E1-E-0, performance tables); GE Vernova HA-class from fact sheets GEA35750/GEA35768 plus F-class and aeroderivatives from product pages; Siemens Energy HL/H/F-class and SGT-800 from product pages. Every entry records its source, and CI asserts the internal consistency |HR − 3600/η| ≤ 30 kJ/kWh for each machine (vendors round HR and η independently) and that combined-cycle output and efficiency always exceed the simple-cycle values. Ratings are net or gross per each vendor's own publication convention — the estimate is a screening number, not a performance guarantee; degradation, ambient conditions, part load and fuel composition all move real-world fuel burn.
11.4 Unit switches (v3.5)
The consumption totals convert with the same named constants as the cargo card: MWh × 3.6 = GJ, GJ ÷ 1.055056 = MMBtu, and each family auto-scales (kNm³ → MMNm³, MMscf → Bscf, MWh → GWh, GJ → TJ → PJ, MMBtu → TBtu). Two energies are tabulated: fuel energy in (Q × hours — what the plant buys) and power sent out (P × hours — what it sells); their ratio is the efficiency, and the yearly fuel energy in TBtu is directly comparable to LNG cargoes (Part IX). Fuel volume in scf uses the mandated 37.3258 scf/Nm³ of Part VII. For the M701JAC example: 733.93 MMNm³ = 27.394 Bscf, fuel energy 29.54 PJ = 28.00 TBtu, sent out 3,610.5 GWh = 13.00 PJ = 12.32 TBtu, fuel-energy input 1,018.18 MW-th = 3,665.5 GJ/h = 3,474.2 MMBtu/h. Gas density in lb/scf converts with 16.9307 kg/Nm³ per lb/scf — see Part IX §9.2 for why that is not 16.0185.
Part XII — PRV Sizing Theory (API 520 §5.6–§5.10)
12.1 Gas/Vapor — Critical Flow (§5.6.3, Eq. 2/5)
12.2 Gas/Vapor — Subcritical Flow (§5.6.4, Eq. 12/15)
12.3 Steam (§5.7, Eq. 21/22)
12.4 Liquid — Certified PRV (§5.8, Eq. 28/29)
12.5 Liquid — Non-Certified PRV (§5.9, Eq. 38/39)
12.6 Two-Phase — Omega Method (§5.10 / Annex C.2.2)
12.7 API 526 Standard Effective Orifice Areas
| Letter | in² | mm² | Letter | in² | mm² |
|---|---|---|---|---|---|
| D | 0.110 | 71.0 | N | 4.340 | 2800 |
| E | 0.196 | 126.5 | P | 6.380 | 4116 |
| F | 0.307 | 198.1 | Q | 11.05 | 7129 |
| G | 0.503 | 324.5 | R | 16.00 | 10323 |
| H | 0.785 | 506.5 | T | 26.00 | 16774 |
| J | 1.287 | 830.3 | |||
| K | 1.838 | 1186 | |||
| L | 2.853 | 1841 | |||
| M | 3.600 | 2323 | |||
12.8 Input Basis — Pressure Units and Abs/Gauge
Part XIII — Data Sources & Standards
Terms of Use
Last updated: August 2026 — Version 3.9
Translations of these Terms of Use are provided for convenience only. The English version is authoritative and governs in the event of any discrepancy.
1. Acceptance of Terms
By accessing or using the O&G Engineering Converter ("Application"), you agree to be bound by these Terms of Use. If you do not agree, you must not use the Application. These terms apply to all users including engineers, operators, researchers, and students.
2. Nature of the Application
The Application is a browser-based engineering utility. The majority of calculations are performed locally within your web browser. However, three features — the Pipe Delta Pressure Calculator and the Flow Regime visualizer (Advanced tab), and the PRV Sizing Calculator (Safety tab) — transmit engineering input data (such as flow rates, pressures, and fluid properties) to serverless API functions hosted on Vercel for server-side computation, including server-side rendering of the flow regime map image. No personally identifiable information is transmitted in these requests. The Application is provided as a convenience tool for engineering estimation and is not intended to replace formal engineering analysis, professional judgment, or regulatory-approved calculation methods.
