Manufacturing

Rotary screw air compressor

Compressed air is the most expensive utility in the plant. The physics shows where it goes.

An air compressor turns electricity into pressure. How much power it takes for the same air is set by the pressure ratio, the filters and the leaks. We read all three from the meter.

300 acfm at 125 psig, 75 hp

Sketch of an oil-flooded rotary screw air compressor: inlet filter, intake valve, screw airend, oil separator tank, cooler and receiver, with numbered sensor points.12345678air into plant75 hp motor
Full-load input
54.1 kW
Specific power
18.0 kW per 100 cfm
Delivery
300 acfm
Setpoint
125 psig

Anatomy and nameplate

Sketch of an oil-flooded rotary screw air compressor: inlet filter, intake valve, screw airend, oil separator tank, cooler and receiver, with numbered sensor points.12345678air into plant75 hp motor
  1. 1Inlet filter pressure drop. Restriction raises the pressure ratio.
  2. 2Airend discharge temperature and pressure. Shows cooler fouling and separator loading.
  3. 3Separator pressure drop. The element clogs. Power climbs for the same air.
  4. 4Oil temperature. Oil flow and cooler health.
  5. 5Motor current. Package power without a new meter.
  6. 6Airend vibration. Rotor and bearing condition.
  7. 7Line pressure. Setpoint and the demand it has to hold.
  8. 8Power meter. The number the baseline is measured against.
Rotary screw air compressor nameplate
Performance
TypeOil-flooded rotary screw, single stage
Free air delivery300 acfm
Rated pressure125 psig
Full-load input power54.1 kW
Specific power18.0 kW per 100 cfm
Rotor profile4 male lobes, 6 female lobes
Male rotor speed3,560 rpm
Discharge temperature185–195 °F
Motor and package
Motor75 hp (55 kW), 4-pole, TEFC
Supply460 V, 3-phase, 60 Hz
Full-load current84 A
Separator pressure drop3 psi new, change at 10 psi
CoolingAir-cooled oil and air coolers, 1.5 kW fan
Oil charge12 gal synthetic

The physics

The equations behind every curve on this page.

Compression work

P = k/(k−1)·p1·Q1·[(p2/p1)^((k−1)/k) − 1] / η

More pressure ratio, more power, for every cubic foot.

Specific power

kW per 100 cfm

18.0 kW per 100 cfm at full load and 125 psig.

Orifice leak

ṁ = Cd·A·p0·√(k/(R·T0))·(2/(k+1))^((k+1)/(2(k−1)))

Flow through a hole grows with the square of its diameter.

What the air costs

Power follows the pressure ratio. The separator sits in the path, so a clogging element adds to it.

Input power against system pressure

300 acfm delivered, new separator element

Input power against system pressure404550556065709010011012013014015054.1 kW at 125 psig49.6 kW at 105 psigSystem pressure setpoint (psig)Input power (kW)
  • Input power at 300 acfm
Each +2 psig
+0.8% power

Every extra pound of pressure is paid for in power on every cubic foot. The rule of thumb that 2 psi costs about 1% falls out of the equation.

P = k/(k−1) · p1 · Q1 · [(p2/p1)^((k−1)/k) − 1] / η

Clogging separator: more power for the same air

Clogging separator: more power for the same air505254565860620510152025Replace element, 10 psiNew 54.1 kWClogged 57.3 kWSeparator pressure drop (psi)Input power (kW)
  • Input power at 300 acfm and 125 psig
Cost of waiting
3.1 kW, about $2,069 a year

The airend has to push against the separator as well as the plant. Output stays the same, the meter climbs.

p_airend = p_line + Δp_separator + Δp_piping

Leaks

Leaks are free air that someone paid to compress. The size of the hole sets the size of the bill.

What a leak costs

What a leak costs02040608002468101214160.000.050.100.150.200.251/8 in: 20 scfm, 3.7 kWHole diameter (in)Air lost (scfm)Power wasted (kW)
  • Air lost at 125 psig
  • Power wasted (right axis)

Flow through a small hole grows with the square of its diameter. A few small leaks can eat the output of a whole compressor.

ṁ = Cd·A·p0·√(k/(R·T0))·(2/(k+1))^((k+1)/(2(k−1)))

Failure modes

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

Failure modeThe physicsSignals that show itHow the twin catches it
Separator element cloggingPressure drop across the element adds to the airend discharge pressure.Separator differential pressure up. Package kW up at the same flow.Expected kW at the current pressure and flow. The gap is the clogging.
Air leaksChoked flow through holes in pipes, couplings and tools.Unloaded-time falls. Idle kW rises. Pressure sags at night.Converts off-shift flow into equivalent hole size and kW.
Cooler foulingLess heat leaves the oil, so discharge temperature rises.Discharge temperature up at the same ambient.Heat balance on the oil circuit. Compares expected and measured temperature.
Airend bearing wearRotor bearing defects ring at fixed geometric frequencies.Vibration lines at bearing frequencies. Rising motor current.Tracks band energy at the predicted frequencies.
Inlet filter loadingLow inlet pressure raises the pressure ratio.Inlet vacuum. Falling delivered flow.Flags the added kW and the day the filter pays for itself.

How we catch it early

A physics twin and a PINN, working on one rotary screw air compressor.

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

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

  1. Step 1

    Measure

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

  2. Step 2

    Physics twin

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

  3. Step 3

    Residual

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

  4. Step 4

    PINN

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

  5. Step 5

    Finding

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

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

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

Why this beats black-box AI

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

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

What the twin checks on this asset

  • Specific power: 18.0 kW per 100 cfm
  • Separator pressure drop against the 10 psi change point
  • Off-shift flow converted to leak size
  • Discharge temperature against the heat balance
Fixed alarmBlack-box AITwin + PINN
Needs failure history to learnNoYes, lotsNo
Says what is wrongNoRarelyYes, by name
Works at a flow it has never seenNoOften failsYes, the equations still hold
Puts a number on the savingNoNoYes, in kWh and dollars
Lead time on slow wearLateVariesWeeks to months
Every number can be tracedYesNoYes, to an equation

What the twin puts a number on

Energy and cost, from the equations.

Full-load input power

54.1 kW

300 acfm at 125 psig

Clogged separator, 18 psi

3.1 kW

$2,069 a year at 6,000 hours

One 1/8 in leak

3.7 kW

20 scfm lost

Pressure setpoint

0.8% per 2 psig

Lower the setpoint where the process allows

What we do for you

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

Our reliability engineers run the twin on your rotary screw air compressor, review every finding, and help your crew fix it. After the repair we compare the meter to the baseline and write the result in the ledger.