Pneumatic Air Motor Selection

1. Engineering answer: select a working point and verify the complete duty

A pneumatic air motor should be selected from the torque and speed required at the driven shaft, then checked against the exact motor's supported performance at the pressure available at its inlet and the applicable exhaust condition. Starting capability, acceleration, variable loads, speed control, transmission losses, air consumption, shaft loading and stopping behaviour are separate checks. A motor that can turn the load once has not necessarily demonstrated that it can start reliably, maintain the required speed or operate economically throughout the production cycle.

Begin with the useful process outcome. A mixer may require sustained rotation under changing material resistance; a conveyor may require breakaway torque after a dwell; a spindle may require a limited operating-speed band; a winding mechanism may change torque with reel radius. These are different load envelopes even when their calculated mechanical powers happen to match. The useful enquiry therefore contains the operating cases, not one isolated kilowatt value. [N1]

The result of a sound selection is a documented motor-and-installation candidate with clear conditions and remaining checks. A missing consumption curve, uncertain breakaway load or unverified stopping requirement should remain visible. No generic air-motor category supplies those missing facts. The calculations and candidate data below are deliberately fictional teaching material; they are not current product ratings, quotations or measured customer outcomes.

2. Keep this page distinct from actuators, tools and the selection worksheet

A motor can be used in an indexing mechanism, but position control, stopping, locking and repeatability then become requirements of the complete machine. Continuous rotation capability does not by itself provide accurate angular positioning.

This division keeps the present article focused on selecting and integrating a motor rather than repeating each neighbouring subject in full.

A powered hand tool is also more than its motor. Tool speed limits, guards, accessory compatibility and the productive process belong to the complete tool specification. Likewise, an air motor connected to a drum does not become an approved hoist or winch simply because the torque calculation appears adequate. Lifting and load retention need their own competent design and equipment evidence.

Use this page to prepare a new application, evaluate a replacement, investigate an underperforming drive or compare pneumatic and other drive options. Do not use its fictional curves as a substitute for a supplier's selection. The required asset is a torque–speed duty comparison that preserves the motor, transmission, air path, control arrangement and evidence basis together. [S4; N1]

3. Define the driven load before choosing a motor family

State where the required torque is measured: motor shaft, gearbox output, roller, impeller, spindle or another driven interface. Record the speed at that same point. An output-shaft requirement cannot be compared directly with a high-speed motor curve before the transmission is accounted for. Give the driven equipment and each operating case stable identifiers so subsequent curve readings remain attached to the correct duty.

Describe normal production, starting after the longest relevant dwell, peak resistance, speed changes, reversal, expected stops and credible abnormal loading. Establish which values are measured, supplier-provided or preliminary assumptions. A process torque measured with warm material may not describe a cold start. Conversely, an isolated jam should not automatically be treated as an acceptable continuous duty. The process owner must define which events the drive is intended to perform and which it must detect and stop. [N1]

Keep the required speed band and the maximum permissible speed separate. A process tolerance describes acceptable operation; a mechanical limit describes a boundary that must not be exceeded. Do not add a percentage margin to a process target and accidentally reinterpret it as permission to exceed the driven machine's limit.

Where duty is genuinely unknown, plan a measurement or supplier clarification. A provisional design case is acceptable when labelled, but it should not disappear into a completed-selection badge. The enquiry should state which unknown can change motor size, gearbox ratio or control architecture.

4. Choose a motor architecture without assuming one universal curve

Motor architecture is a starting filter, not the final selection. Establish whether a vane, piston, turbine or geared assembly has documented performance suited to the required speed, torque and environment. Different designs need different operating evidence. A curve convention used for one ungoverned vane motor must not be imposed on every pneumatic motor.

Ingersoll Rand's published selection guidance distinguishes vane and radial-piston motors and describes its piston arrangements as candidates for lower-speed, high-torque duties. Its guidance also distinguishes governed and ungoverned behaviour. These are manufacturer-specific selection directions, not a universal ranking or a guarantee for an unspecified model. [R5]

Avoid the assumption that the physically largest motor is the most robust answer. A different motor may meet the desired point only after throttling or pressure adjustment; that changes the conditions under which starting and running must be checked. Extra capacity can also create excess speed, more airflow or larger transmission loads.

