Successfully Added
The product is added to your quote.
Nothing is sent to sales until you ask.
A VFD is sized by current, not horsepower. Take the full-load amps (FLA) stamped on the motor nameplate, add margin for the load's duty class, correct for ambient temperature and altitude, and then pick the smallest drive whose continuous output current rating at your duty class meets or exceeds that number. Horsepower labels are a shorthand that only holds true for a standard 4-pole NEMA Design B motor on a variable-torque load — the moment your load is a conveyor, a hoist, a compressor, or a rooftop unit in Arizona, the HP label starts lying to you.
Continuous drive amps needed = Motor FLA × Duty margin ÷ (Temperature factor × Altitude factor × Carrier factor)
Every VFD datasheet in the world leads with a horsepower or kilowatt number, and almost every one of them qualifies it in a footnote. The horsepower rating is a marketing convenience. What the drive's power stage can actually deliver is a continuous RMS output current at a stated ambient temperature, a stated carrier frequency, and a stated duty class. That current figure is the real rating. Everything else on the front page of the datasheet is derived from it by assuming a typical motor.
The problem is that "typical" is doing a lot of work in that sentence. A 25 HP motor at 460 V might draw 32 A or 34 A if it's a modern premium-efficiency NEMA Design B unit. The same nameplate horsepower on an older rewound motor, a high-slip design, an 8-pole low-speed motor, or an inverter-duty motor with a different power factor can pull 38 A or more. A 25 HP drive sized purely on the label may be running at 105% of its continuous rating the day it's commissioned — and it will pass a factory acceptance test and then trip out three months later on the first hot afternoon.
Reading amps also protects you from the two most expensive assumptions in drive selection. The first is that the motor is loaded to its nameplate. Frequently it isn't — a 40 HP motor pulling a measured 38 A on a 460 V system is doing roughly 30 HP of work, and you may not need a 40 HP drive at all. The second is that the motor is only loaded to its nameplate. On a jammed conveyor or a hoist catching a load, the drive sees far more than FLA, and the question is no longer "what's the continuous rating" but "how much overload current can this drive pass for how many seconds".
So the honest sizing question is never "what horsepower drive do I need?" It is: "what is the highest continuous current this drive will be asked to deliver, in the environment it will actually live in, and what peak does it need to survive on top of that?" Answer those two and the frame size falls out on its own.
Myth
"It's a 30 HP motor, so I need a 30 HP drive."
Reality
A 30 HP label maps to anywhere from 38 A to 44 A depending on motor design, and the drive's own 30 HP rating may only be valid at 40 °C, at a 2 kHz carrier, on Normal Duty. Change any one of those and the same drive is a 25 HP drive.
Myth
"The motor has a 1.15 service factor, so I get 15% free headroom."
Reality
Service factor is defined for sine-wave utility power. Inverter output adds harmonic heating and reduces cooling at low speed, which is why most motor manufacturers state that the service factor of an inverter-fed motor should be treated as 1.0. Size the drive above FLA on purpose — don't borrow capacity from a service factor that no longer exists.
Six fields on the nameplate decide the drive. Everything else is useful context.
1 — HP / kW. The mechanical output rating. Use it as a sanity check on your amp figure, not as the sizing input.
2 — Volts. Often shows a dual rating (230/460) for a nine-lead motor. Confirm which way the motor is actually wired in the peckerhead before you order anything — a motor connected for 230 V draws double the amps of the same motor connected for 460 V, and a drive sized off the wrong column will be exactly half the size it needs to be. This is the single most common sizing error we see on quotes.
3 — FLA / Amps. Full-load amps at the corresponding voltage. This is your sizing input. If two voltages are listed, two amp figures will be too — take the one matching your supply.
4 — RPM. Tells you the pole count, which tells you what to expect from the amps. A 4-pole motor is roughly 1750 RPM at 60 Hz, 6-pole roughly 1160, 8-pole roughly 870. Low-speed motors draw disproportionately high current for their horsepower and are a frequent source of "the drive is the right HP but keeps tripping".
5 — SF (Service Factor). Typically 1.00, 1.15, or 1.25. On a VFD-fed motor, treat it as 1.00 and take your headroom from the duty margin instead. See amps vs horsepower.