3. Disclaimer of Warranties
The Application is provided "as is" and "as available" without warranties of any kind, express or implied, including but not limited to:
- Accuracy, correctness, or completeness of any calculation result
- Fitness for any particular engineering, commercial, or regulatory purpose
- Uninterrupted or error-free operation
- Compliance with any specific national or international standard
While the calculations are based on JIS K 2301:2011, ISO 6578:1991, and other referenced standards, users must independently verify all results before applying them to any operational, safety, commercial, or regulatory context. Several tools deliberately use simplified or screening-level models and are provided for qualitative orientation only — they must not be used as a basis for design, operational, or safety decisions. These include: the Flow Regime visualization (approximate regime-map boundaries + a conceptual 3D animation), the Pipe Delta Pressure two-phase calculation (selectable screening correlations — HEM default, Lockhart-Martinelli, Müller-Steinhagen-Heck, Friedel — none with slip in the static head or an acceleration term), the erosional-velocity check (API RP 14E empirical screening), and the Z-Factor Estimator (Papay correlation, valid only within its stated Pr/Tr envelope). Where an input falls outside a method's validated range, the Application flags the result as extrapolated; such results are indicative only.
Screening tools added in v2.8–v3.1 carry the same qualification: the Gas Property Estimator (Lee-Gonzalez-Eakin viscosity, sonic velocity and Joule-Thomson coefficient on the Papay Z-factor), the Steam Properties card (IAPWS-IF97 industrial formulation, Regions 1/2/4 only), the NPSHa card (available head only — it deliberately issues no margin verdict against the pump vendor's NPSHr), the Compressor card (isentropic and perfect-gas polytropic relations with a one-pass real-gas correction; not a test-code procedure), and the GT Fuel estimator and catalogue. The GT catalogue figures are indicative ratings transcribed from vendors' publicly available materials at ISO conditions; they may be superseded by later vendor revisions, are net or gross per each vendor's own convention, and do not constitute a performance guarantee — actual fuel consumption depends on site conditions, degradation, part load and fuel composition. Always confirm against the current vendor datasheet.
4. Limitation of Liability
To the maximum extent permitted by applicable law, the developer (Naoto Yamabe) and any associated parties shall not be liable for any direct, indirect, incidental, special, consequential, or punitive damages arising from:
- Use of or reliance on any calculation result from this Application
- Errors, omissions, or inaccuracies in the Application's output
- Any operational, commercial, safety, or financial decisions made based on Application output
- Loss of data, business interruption, or equipment damage
5. Intended Use and User Responsibilities
This Application is designed for use by qualified professionals familiar with oil and gas engineering principles. Users are solely responsible for:
- Validating input data and its suitability for the selected calculation method
- Cross-checking outputs against authoritative reference materials and standards
- Ensuring compliance with applicable local, national, and international standards and regulations
- Applying appropriate safety factors and engineering margins in practice
6. Intellectual Property and Licence
The Application is open source, released under the MIT Licence. You are free to use, copy, modify and redistribute it, including commercially, provided the copyright notice and the licence text are retained. The full text is in the LICENSE file of the public repository. Copyright © 2026 Naoto Yamabe. (Changed in v3.3 — earlier versions of these Terms prohibited redistribution and commercial use without written permission. That restriction no longer applies.)
The licence covers the Application’s own code, design and documentation. It grants no rights in third-party material the Application cites or transcribes. The underlying calculation methods and physical constants are derived from publicly available engineering standards (JIS K 2301, ISO 6578, API 520 / API 526, API RP 14E, NORSOK P-001, IAPWS-IF97, Crane TP-410 and others), which remain the property of their respective issuing bodies; their texts, tables and figures are not reproduced here. Anyone redistributing or building on the Application remains responsible for their own compliance with the terms attached to that material.
Third-party names and data. The GT Fuel catalogue refers to gas-turbine manufacturers and model designations (including MHI / Mitsubishi Power, GE Vernova and Siemens Energy) solely to identify the equipment; those names and marks are the property of their respective owners, and the performance figures shown are transcribed, with source citations, from the manufacturers' publicly available brochures, fact sheets and product pages. The Application is an independent tool that is not affiliated with, sponsored by, or endorsed by any of these manufacturers. The catalogue thumbnails are original schematic illustrations created for this Application and are not manufacturer imagery. The LNG reference compositions (v3.3) badged PUBLISHED are transcribed, with citation, from GIIGNL Information Paper No. 1, “Basic Properties of LNG”, Table 1 (data credited to the 2018 GIIGNL Annual Report), and remain the property of GIIGNL; those badged ASSUMED are the developer’s own engineering assumptions and are labelled as such in the Application.