The candidate-family decision should explain why the family is being investigated and what remains unresolved. Do not infer current availability, local stock, authorised service, hazardous-area suitability or a particular service life from the family name. Those are separate product and commercial evidence items.

5. Separate the performance quantities that are often confused

The torque required by the load is not the torque available from the motor. The available value must be tied to speed, pressure, direction and configuration. Similarly, mechanical shaft power is not compressor electrical power or a direct measure of air consumption. Keeping the quantities separate prevents a plausible calculation from disguising a mismatched motor.

At zero shaft speed the mechanical output power is zero even when torque exists and air is being consumed. At a nonzero speed, the same power can correspond to very different torques. Neither statement tells the designer how much torque is available to start the real load.

Record the meaning of a quoted “rated speed.” It may describe the speed at maximum power, an output-shaft rating or another declared point. Where a document uses an abbreviation without a definition, request clarification instead of choosing whichever interpretation makes the candidate appear suitable. Preserve the original label in the evidence register.

6. Calculate mechanical power with a named shaft and consistent units

For steady rotation, or instantaneous compatible measurements, mechanical power follows:

• P_shaft,W = 2π × n_rpm × τ_Nm / 60
• P_shaft,kW = 2π × n_rpm × τ_Nm / 60,000

The inverse running-duty relationship is τ_Nm = 60,000 × P_shaft,kW / (2π × n_rpm), with n greater than zero. The factor of 1,000 is essential when changing watts to kilowatts.

Use the formula to cross-check an enquiry or supplier row. Do not multiply maximum torque by maximum speed when those values belong to different operating conditions. The product is meaningful only when torque and speed occur together at the named boundary. A catalogue arithmetic discrepancy also needs a rounding and definition check before it is described as a measurement error.

For a changing duty, average torque multiplied by average speed is not generally equal to average power. The applicable relationship is the time average of the compatible torque–speed product. Likewise, mechanical energy is the time integral of mechanical power. If the data contain only a process torque at one speed, the result describes that process state, not all acceleration, idle and braking energy. [N2]

The power identity is not a motor-performance model. It cannot determine the working curve, starting capability, air consumption or permitted bearing load. Those still require the motor and driven-equipment evidence. It is particularly misleading to calculate a small steady power at a low speed and conclude that any motor with that power label will deliver the required high torque.

7. Find the load–motor intersection, not merely a positive torque margin

A motor connected to a load tends to change speed according to the difference between available motor torque and opposing load torque. In a simplified single-shaft model, J × dω/dt = τ_motor − τ_load. A steady operating point requires equality of the two torques. Friction and any other resisting contribution must either be included in the load term or recorded separately, not counted twice. [N2]

A positive available-torque margin at the desired speed shows capacity to accelerate through that point; it does not prove that the shaft will remain there. Without a suitable control or another change in the load, the speed may continue rising until a different intersection is reached. This is why “the motor exceeds the required torque at target speed” is an incomplete speed-control conclusion.

For a local screening analysis, a negative slope of net torque with increasing speed tends to oppose a small speed disturbance. That observation assumes a suitable quasi-steady model, positive effective inertia and no dominant controller or pneumatic delay. It is not a stability certification. A real system can include compliant shafts, changing chamber pressures, delayed regulation and time-dependent process loads. [N2]

The selection record should therefore show the candidate torque curve, the required load envelope and the proposed method for achieving the speed band. Where only a few supplier points exist, do not fabricate a smooth curve through unsupported regions. Ask for data or retain an explicitly bounded interpolation for preliminary study.

This approach also explains why a larger motor may require more control rather than simply provide more reserve. The engineering question is whether the complete drive can meet the speed–torque envelope through the required operating states, not whether one candidate value is larger than one load value.

8. Check starting torque independently from stall and running torque

Starting can govern selection even when the continuous shaft power is modest. Record breakaway resistance, direction, dwell condition and any process load present at the instant of start. Where measured starting loads vary, retain the measurement range and the agreed selection basis instead of replacing them with one favourable observation.