6 — Hz / Design / Insulation Class / "Inverter Duty". A Design B motor is the normal case. Design C and D motors have high starting torque and correspondingly high inrush. If the nameplate does not say inverter duty or reference NEMA MG 1 Part 31, note it — you may need an output filter, which we cover in harmonics and cable length.
Uploading a nameplate photo to the calculator above reads fields 1–5 automatically.
Nameplate unreadable, painted over, or the motor is inaccessible? Clamp an ammeter on one motor lead while the machine runs its heaviest normal cycle and record the highest steady reading. A measured amp figure beats a nameplate figure in every case — the nameplate tells you what the motor can draw, and the clamp tells you what your process actually demands. If you can't do either, our application engineers can work backward from the machine, the driven equipment, or the drive that's already in the panel.
Note the voltage and phase at the panel where the drive will land: 208, 230, 240, 380, 400, 460, 480, 575, or 600 V, and single- or three-phase input. Drive voltage classes are hard boundaries, not preferences — a 230 V class drive on a 480 V feeder is destroyed on power-up, and a 480 V class drive on a 208 V feeder will undervolt-fault or fail to produce rated torque. If your input is single-phase, jump ahead to voltage, phase and supply before you do anything else, because the derate is severe.
From the nameplate, matched to how the motor is wired. Where you have a clamp meter and a running machine, use the measured peak steady current instead.
Variable torque (fans, centrifugal pumps), constant torque (conveyors, positive-displacement pumps, extruders, compressors), or high inertia and high breakaway (hoists, crushers, centrifuges, mills, reciprocating loads). Duty margin by load type has the margin table.
Normal Duty gives you roughly 110% overload for 60 seconds. Heavy Duty gives roughly 150% for 60 seconds, and costs you frame size. Normal Duty vs Heavy Duty explains which one your load demands. Getting this wrong is the most expensive mistake on this page.
Ambient temperature above the drive's rated ambient, installation altitude above 3,300 ft (1,000 m), and any carrier frequency above the factory default all reduce the drive's continuous current. Derating has the factors and a worked combination.
Open the manufacturer's rating table — not the HP column on the marketing page — and find the smallest frame whose continuous output amps at your duty class equals or exceeds your derated requirement. If you land within about 5% of a breakpoint, go up. The price delta between adjacent frames is almost always smaller than one unplanned production stop.
Verify the drive's overload profile covers your worst-case transient (braking and inertia). Confirm the enclosure can shed the drive's heat (enclosure and panel heat). Check your motor lead length against the drive's limit and add a filter if you're over (harmonics and cable length). Then size input conductors and protection to code.
Nearly every modern VFD ships with two published current ratings, and the gap between them is wide enough to change your part number. The industry names vary — ABB uses I_N / I_Hd and light-overload versus heavy-overload, Yaskawa publishes Normal Duty and Heavy Duty, Schneider and Invertek describe it as standard versus high overload — but the concept is identical. The same power stage is certified for a higher continuous current if you promise not to demand much overload from it, and a lower continuous current if you need it to survive hard transients.
Ignoring duty class is how a correctly-calculated drive still ends up undersized. An engineer sizes a 40 HP drive for a 40 HP conveyor, the datasheet's 40 HP figure is the Normal Duty rating, the conveyor jams on a Monday morning and demands 180% torque, and the drive current-limits, ramps down, and trips. Nothing was wrong with the arithmetic. The wrong column was read.
Typically 110% of rated current for 60 seconds, sometimes 120% for 3 seconds. Written for loads whose torque falls with speed and that never see a shock: centrifugal fans, centrifugal pumps, blowers, cooling towers, rooftop units. On these loads breakaway torque is low and the drive never needs to push hard. ND is the higher continuous current of the two — you get more amps out of the same frame.
