Allan Ronald Dones
Rotating Equipment Engineering

Compressor Performance Estimator

Single-stage centrifugal and reciprocating compressor estimation — polytropic head, discharge temperature, gas and shaft power, impeller sizing, volumetric efficiency and cylinder bore. Every parameter is carried in SI units on the left and English units on the right, so the sheet reads the same whichever system the datasheet arrives in.

Developed by Allan Ronald Dones — Senior Mechanical Engineer, Rotating Equipment Specialist

Enter a value on either side of the divide — the opposite unit system updates automatically and the whole sheet recalculates.
SI units — left of the divide English units — right of the divide Input field Calculated output + Added — not in the source workbook Everything runs in your browser — no data leaves this page.

1Centrifugal Compressor — Single Stage Body

Polytropic path. Each case column is independent — use them for the operating envelope (minimum, normal, maximum) or for alternative suction pressures.

2Reciprocating Compressor — Single Stage

Adiabatic path by default (compression efficiency = 1), with the shaft power taken through a separate mechanical efficiency exactly as in the source sheet. Cylinder sizing assumes double-acting cylinders.

3Screening Checks

Rules-of-thumb applied to the calculated results — a sanity filter for a selection study, not acceptance criteria. Nothing here is mandated by API 617 or API 618; the vendor's proposal and the purchase specification govern.

4Method & Correlations

Every relation below is evaluated internally in English units — the constants are English-based — and the SI column is a unit conversion of the result.

QuantityRelation (English units)
Mass flowW = SCFM · MW / 379.5 — 379.5 scf/lbmol at 14.696 psia and 60 °F
Inlet volume flowQs = W · (1545/MW) · Zs · Ts / (144 · Ps)
Pressure ratior = Pd / Ps
Temperature exponentM = ((k−1)/k) / ηp — the polytropic exponent group (n−1)/n
Discharge temperatureTd = Ts · rM (absolute)
Polytropic headHp = Zavg · R · Ts · (rM − 1) / M, with R = 1545/MW
Gas powerGHP = W · Hp / (33 000 · η) — η = polytropic efficiency (centrifugal) or mechanical efficiency (reciprocating)
Shaft powerBHP = GHP + mechanical loss allowance
Speed / impeller diameterN = (1300/D) · √(Hp / (nimp · μ)) — the constant 1300 is 720·√g/π, so the relation is the exact restatement of Hstage = μ·u²/g with D in inches
Flow coefficientφ = 700 · Qavg / (N · D³) — the constant 700 is 1728·4/π²; Qavg is the mean of inlet and discharge volume flow
Tip speed +u = π · D · N / 720 ft/s, with D in inches
Machine Mach no. +Mu = u / a, a = √(k · gc · Zs · R · Ts), gc = 32.174
Volumetric efficiencyVE = 1 + C − C · r(1/k) — clearance volume only, no valve or gas-passage losses
Cylinder boreDcyl = √( 1728 · 4 · Qs / (S · N · π · 2 · ncyl · VE) ) — the factor 2 is the double-acting assumption
Piston speedvp = 2 · N · S / 12 ft/min, with S in inches
1 MMSCFD = 1 116.252 Nm³/h 1 lb/min = 27.2155 kg/h 1 ACFM = 1.699011 m³/h 1 ft·lbf/lbm = 2.98907 J/kg 1 hp = 0.74570 kW

§Notes & Conventions

  1. Scope — estimation, not selection. This is a single-stage screening calculation. It sizes nothing and guarantees nothing: it tells you roughly how much head, how much power, how hot the discharge runs and how big the machine has to be, which is what you need before a vendor is in the room. The certified vendor performance curve governs.
  2. Compressibility. Zs and Zd are entered, not calculated — take them from a process simulation or an equation of state at the actual suction and discharge conditions. The head relation uses their arithmetic mean, Zavg = (Zs + Zd)/2. Because Zd depends on the discharge temperature this calculation produces, a second pass with an updated Zd is worthwhile where the ratio is high.
  3. Constant k. A single ratio of specific heats is applied across the stage. For real gases at high pressure ratio, use the average of the suction and discharge values rather than the suction value alone.
  4. Rankine offset. Absolute temperature uses the source workbook's °R = °F + 460 rather than the exact 459.67. The SI absolute column is converted from that same value so the two halves of the sheet stay consistent with each other; the effect on head and power is below 0.1%.
  5. Reciprocating efficiency. The source sheet sets the compression efficiency to 1 — an adiabatic path — and takes the shaft power through a separate mechanical efficiency in the power denominator. Both are exposed here. Leaving the compression efficiency at 1 reproduces the workbook exactly.
  6. Volumetric efficiency. The clearance-only expression neglects valve losses, gas passage losses and suction gas heating. Real machines typically run 3 to 8 percentage points below the value shown, so the calculated bore is on the optimistic side.
  7. Cylinder count. The bore relation assumes double-acting cylinders. For single-acting service, halve the cylinder count entered — or read the result as the bore of an equivalent double-acting machine.
  8. Empirical constants. The 1300 and 700 constants in the impeller relations are exact restatements of 720√g/π and 1728·4/π², so they carry no hidden fudge factor. What they do carry is the assumption of a simple head coefficient μ applied identically to every impeller in the body.
  9. Deviation from the source workbook. Two rounded constants in the original — π taken as 3.1416 in the bore relation, and 3.281 ft/m in the piston speed conversion — are replaced here by their exact values. The difference is roughly 1 part in 10⁶. Everything else reproduces the workbook to the last digit.
  10. Discharge volume flow. Qd is evaluated rigorously as Qs · (Td · Zd · Ps) / (Zs · Ts · Pd), which is what the source sheet uses inside the flow coefficient. Where Zs = Zd this is identical to the sheet's separate Qd/N = (Qs/N)·rM/r row.
Not covered. Multi-stage or multi-section trains with intercooling and sidestreams, rotordynamics and lateral critical speeds, surge and stonewall limits, real-gas Schultz polytropic head corrections, rod loading and rod reversal, pulsation study requirements, and driver selection beyond the sizing margin. Each of those is a separate exercise.