Atlas Copco's vane-motor guidance explains that starting torque varies with vane position and identifies minimum starting torque separately from running performance. This supports requesting the applicable minimum value rather than estimating it from the top of a torque curve. Its general stall approximations are not adopted here as guaranteed starting ratings. [R1]

Calculate separate margins: available minimum starting torque minus required starting torque, and available running torque at the applicable point minus required running torque. A positive running margin cannot repair a negative starting margin. If a gearbox is present, the starting comparison needs the applicable starting transmission behaviour; a running efficiency is not automatically valid at breakaway. [S1; N2]

A statement that a particular air motor tolerates stalling does not authorise an uncontrolled jam in the driven machine. Couplings, gears, workpieces and mechanisms still experience torque; air may continue flowing; release of the jam may produce sudden acceleration. Record whether stall is prohibited, expected briefly or part of an explicitly supported process, and define the authorised response.

Similarly, do not use motor stall as a substitute for a load-holding brake. The required holding time, possible back-driving and energy-loss condition need a separate mechanical and control assessment. The present selection guide does not establish a safe retention function.

9. Read supplier curves as a complete evidence set

For each candidate, retain the exact motor and gearbox designation, direction, performance-sheet revision, inlet pressure, exhaust condition, lubrication arrangement and interpretation of the output shaft. Capture the torque, speed, power and consumption data that belong together. A family brochure can identify candidates but may not settle the offered configuration.

Atlas Copco's catalogue-reading guidance presents torque, power and consumption against speed for a motor/gear unit and obtains minimum starting information separately from tabulated data.

Check whether a curve describes a bare motor or an integrated gearmotor. A gearbox already included in an output curve must not be applied again. Also check whether values are measured, declared, interpolated or estimated. Preserve the pressure and direction for every row; a strong clockwise row cannot silently satisfy a reverse-direction requirement.

For preliminary linear interpolation between two supported points, a transparent expression is y = y₁ + (y₂−y₁)(n−n₁)/(n₂−n₁). Both endpoints must belong to the same quantity, pressure and configuration, with distinct speeds. This is an analyst's interpolation, not new manufacturer data. Do not extrapolate outside the range or interpolate through a control-mode change. [N2]

Request uncertainty, tolerances or performance variation where they affect the decision. A nominal curve just touching the load requirement leaves no demonstrated allowance for variation. Keep a nominal match, a supported guaranteed envelope and actual acceptance measurements as different evidence states.

10. Translate the duty through the gearbox without double counting

Define reduction ratio as i = n_motor/n_load. For forward motoring under a stated running efficiency η_g, n_motor = i × n_load and τ_motor,required = τ_load/(i × η_g). The motor-side shaft power is P_load/η_g.

The running efficiency belongs to the actual gearbox, speed, load and direction. Starting friction, reverse transmission and back-driving can require different evidence. Do not assign one optimistic efficiency to every state. Where only a preliminary estimate is available, show a sensitivity range and keep final selection conditional.

Gearboxes also have permissible input speed, output torque, thermal duty, shock, mounting and bearing limits. A motor that meets the translated torque is not proof that the gearbox can survive the duty. A transmission selected for steady operation may have a different starting or cyclic limit. Supplier service factors and duty classifications must retain their original definitions rather than being stacked with unexplained extra margins.

For a rigid ideal kinematic model, load inertia reflected to the motor shaft is J_load/i². Add the relevant motor and transmission inertia at the same reference shaft. This inertia transformation is not a complete transient loss model; retain the chosen loss treatment separately. Do not apply gearbox efficiency to kinetic energy in an undocumented way. [N2]

Atlas Copco's shaft-loading guidance treats combined radial and axial load as an independent selection issue. Accordingly, record pulley or sprocket geometry, load application distance and coupling arrangement. A permissible torque value cannot substitute for a bearing-load curve. [R6]

11. Establish the air conditions at the motor while it operates

Use dynamic inlet pressure at the motor connection, not the compressor-room setpoint. Record the simultaneous motor load and speed when measuring pressure so the value belongs to the actual working state.

Assess the lowest credible inlet condition for the required performance and the highest possible condition for equipment limits and overspeed response. A regulator setting at no flow is not a minimum dynamic supply. Conversely, a low pressure observed during an unresolved restriction should not automatically become a permanent justification for a larger compressor.

The exhaust boundary also matters. Long exhaust piping, silencers and shared exhaust arrangements can affect backpressure. Atlas Copco's installation guidance explicitly addresses the inlet and exhaust paths; its generic line dimensions and lubricant quantities are not treated here as universal design values. [R2]

The physical pressure difference across the motor is p_in,abs − p_out,abs. However, matching this difference alone does not prove equivalence to a supplier curve measured with atmospheric exhaust. Inlet density, expansion and flow can differ. Obtain a supported exhaust-pressure condition or an approved correction rather than substituting differential pressure into a gauge-pressure rating. [N2]

Include the relevant valve state, filter condition and temporary connections in the supply record. A motor accepted on a short workshop hose may perform differently after installation on a long moving connection. The selection and acceptance boundaries must represent the as-installed path.