Typically 150% of rated current for 60 seconds, often 180–200% for 2–3 seconds. Written for loads that demand full torque at zero speed or that can shock the drivetrain: conveyors, hoists, crushers, mixers, positive-displacement and screw pumps, extruders, reciprocating compressors, mills. HD publishes a lower continuous current from the same hardware — the extra silicon is held in reserve for the transient.
| If your load is… | Duty class to size on | Why |
|---|---|---|
| Centrifugal fan, blower, cooling tower fan | Normal Duty | Torque falls as the cube of speed; breakaway is trivial |
| Centrifugal pump, circulating pump | Normal Duty | Same affinity-law behaviour; no shock loading |
| Rooftop unit / AHU supply and return fans | Normal Duty | Add margin for dirty filters and closed dampers, not for shock |
| Belt or chain conveyor | Heavy Duty | Full torque from standstill; jams and slugs are routine |
| Positive-displacement, progressive-cavity, or gear pump | Heavy Duty | Torque is flat versus speed; cold viscous product spikes it |
| Screw or reciprocating compressor | Heavy Duty | Constant torque plus cyclic pulsation |
| Mixer, agitator, extruder | Heavy Duty | High breakaway on settled or cured material |
| Hoist, crane, elevator, winch | Heavy Duty + braking | Overhauling loads, and you need torque before the brake lifts |
| Crusher, hammer mill, shredder | Heavy Duty, often oversized further | Uncontrolled shock loading from tramp material |
| Centrifuge, flywheel, large fan with high J | Heavy Duty | Long accelerations at current limit |
Rule of thumb: if the load can stall, jam, reverse-drive the motor, or has to break away under full load, size on Heavy Duty. If it can't, Normal Duty is not just acceptable, it's the economical choice — and on a large fan or pump it can save you an entire frame size.
Margin is not superstition and it is not a safety factor bolted on out of caution. Every percent you add is paying for a specific, identifiable current you expect the drive to deliver that the nameplate FLA doesn't account for: filter loading on a fan, cold product in a pump, a slug on a conveyor, the inrush of a clutch engaging. Decide what you're paying for and the number stops being arbitrary.
These are starting points, not rules. Where you have measured amps from the running machine rather than a nameplate figure, you can safely take the lower end of each band — the measurement has already absorbed most of the uncertainty the margin exists to cover. Where you're specifying a machine that doesn't exist yet, take the upper end.
| Load type | Examples | Margin on FLA | Duty class | What the margin is buying |
|---|---|---|---|---|
| Variable torque, clean | Centrifugal fans, cooling tower fans, centrifugal pumps in fixed systems | +10% | Normal Duty | Meter tolerance, voltage sag, motor-to-motor variation |
| Variable torque, real world | RTU and AHU fans, dust collectors, variable-system pumps | +15% | Normal Duty | Dirty filters, closed dampers, throttled valves, seasonal density change |
| Constant torque, steady | Belt conveyors under even load, mixers on thin product, machine tool spindles | +15% | Heavy Duty | Breakaway torque, belt tensioning, friction drift as bearings age |
| Constant torque, cyclic | Positive-displacement pumps, screw compressors, extruders, packaging machinery | +20% | Heavy Duty | Pulsation, viscosity change, cold starts, product density variation |
| High inertia, long accel | Large fans with heavy wheels, centrifuges, flywheel presses | +25% | Heavy Duty | Sustained operation at current limit through a long ramp |
| High breakaway / shock | Hoists, cranes, crushers, shredders, reciprocating loads | +25–35% | Heavy Duty + braking | Overhauling torque, tramp material, torque before brake release |
| Frequent start/stop | Anything cycling more than ~6 starts/hour | add +5% to the above | Heavy Duty | Thermal cycling of the IGBT junction and DC bus capacitors |
| Multiple motors on one drive | Parallel fan arrays, multi-pump skids | sum of all FLA, then +25% | Heavy Duty | No electronic motor overload protection is possible — each motor needs its own overload relay, and the drive cannot current-limit per motor |
Torque rises with the square of speed, power with the cube. Halve the speed of a centrifugal fan and it needs roughly a quarter of the torque and an eighth of the power. Breakaway is negligible, so the drive is never asked for more than rated current in normal operation. This is why Normal Duty exists and why it's safe here. It's also where the energy savings live: a fan at 80% speed draws about 51% of full power, which is the whole reason the drive is being bought.
Torque is flat across the speed range, power rises linearly. A conveyor moving the same load at half speed still needs the same torque; it just does half the work per second. The drive must produce full rated current at 10 Hz as readily as at 60 Hz, which is thermally harder on both drive and motor and is why these loads need Heavy Duty and often need a separately-powered blower on the motor for sustained low-speed running.