LNG vessel data and photographs (v3.5). The LNG Cargo Estimator lists representative LNG carriers by name, IMO number, owner, builder, capacity and type solely to identify the vessels; each entry cites the public page it was taken from, and vessel names and owners’ marks remain the property of their respective owners. Appendix 3 of the IGU World LNG Report 2026 — the table of the global active LNG fleet (data: Rystad Energy) — is reproduced in the vessel list with the International Gas Union’s written permission of 24 August 2026, with full attribution and a link to the original source; the data is not to be re-extracted from the Application. Vessel photographs are Wikimedia Commons works reproduced under their stated Creative Commons or public-domain terms with the author credited on screen (see assets/vessels/CREDITS.json); the remaining thumbnails are original schematics created for this Application. The Application is not affiliated with, sponsored by, or endorsed by any shipowner, shipbuilder, classification society, MarineTraffic, VesselFinder or the IGU.
7. Modifications and Updates
The developer reserves the right to modify, update, or discontinue the Application at any time without notice. These Terms of Use may be updated periodically; continued use of the Application constitutes acceptance of any revised terms.
8. Governing Law
These Terms shall be governed by and construed in accordance with the laws of Japan. Any disputes arising from the use of this Application shall be subject to the exclusive jurisdiction of the courts of Japan.
Privacy Policy
Last updated: August 2026 — Version 3.8
Translations of this Privacy Policy are provided for convenience only. The English version is authoritative and governs in the event of any discrepancy.
1. Overview
The O&G Engineering Converter is committed to protecting your privacy. This policy describes how the Application handles information. As a client-side application, it is architecturally designed to minimize data collection and external transmission.
2. Data We Do NOT Collect
The Application does not collect, transmit to the developer, or remotely store any of the following:
- Personally identifiable information (name, email, IP address)
- Gas composition data or engineering inputs you enter (these are kept only in your own browser — see Section 3 — except the transient numerical parameters sent to the calculation APIs in Section 4)
- Calculation results or session history on any server
- Browser fingerprinting or device identifiers
- Cookies — of any kind, for any purpose
- Advertising, ad networks, or data sold or shared with third parties
Changed in v3.2. Earlier versions of this policy also stated that the Application collected no usage analytics of any kind. That is no longer true: the Application now records anonymous, cookieless usage statistics so the developer can see which calculators are actually used and prioritise accordingly. Nothing you type is ever included. Section 7 sets out in full what is recorded.
3. Browser Local Storage
The Application uses your browser's Local Storage (a built-in browser feature) to save (a) custom converter modules you create in the General tab and (b) — new in v2.4 — your most recent calculator inputs and UI preferences, so the page restores your last session automatically. This data:
- Is stored only on your local device, within your browser
- Never leaves your device or browser, and is never transmitted to any external server
- Can be cleared at any time by clearing your browser's local storage or site data
- Contains only your custom module definitions and the numerical inputs / unit selections you typed into the calculators
Share links (v2.4). The "Share" button encodes your current inputs into the page URL (after the # fragment) and copies it to your clipboard. This is performed entirely in your browser; the fragment is never sent to the server. Anyone you give the link to can reconstruct those inputs, so avoid sharing links containing data you consider sensitive.
Share links now carry custom modules (v2.8). A share link also encodes the definitions of any custom converter modules you have created — their title, unit labels and conversion factor. When someone opens such a link, those modules are added to their browser's local storage and persist after reload, so they remain usable. A short on-screen notice states how many were added. Imported module text is sanitised on arrival and their internal identifiers are regenerated, so a link cannot overwrite or delete modules the recipient already had. Anyone can remove an imported module with its TERMINATE button, or by clearing site data.
The bug-report form is excluded (v2.8). Anything typed into the Report tab — your name, the description, the report type — is no longer written to local storage or encoded into share links. Previously those fields were captured along with the calculator inputs. That text is now used only when you choose to open your email client to send the report.
Language preference (v2.6). Your chosen interface language is also kept in local storage (a two-letter code such as ja) so the page opens in that language next time. Like everything else in local storage it never leaves your browser; a share link may carry the sender's language so the recipient sees the same view.