12. Select pressure regulation, throttling and feedback deliberately

Pressure regulation and flow restriction are different ways of modifying performance. Neither should be chosen merely because a valve is already available. State whether the application prioritises starting torque, a speed band, torque limitation, consumption or a particular transient response, then obtain the corresponding motor and control data.

Atlas Copco's modification guidance distinguishes the effects of throttling and pressure regulation, including the tradeoff between retaining starting torque and reducing running speed. It also distinguishes inlet and exhaust throttling. Those observations are used here as selection questions; its pressure-correction table is not reproduced as a universal motor law. [R3]

Recheck starting after selecting a control arrangement. A candidate that starts on an unrestricted supply may fail with a different regulator setting or delayed pressurisation. Conversely, an exhaust restriction chosen to influence speed can alter backpressure and restart behaviour. The controlled performance evidence should identify the actual arrangement, not only the bare motor.

For a narrow speed band under changing load, specify measurement and regulation requirements. Record sensor location, speed resolution, response, permitted overshoot, settling time and how invalid feedback is handled.

Do not describe the teaching calculations as a tuning method. Controller gains, valve authority and pneumatic dynamics belong to the supported implementation and its tests.

13. Treat low-speed operation, loss of load and overspeed as different cases

The required low speed must be supported under load, not inferred from the fact that a throttled motor can rotate slowly during an unloaded demonstration. Ask whether starting repeatability, smoothness, speed regulation and lubrication remain suitable at the intended point. Where low-speed behaviour is uncertain, compare gearing or another motor architecture instead of simply restricting the inlet further.

A sudden reduction in process resistance can move the operating point toward a higher speed. Record the load-loss case separately from the normal working point. Include the driven machine's maximum permissible speed and the response required from governors, controls or other protective measures. A catalogue free-speed value is descriptive evidence, not universal permission to run unloaded. [N1]

This caution is material because supplier guidance differs by design. Ingersoll Rand's selection document expressly restricts unloaded running for the ungoverned motors it describes, while other manufacturer guidance discusses operation across a broader curve. The correct rule comes from the exact motor and assembly, not from combining favourable statements from different catalogues. [R5]

Where a broad speed range is required, define the torque and duty at several representative speeds. “Variable speed” is not a complete specification. Establish which points are continuous, temporary, start-only or prohibited. Specify whether the minimum operating point is stable indefinitely or only passed through during acceleration.

Replacement decisions should preserve these restrictions. A physically interchangeable motor with a different free speed, governor or gearing can change the load-loss consequence even when its normal power matches. Include that change in the mechanical and controls review.

14. Specify air preparation, exhaust and the installation environment

Record the selected motor's filtration, moisture, lubrication, ambient and material requirements. A lubrication-free variant is a configuration choice with its own performance and maintenance evidence, not a statement that every air stream or environment is suitable.

Define where exhaust is allowed to go and what it may carry. Oil mist, moisture and noise can matter to the surrounding process. Routing exhaust away from a sensitive area may be necessary, but that routing must be considered in the pressure and performance boundary. Do not solve contamination by creating an unexamined exhaust restriction. [N1]

Expansion cooling and environmental exposure should be included in the supplier enquiry where low temperatures, condensation or icing are credible concerns. Describe the actual environment rather than assigning a universal dew-point or temperature limit. Wet washdown, corrosive surroundings, abrasive contamination and outdoor storage also require configuration-specific evidence.

A motor powered by air is not automatically suitable for an explosive atmosphere. The complete assembly can include hot surfaces, mechanical contact, accessories and external ignition risks. The project must obtain the required suitability evidence for the actual atmosphere and installation. Do not infer approval from the absence of an electrical winding or from a generic product-family description.

Maintain safe access to the preparation assembly, isolation points and exhaust equipment. Maintenance personnel should be able to identify the approved filter, lubricant requirement and connection arrangement without relying on memory. Changing these components can change performance and should be recorded against the accepted configuration.