Torque demand is dominated by acceleration, not by the load. The governing term is T = J × (dω/dt), and on a large centrifuge or a heavy fan wheel that acceleration torque dwarfs the running torque. The drive spends the entire ramp sitting in current limit. Either accept a long ramp time, or size up so the ramp finishes before the drive's overload timer does. Long ramps are usually the cheaper answer.
A drive's published current rating comes with an asterisk, and the asterisk is the conditions it was tested under. Move the drive somewhere hotter, higher, or ask it to switch faster, and the same hardware safely delivers less current. Derating is not optional and it is not conservative engineering — it is the manufacturer telling you the rating is void outside the stated envelope.
Always start from your specific drive's rating conditions, because they differ by family. A typical open-chassis or IP20 unit is rated to 50 °C; put the same power stage in a NEMA 1 / IP21 kit and the rating usually drops to 40 °C because the vents are restricted.
| Ambient at the drive's air inlet | Typical effect on continuous current |
|---|---|
| At or below rated ambient (40 °C / 104 °F for most NEMA 1 units) | No derate — full rating |
| Each °C above rated ambient | −1% to −2% per °C, family-dependent |
| Above 50 °C / 122 °F | Many families are not rated at all; check for a high-ambient variant |
| Below 0 °C / 32 °F | No current derate, but check the minimum operating temperature and condensation rating — enclosure heaters are often required |
The trap here is that "ambient" means the air temperature at the drive's own intake, not the temperature of the room. A drive inside a sealed NEMA 12 enclosure in a 35 °C plant is not living at 35 °C — it is living at 35 °C plus whatever its own losses have added to the air inside that box. Enclosure and panel heat shows how to work that number out.
| Installation altitude | Typical effect |
|---|---|
| Up to 3,300 ft / 1,000 m | No derate |
| Each 330 ft / 100 m above that | −1% continuous current |
| Above 6,600 ft / 2,000 m | Current derate plus a voltage-class restriction on many families — insulation clearances and creepage distances no longer meet the original rating |
| Above 13,000 ft / 4,000 m | Generally outside standard product ratings entirely |
Two mechanisms are at work and they're worth understanding separately. Thin air carries away less heat, which costs you current. Thin air is also a worse dielectric, which costs you voltage withstand. That second effect is why a 480 V drive in Denver is fine but a 600 V drive at 8,000 ft may need to be specified differently.
Raising the carrier frequency makes the motor quieter and the current waveform smoother. It also makes the IGBTs switch more often, and every switching event is a fixed energy loss. Doubling the carrier roughly doubles switching losses.
| Carrier frequency | Typical effect on continuous current |
|---|---|
| Factory default (usually 2–4 kHz on larger frames, 4–8 kHz on small ones) | Full rating |
| ~8 kHz | −10% to −20% |
| ~12 kHz | −20% to −35% |
| 16 kHz and above | −35%+, and frequently limited to small frames only |
Most drives will silently reduce the carrier on their own when they get hot, which protects the drive but changes the motor's acoustic signature — a common and puzzling "the drive got noisy in August" complaint. If audible noise is a genuine requirement, size the drive for the carrier you intend to run, not the default.
Combined derating is multiplicative, not additive. Multiply the factors together; don't add the percentages.
Worked example — a rooftop supply fan in Albuquerque:
Motor FLA (460 V, 50 HP) 65.0 A Variable torque, real-world margin +15% 74.8 A required at ideal conditions Rooftop ambient, design peak 50 °C 10 °C above the drive's 40 °C rating at −1.5%/°C factor 0.85 Altitude 5,300 ft (1,615 m) 615 m above 1,000 m at −1%/100 m factor 0.94 Carrier left at factory default factor 1.00 Combined derating factor: 0.85 × 0.94 × 1.00 = 0.80 Continuous drive current required: 74.8 A ÷ 0.80 = 93.5 A
A 50 HP Normal Duty drive at 460 V is typically rated around 72–77 A continuous. This installation needs a 60 HP frame — roughly 88–96 A continuous at Normal Duty. Sized on the horsepower label alone, this job ships a full frame size too small, passes commissioning in April, and starts nuisance-tripping in July.