4. Server-Side Calculation APIs
Three features in this Application — the Pipe Delta Pressure Calculator, the Flow Regime visualizer, and the PRV Sizing Calculator — send engineering input data to serverless functions hosted on Vercel for computation. The data transmitted consists exclusively of numerical engineering parameters (e.g. pressures, flow rates, fluid properties). For the Flow Regime visualizer, the server also renders and returns a map image generated from those parameters. This data is processed transiently to return a calculation result and is not logged, stored, or associated with any user identity by the Application. Standard Vercel infrastructure logging may capture request metadata (see Section 5 below).
5. Hosting & Third-Party Services
The Application is hosted on Vercel. As with any web hosting service, Vercel may collect standard server access logs (such as request timestamps and IP addresses) as part of their infrastructure operations. This is outside the control of the Application developer. Please refer to Vercel's Privacy Policy for details on their data handling practices.
The Application also loads open-source libraries from public content delivery networks (CDNs): the Tailwind CSS framework (cdn.tailwindcss.com) and, only when the Flow Regime 3D animation is first used, the Three.js 3D library (cdnjs.cloudflare.com). Like any web resource request, these providers may log standard request metadata (e.g. IP address, user agent). No engineering input data or calculation results are sent to any of them.
Vercel Web Analytics (v3.2). The Application additionally loads Vercel's cookieless analytics script, served from the Application's own domain. Vercel is the sole processor of this data and it is not shared with anyone else. Section 7 describes in full what is and is not recorded.
6. Report / Feedback Feature
The Report tab generates a mailto: link that opens your local email client. No data is transmitted through the Application — your email client handles the message. The developer (Naoto Yamabe) may receive your email address and the content you write if you choose to send the email. This information will be used solely to respond to your report or request and will not be shared with third parties.
7. Cookieless Usage Analytics (v3.2)
The Application uses Vercel Web Analytics, a privacy-focused analytics product provided by the Application's host. It sets no cookies, reads nothing from your device's storage, and performs no cross-site tracking. There is no Google Analytics, no advertising pixel, and no third-party tracker of any kind.
Two categories of information are recorded:
- Page views — the page address, the referring site (e.g. that you arrived from LinkedIn or a search engine), your country, and your device type, operating system and browser.
- Anonymous interaction events — four kinds only: which tab you open (including which sub-section of the Advanced tab), which calculator you use (recorded as a fixed name such as steam-if97 or pipe-dp), which interface language you select, and whether you used Export PDF, Share, or the report form.
What is never recorded. No value you type or select is transmitted — not a gas composition, not a pressure, not a flow rate, not a calculation result, and not one character of the Report tab, which is excluded from analytics entirely. An event records only that a calculator was used, never what it was used on. For the live-updating converters, at most one event is recorded per calculator per visit, no matter how much you type.
How visits are counted. To count a visit once without a cookie, Vercel derives a non-reversible hash from the incoming request (IP address and user agent). The raw IP address is not retained by the analytics product, and the hash is rotated, so your visits cannot be linked together across days or to your activity on any other website. No profile is built and no identifier is stored on your device. Vercel acts as the data processor; see Vercel's Web Analytics privacy documentation for the authoritative description.
Opting out. The analytics script is a standard, widely recognised script file. Any content blocker or privacy-focused browser will block it, and the Application is designed to work identically when it is blocked — every calculator functions normally and no error is shown.
8. Children's Privacy
The Application is intended for professional use by adults. It does not knowingly collect information from children under the age of 16. If you believe a child has submitted information through the Application, please contact the developer.
9. Your Rights
The Application collects no personal data, so there is no personal data to access, rectify, delete, or export. The usage analytics described in Section 7 are anonymous and carry no identifier that could be used to locate, isolate, or delete "your" records — which also means a deletion request cannot be fulfilled, because there is nothing to match it against. If you would prefer not to be counted at all, block the analytics script as described in Section 7. To remove your locally stored custom modules, use your browser's developer tools or site data clearing options (Settings → Privacy → Clear Site Data for the relevant domain).
10. Contact
For privacy-related questions or concerns, please contact the developer via the Report tab or at petro.naoto@gmail.com.
System Report & Request Form
Found a calculation bug or have an idea for a new module? Fill out the form below. This will automatically draft an email to the developer (Naoto Yamabe).