15. Check mounting, shaft loads and the driven-machine interface

Record shaft diameter and form, key or spline, usable engagement, coupling alignment, fastening and mounting orientation. The interface should transmit the required torque without imposing unsupported bending, axial loading or distortion. Procurement must preserve these attributes, not only motor power and port size. [N1]

For a belt, chain or gear attached directly to the output shaft, identify the forces and their distance from the bearing reference. A tension difference that creates torque is not the whole radial load: both belt or chain tensions may contribute to bearing loading. Use the actual mechanical arrangement and supplier load limits rather than estimating bearing suitability from transmitted torque alone.

A flexible coupling is not a universal remedy for misalignment. Its allowable angular, parallel and axial displacement, transmitted torque and speed must match the selected components. Mounting stiffness and alignment should be checked in the final machine condition, particularly where thermal movement, moving bases or variable external loads are involved.

Separate the motor's internal bearings from the support needed by an agitator, conveyor roller or process shaft. An overhung impeller or long shaft may need independent support even if the coupling can transmit torque. The duty register should name who owns that mechanical design and which drawings establish the accepted interface.

Inspection and maintenance access are part of selection. A compact motor located where the exhaust, lubrication point or mounting cannot be inspected may increase lifecycle difficulty. Record access, removal space, replacement identification and the acceptance checks needed after a motor or gearbox is replaced.

16. Include acceleration, stopping, reversal and retained energy

For a simplified rigid-shaft acceleration case, required torque is the process resistance plus J_equivalent × α, with all terms referred to one shaft. Angular acceleration is (ω_final−ω_initial)/t for a specified constant-acceleration teaching model. If motor torque or load changes materially with speed, the acceleration should be evaluated over the actual envelope rather than assuming one constant surplus. [N2]

Starting torque addresses initial breakaway; it does not automatically demonstrate adequate acceleration through the complete speed range. A motor may start the load but take too long to reach the productive speed because the available torque falls as speed increases. Conversely, a high transient torque requirement does not necessarily imply high continuous process power.

The rotating energy at speed is E_k = ½Jω². The stopping arrangement must deal with the energy and the required deceleration while respecting mechanical and thermal limits. Do not infer brake energy capacity from a holding-torque value. If an external process can drive the shaft during stopping, account for its contribution separately. [N2]

Reversal must have an authorised sequence and a supported motor, gearbox and driven mechanism. The ability to reverse a motor's direction does not establish permission to reverse an inertial load instantaneously. Specify whether a stop must be confirmed before reversal, the allowed torque transient and the feedback required.

Loss of air does not remove kinetic energy, gravity or external driving forces. Nor does air restoration authorise a new start. Isolation, retained pressure, braking and restart behaviour require the responsible machinery-design assessment. This guide provides the duty information for that assessment, not an executable protective circuit.

17. Match the selection to the process rather than to a generic application label

For mixing, record material states, impeller geometry, speed range, fill condition and startup history. Use measured or process-designer load evidence rather than assuming that all mixers follow one torque law. A speed adequate for one formulation may be unsuitable for another, and an empty-vessel condition may change the speed-control challenge.

For conveying or feeding, separate steady material resistance from breakaway, acceleration and possible plugging. State whether reversal is a permitted recovery action or a prohibited manoeuvre. Include the mechanical transmission and external bearing arrangement; an air-motor choice alone does not settle the conveyor design. [N1]

For winding, tension and speed can vary with effective radius. Translate the workpiece requirement to shaft torque and speed at the relevant radius, then examine the resulting envelope. Do not evaluate the empty reel for speed and the full reel for torque as though those maxima occur simultaneously without checking the actual process schedule.

For drilling, grinding or finishing integration, retain the complete machine's permitted speed and accessory conditions. A lower loaded speed can result from process resistance, supply restriction or motor condition. A motor-only selection must not override guarding or tool requirements. Where the task is an existing certified tool, preserve the exact approved configuration.

These application descriptions are requirement prompts, not substitute performance models. The useful output is an agreed duty set and a list of motor evidence needed to support it. When the process remains insufficiently defined, hold the final selection and resolve the application rather than adding arbitrary torque and airflow margins.

18. Build air demand and energy estimates from supported operating states

Use model-specific consumption at the relevant speed, torque, inlet pressure and exhaust condition. Preserve the flow reference pressure, temperature and moisture convention where specified. L/s, m³/min and cfm are units; they do not by themselves identify the gas reference state. Convert units only after the underlying quantity has been identified.