| Class | Nominal supply voltages | Typical DC bus | Common North American use |
|---|---|---|---|
| 200 V class | 200, 208, 220, 230, 240 V | ~325 VDC | Legacy plants, light commercial, small machine tools |
| 400 V class | 380, 400, 415, 440 V | ~565 VDC | Imported OEM machinery, European-built equipment |
| 480 V class | 440, 460, 480 V | ~650-680 VDC | The default for US industrial |
| 600 V class | 575, 600, 690 V | ~815 VDC | Canadian industrial, long feeders, mining, large HP |
Note that 460 V and 480 V drives are usually the same product with a different label, and that many 400 V class drives will accept 480 V — but many is not all, and the ones that won't fail immediately and without warranty. Confirm the input voltage range on the specific datasheet, not the class name.
Higher voltage is cheaper to distribute: the same horsepower at 575 V draws roughly 80% of the current it does at 460 V, which means smaller conductors, smaller breakers, and less voltage drop over a long run. On any job with a feeder over a few hundred feet, this is worth a conversation before the panel is built.
This comes up constantly on small pumps, dust collectors, machine tools and light commercial work — and it is where most self-service sizing goes wrong.
Many three-phase drives will accept single-phase input, but only at a substantially reduced rating. The reason is straightforward: the input rectifier and the DC bus capacitors were designed for the smooth six-pulse ripple of a three-phase feed. On single-phase, only two of the three input legs are energised — four of the six diodes carry the entire load while two sit idle — and the bus ripple roughly triples, so both the conducting diodes and the bus capacitors run far hotter for the same output power.
Practical guidance: derate a three-phase drive to approximately 50% of its nameplate output current on single-phase input, expect to require a specific parameter change to disable input phase-loss detection, and expect a line reactor to be either recommended or mandatory. Confirm all three on the datasheet before ordering — some families explicitly forbid single-phase input, and using them that way voids the warranty and typically kills the drive within weeks.
Where single-phase supply is the permanent reality, a purpose-built single-phase-input drive is nearly always the better buy. It's rated honestly, needs no derate, and costs less than the double-size three-phase unit it replaces. Our team can identify which families in stock are rated for it.
The amp figures below are the NEC Table 430.250 full-load current values for three-phase AC induction motors. Two important caveats before you use them:
| HP | kW (approx) | FLA @ 208 V | FLA @ 230 V | FLA @ 460 V | FLA @ 575 V | Typical drive continuous A @ 460 V, ND | Typical @ 460 V, HD |
|---|---|---|---|---|---|---|---|
| 1 | 0.75 | 4.6 | 4.2 | 2.1 | 1.7 | 2.1–2.5 | 1.7–2.1 |
| 1.5 | 1.1 | 6.6 | 6.0 | 3.0 | 2.4 | 3.0–3.5 | 2.4–3.0 |
| 2 | 1.5 | 7.5 | 6.8 | 3.4 | 2.7 | 4.0–4.8 | 3.4–4.0 |
| 3 | 2.2 | 10.6 | 9.6 | 4.8 | 3.9 | 5.6–6.0 | 4.8–5.6 |
| 5 | 3.7 | 16.7 | 15.2 | 7.6 | 6.1 | 9.0–9.6 | 7.6–8.0 |
| 7.5 | 5.5 | 24.2 | 22 | 11 | 9 | 12–14 | 11–12 |
| 10 | 7.5 | 30.8 | 28 | 14 | 11 | 17–18 | 14–17 |
| 15 | 11 | 46.2 | 42 | 21 | 17 | 24–27 | 21–24 |
| 20 | 15 | 59.4 | 54 | 27 | 22 | 31–34 | 27–31 |
| 25 | 18.5 | 74.8 | 68 | 34 | 27 | 38–41 | 34–38 |
| 30 | 22 | 88 | 80 | 40 | 32 | 45–48 | 40–45 |
| 40 | 30 | 114 | 104 | 52 | 41 | 60–65 | 52–60 |
| 50 | 37 | 143 | 130 | 65 | 52 | 72–77 | 65–72 |
| 60 | 45 | 169 | 154 | 77 | 62 | 88–96 | 77–88 |
| 75 | 55 | 211 | 192 | 96 | 77 | 105–124 | 96–105 |
| 100 | 75 | 273 | 248 | 124 | 99 | 145–156 | 124–145 |
| 125 | 90 | 343 | 312 | 156 | 125 | 170–180 | 156–170 |
| 150 | 110 | 396 | 360 | 180 | 144 | 208–240 | 180–208 |
| 200 | 150 | 528 | 480 | 240 | 192 | 260–302 | 240–260 |
| 250 | 185 | — | 604 | 302 | 242 | 320–361 | 302–320 |
| 300 | 220 | — | 722 | 361 | 289 | 390–414 | 361–390 |
| 400 | 300 | — | 954 | 477 | 382 | 515–590 | 477–515 |
| 500 | 375 | — | 1180 | 590 | 472 | 650–720 | 590–650 |
Reading the table: find your HP row, take the FLA column matching your supply voltage, apply your margin and derating from margin and derating, then move right and choose the drive column matching your duty class. If your requirement lands between two rows, take the higher one.