For a state-based teaching schedule with reference flow q_j in m³/min and duration t_j in minutes, V_cycle,ref = Σ(q_j × t_j). Divide by the full cycle duration to obtain cycle-average reference flow. State whether a duration includes productive running, unloaded enabled time, a stalled interval or an isolated stop. Missing state consumption is not zero. [N2]

For several motors, preserve actual overlap. Sum compatible flows during each defined event and separately integrate the full period. Multiplying each motor's already averaged consumption by another unexplained diversity factor can understate the demand. A high simultaneous-start case and an annual average answer different questions.

Shaft energy remains distinct from electrical energy used by the compressor station. A plant-energy estimate needs compatible electrical measurements or a justified system-specific model, treatment loads and the effect of controls. Do not divide shaft output by a guessed pneumatic efficiency, or multiply it by a universal compressor penalty, to present verified energy savings.

For a commercial comparison, use the actual duty and confirmed air-volume or energy basis alongside maintenance, controls and installation costs. A motor with lower purchase cost can require a different air path or operating regime. The selection record should show those differences without inventing current tariffs, quotations or supplier availability.

19. Compare candidates through independent decision rows

Use a torque–speed duty matrix rather than an overall score that allows low cost to cancel a technical shortfall. A candidate can meet normal running but fail starting, or satisfy both while its maximum-speed protection remains unresolved. Keep each result visible with the source and the condition to which it applies.

A nominal equality is not automatically a robust selection. The acceptance basis should explain treatment of tolerances, load variability and any margin. Record margin as a stated engineering provision, not as a hidden multiplier applied repeatedly to the same uncertainty. [N1]

A row-level PASS means that particular criterion has been supported. It must not imply complete machine approval. Where one mandatory criterion fails and another is missing, report both rather than letting one general status hide either issue. [S1; N1]

20. Worked example 1: normal power passes, starting and speed control do not

Fictional teaching case. A process needs 4.00 N·m at 1,500 rpm, with an operating-speed band of 1,425–1,575 rpm. Breakaway requires 7.00 N·m. All candidate values below are invented and assumed to refer to the same motor-output boundary, direction and defined dynamic air conditions. They are not real supplier data.

Required running shaft power is 2π × 1,500 × 4 / 60,000 = 0.628319 kW. This value alone does not choose a motor. Candidate A supplies 4.4 N·m at 1,500 rpm but has a stated minimum start of only 5.5 N·m. Its running torque margin at that speed is +0.4 N·m, while its starting margin is −1.5 N·m. A favourable running row therefore cannot support starting this load.

For the stated constant 4 N·m load and the fictional B curve, the original equilibrium is 1,800 rpm. That is 225 rpm above the permitted upper operating speed. At 1,500 rpm the +1 N·m surplus does not prove speed suitability; it drives further acceleration in the simple model.

The controlled B data create a nominal intersection at 1,500 rpm and preserve a stated start margin of 0.4 N·m. That is a candidate to evaluate, not a final selection. The remaining work includes pressure and load variation, permitted overshoot, mechanical limits and verification of the actual control arrangement. A control adjustment is acceptable only when its starting and running evidence is retained together. [N2]

21. Worked example 2: gearbox running efficiency cannot settle breakaway

Fictional teaching case. A driven shaft requires 60 N·m at 100 rpm and 90 N·m to break away. A 15:1 reduction is proposed. Assume a documented teaching running efficiency of 0.88 and, separately, an equivalent starting torque-transfer factor of 0.80. These values are assumptions for this example, not recommended defaults.

Motor running speed is 15 × 100 = 1,500 rpm. Required running motor torque is 60/(15 × 0.88) = 4.545455 N·m. Load power is 2π × 100 × 60/60,000 = 0.628319 kW. Required motor shaft power before gearbox losses is 0.628319/0.88 = 0.713998 kW.

Candidate C may cover the running curve yet fail breakaway through this transmission. Applying the running efficiency at startup produces the opposite conclusion, which demonstrates why the two states need separate evidence. Candidate D passes only the nominal start comparison; its running curve and every other duty row remain to be checked.

The 0.80 starting factor is a compact representation of the stated teaching assumption, not a universal gearbox law. A supplier may instead specify breakaway loss, output starting torque or another method. Use that method faithfully. For an integrated gearmotor with declared output starting torque, compare directly at the output and do not apply the gearbox again.