Application: variable-air-volume supply fan on a packaged rooftop unit, Albuquerque NM. 50 HP, 460 V, 3-phase, NEMA 1 drive in a rooftop-mounted enclosure with solar gain.
Nameplate FLA @ 460 V 65 A Load class: variable torque, real-world +15% 74.8 A Duty class: Normal Duty (110%/60 s is ample) Design ambient at drive inlet 50 °C → factor 0.85 Altitude 5,300 ft → factor 0.94 Carrier: factory default → factor 1.00 Required continuous current: 74.8 ÷ (0.85 × 0.94) = 93.5 A
Result: a 60 HP frame, not a 50 HP frame. An HVAC-dedicated family is the right fit here — it ships with the fan and pump control macros, PID, real-time clock scheduling and BACnet/Modbus already in the firmware, which saves a day of commissioning.
What we'd quote from stock: an ABB ACH580 or ACH550, or a comparable Invertek HVAC unit. → HVAC Drives · ABB VFDs
Application: progressive-cavity sludge pump, indoor lift station, 230 V, 3-phase. Product viscosity varies with temperature and the pump must break away against a settled column.
Nameplate FLA @ 230 V 54 A Load class: constant torque, cyclic +20% 64.8 A Duty class: Heavy Duty (150%/60 s needed for breakaway) Ambient 40 °C, indoor, at drive rating → factor 1.00 Altitude: sea level → factor 1.00 Carrier raised to 8 kHz for noise → factor 0.85 Required continuous current: 64.8 ÷ 0.85 = 76.2 A
Result: a 30 HP frame rated at Heavy Duty, not a 20 HP unit. Note how much the carrier frequency choice cost — nearly a whole frame size bought purely for acoustic comfort. If the pump room isn't occupied, leaving the carrier at default saves the customer real money.
Also specify: sensorless vector or closed-loop vector control, not V/Hz. A PC pump needs full torque at 3–5 Hz and V/Hz control won't deliver it reliably.
Application: a loaded belt conveyor running downhill from a surge bin, 460 V, 3-phase, drive mounted in a NEMA 4X enclosure outdoors. Once product is on the belt, gravity drives the load faster than the motor wants to turn — the motor is generating for most of the run.
This one is instructive because the sizing calculation says one thing and the application says another.
Nameplate FLA @ 460 V 40 A Load class: constant torque, shock +30% 52 A Duty class: Heavy Duty (150%/60 s for a fully loaded restart) Ambient 45 °C inside a sealed 4X box in sun → factor 0.93 Altitude: sea level → factor 1.00 Carrier: factory default → factor 1.00 Required continuous current: 52 ÷ 0.93 = 55.9 A
Result: a 40 HP frame at Heavy Duty — but the amp calculation is the easy half. Three things it doesn't show:
The same reasoning applies to any load that reverse-drives the motor — hoists and cranes, centrifuge coast-downs, unwind stands, test dynamometers. Size the power stage on amps, then size the braking path on duty cycle, which is a separate calculation entirely (braking).
What we'd quote from stock: a Heavy Duty general-purpose drive plus a regenerative front end. → Regenerative Drives
Application: shop dust collector, 230 V single-phase service only. Customer wants soft start and speed control.