This example also excludes acceleration torque. Meeting breakaway means motion can begin under the stated nominal condition; it does not establish the time to reach 100 rpm. Add the relevant inertia and load-speed evidence before closing acceleration. [N2]

22. Worked example 3: a low continuous power hides demanding acceleration and braking

Fictional teaching case. A directly coupled drive has equivalent inertia J = 0.12 kg·m², constant resisting process torque of 3 N·m and a target speed of 1,200 rpm. The required acceleration time is 1.5 seconds. Assume a rigid shaft and constant acceleration for this screening calculation, with no omitted load or inertia.

Target angular speed is 2π × 1,200/60 = 125.663706 rad/s. Required angular acceleration is 125.663706/1.5 = 83.775804 rad/s². Inertial torque is 0.12 × 83.775804 = 10.053096 N·m. Total required accelerating torque is therefore 13.053096 N·m, although the steady process shaft power is only 0.376991 kW.

Suppose a candidate's available accelerating torque never exceeds 11.5 N·m over the relevant path. With the stated 3 N·m resistance, net torque is at most 8.5 N·m. Even the optimistic constant-surplus time is Jω/8.5 = 1.774076 seconds. The candidate therefore cannot meet the specified 1.5-second acceleration under this model. If the actual surplus falls lower, acceleration takes longer; the bound is not a prediction of its exact time.

Now specify a controlled stop from 1,200 rpm in 0.6 seconds with constant deceleration. Required total retarding torque is Jω/0.6 = 25.132741 N·m. If the 3 N·m process resistance is reliably present throughout the stop and assists braking, the separate brake contributes 22.132741 N·m. Without that evidence, the brake assessment cannot rely on the reduction.

Stored rotational energy is ½ × 0.12 × 125.663706² = 947.482023 J. Removing it in 0.6 seconds corresponds to average total retarding power of 1.579137 kW; constant retarding torque gives an initial total retarding power of 3.158273 kW. These values concern the complete retarding action, not automatically the brake alone. Repetition, heat dissipation, allowable torque and the approved stopping function remain separate checks. [N2]

23. Worked example 4: operating-state consumption versus shaft energy

Fictional teaching case. A motor repeats a six-minute cycle ten times per hour. All flow values below are assumed to be supported on the same reference-air basis. The stopped state is assumed to be verified isolated with zero consumption in this baseline; that is an explicit example condition, not a general assumption about stopped motors.

Cycle-average flow is 3.475/6 = 0.579167 m³/min. Ten cycles consume 34.75 reference m³/h. The startup flow remains 1.20 m³/min, more than twice the average. The branch must be assessed against the relevant simultaneous event; annual average flow does not demonstrate that startup pressure survives.

Suppose the productive point is 5 N·m at 1,200 rpm. Process shaft power is 0.628319 kW. Productive running occupies 30 minutes each hour, so the corresponding productive shaft energy is 0.314159 kWh per hour. This excludes acceleration and other mechanical states. It is not compressor electrical energy and cannot be divided by the air volume to establish a complete motor efficiency.

If later measurements show 0.08 m³/min during the supposedly isolated two-minute interval, the additional volume is 0.16 m³/cycle or 1.60 m³/h. The revised hourly consumption is 36.35 m³/h, and the revised average is 0.605833 m³/min. The increase is 4.604317% relative to the example baseline. Investigation must establish whether the flow is leakage, an intentional use, an isolation issue or a measurement problem.

For two identical machines, simultaneous startup would demand 2.40 m³/min on this basis. Staggering the starts might change the peak, but it does not automatically change the total cycle volume. Any sequence change must remain compatible with production and the approved circuit requirements. [N2]

24. Verify the candidate with the installed load and declared acceptance conditions

Prepare an acceptance plan before the trial. Identify the exact motor, gearbox, regulator, valves, hoses, silencers, lubricant arrangement, sensor points and driven equipment. State the required normal speed band, starting conditions, permitted acceleration time and other application criteria. Acceptance should refer to the configuration that will remain installed, not a temporary test arrangement that hides a restriction. [N1]

Measure speed with the actual productive load and record compatible torque evidence or its agreed process proxy. Measure dynamic inlet pressure and the relevant exhaust condition. Observe the required start cases, including direction and dwell conditions where they matter. Do not improvise a jam, overspeed or loaded reversal to gather data; abnormal-function verification needs an approved method and competent personnel.

Record command, motion onset, productive speed and completion as different events. A controller output timestamp does not prove shaft movement. A speed signal can confirm rotation without proving the process result. Include signal validity, sample timing and measurement limitations so a late or incomplete event is not misclassified as a successful cycle.

After acceptance, preserve the as-left settings and evidence. A change in motor variant, gearbox ratio, hose length, silencer, lubrication or software may invalidate part of the result. Retest the affected conditions rather than automatically rewriting the original baseline.

When performance deteriorates, compare the accepted configuration before increasing supply pressure. Check process resistance, inlet and exhaust conditions, mounting, transmission and motor condition. An investigation can identify an installation restriction or changed load without concluding that the motor was originally undersized. Keep the observed symptom, proposed cause and verified correction distinct.

25. Three editable engineering records

The following Word tables are manual templates. Replace the entry prompts with project information and add rows as needed. They do not automatically calculate, certify a motor or authorise operation. Maintain links to the actual evidence rather than pasting an isolated number without its conditions.

Record A — Driven duty and air-supply definition

Record B — Candidate torque–speed duty comparison

Record C — Trial, acceptance and controlled change

The record owner should distinguish complete evidence from completed paperwork. A candidate with every cell filled by assumptions remains provisional. Equally, an unresolved optional enhancement should not obscure a supported core finding; identify whether an open item is mandatory for the specific decision.

26. Engineering FAQs

Can an air motor be selected from kW or horsepower alone?

No. Power combines torque and speed, and the same power can represent very different duties. Define both at the driven shaft, translate any transmission and check the applicable motor curve. Starting, acceleration, supply conditions and mechanical limits remain separate.

Why does a motor run the load but fail to start it?

The required breakaway torque may exceed the available minimum starting torque even though the running point is covered. A different initial process condition, air path or gearbox starting resistance may also matter. Compare the actual starting requirement with the exact supported start condition instead of substituting stall torque or normal running torque. [S1; N1]

Does more torque at my target speed mean a better match?

Not automatically. A torque surplus can accelerate the load to a different speed. Check the load–motor intersection and the proposed speed-control method. More capacity may be useful for an identified transient, but it does not by itself establish steady-speed accuracy. [N2]

Is maximum power always at half of free speed?

That is a familiar characteristic of a simplified ungoverned vane-motor curve, not a universal selection rule. The applicable curve and control arrangement govern. Do not manufacture missing operating points from the half-speed relationship or transfer it to every piston, turbine, geared or governed motor. [R1, R5]

Can a pressure regulator provide precise speed control?

A regulator controls pressure within its supported behaviour; motor speed also depends on load and the full air path. A tight speed requirement needs an appropriate control and feedback specification. Recheck starting and transient response in the chosen arrangement.

Can a large silencer reduce motor performance?

Its physical size does not settle its flow behaviour. The exhaust arrangement can introduce backpressure and must match the motor's operating conditions. Select and verify the actual silencer and piping; do not remove necessary noise control merely to obtain a favourable speed reading. [R2; N1]

Does an air motor need lubricated air?

The exact motor variant governs. Record whether lubrication is required, permitted or incompatible with the application, and retain the specified air-quality and maintenance conditions. A lubrication-free label is not a universal statement about process cleanliness or all internal lubricants. [S1; N1]

Can the motor hold a suspended or externally driven load when air is removed?

That cannot be assumed from stall torque, gearing or a closed control valve. Holding, back-driving, retained energy and restart need a separately engineered solution appropriate to the machine. An ordinary motor-selection calculation is not a lifting or braking approval. [S3; N1]

Can I calculate compressor demand from shaft power?

Not reliably without supported motor consumption or an appropriate verified model. Use the exact operating-point flow, preserve its reference state and integrate the actual duty. Compressor electrical energy is a separate system-boundary assessment, not the shaft-power result.

What should be submitted for a motor-selection review?

Provide the driven duty and shaft, speed range, running and starting torque, load variability, inertia, duty schedule, gearing, direction, inlet and exhaust conditions, quality requirements, mounting loads and required control behaviour. Attach the candidate's current performance and installation documents. Identify estimates and unresolved requirements openly. [S1; N1]