Nameplate FLA @ 230 V, 3-phase motor 22 A Load class: variable torque, real-world +15% 25.3 A Duty class: Normal Duty → 25.3 A required at output Single-phase input derate on a 3-phase drive × 0.50 Equivalent three-phase drive rating needed: 25.3 ÷ 0.50 = 50.6 A
Result: sized this way, the job needs a 20 HP-class three-phase drive to run a 7.5 HP motor. At 230 V, 50.6 A lands between the 15 HP figure (42 A) and the 20 HP figure (54 A), so 20 HP is the frame — nearly three ratings above the motor. That is exactly why this scenario gets quoted wrong so often.
The right answer is a drive family explicitly rated for single-phase input at 7.5 HP. It carries no derate, needs no phase-loss workaround, and costs a fraction of the oversized alternative. If no single-phase-rated unit is available in the required rating, the next option is a rotary phase converter feeding a normally-sized three-phase drive — a conversation worth having with an engineer, not a spec sheet.
Application: a 15 HP drive on a mixer has failed. The model is discontinued and the motor nameplate is illegible under 20 years of product buildup. Nothing can be measured because the line is down.
This is the most common sizing job we actually handle, and no calculator can do it from a horsepower figure alone. The method is different: you size from the drive that's there, then verify against the machine.
1. Read the OLD DRIVE's nameplate, not the motor's. Record: output amps, output voltage, input voltage/phase, HP/kW rating, and the duty class if stated. 2. Take the old drive's rated CONTINUOUS OUTPUT AMPS as your floor. Whoever commissioned this machine sized it for the load; if it ran for 20 years, the rating was adequate. 3. Check the OVERLOAD SETTING in the old drive's parameters if it still powers up, or in the machine documentation. A motor overload parameter set to 34 A tells you the motor's real FLA is 34 A (check the units — some families enter this as a percentage of drive rating) — the number you couldn't read off the nameplate. 4. Check the FAULT HISTORY. Repeated overcurrent, overload, or overtemperature faults mean the original was UNDERSIZED. Do not replicate it — go up a frame. 5. Confirm the physical envelope: frame dimensions, conduit entry, keypad cutout, and whether the replacement can reuse the existing backplate and wireway. 6. Confirm the control interface: analog in/out counts, relay and digital I/O, and the fieldbus the PLC expects (Modbus RTU/TCP, EtherNet/IP, PROFINET, BACnet). A drive that fits the panel but can't talk to the PLC has not solved the problem.
Then check one thing the old drive can't tell you: whether the application changed. Machines get sped up, product formulations get thicker, and duty cycles get more aggressive over twenty years. If the mixer is running a heavier product than it did in 2005, the original rating is the wrong target.
This is what we do all day. If your drive is obsolete, discontinued, or the manufacturer has gone end-of-life, we hold surplus and refurbished stock for legacy platforms and can cross-reference to a current equivalent. → New, Refurbished, or Repaired? · Allen-Bradley PowerFlex Alternatives
An induction motor becomes a generator any time the load turns it faster than the drive's commanded frequency. That happens more often than people expect: lowering a hoist, decelerating a large fan wheel, a downhill conveyor, a centrifuge coasting down, a vehicle test dynamometer, any web tension unwind. The energy flows backward into the drive's DC bus, the bus voltage climbs, and at the trip threshold the drive faults out — which on a hoist means the load is now held only by the mechanical brake.
Sizing a drive for these applications means sizing two things: the power stage for the motoring current, and a path for the regenerated energy.
| Method | How it works | Best for | Trade-off |
|---|---|---|---|
| Extended ramp / DC bus ride-through | Decelerate slowly enough that losses absorb the energy | Light inertia, no time pressure, no overhauling load | Free, but slow — and useless on a hoist, where the load overhauls continuously |
| Dynamic braking (DB) resistor | A chopper transistor dumps bus energy into a resistor as heat | Occasional braking, short cycles, hoists, most crane duty | Wastes the energy, adds heat to the room or roof, resistor must be duty-rated |
| Regenerative / line-regen unit | Inverts bus energy back onto the AC line | Frequent or continuous braking: test stands, centrifuges, high-duty cranes, downhill conveyors | Higher capital cost, more panel space, needs harmonic consideration |
| Active front end (AFE) | Fully bidirectional IGBT rectifier — regenerates and controls harmonics | High-power, high-duty-cycle, harmonic-sensitive sites | Most expensive; often the only viable answer above ~100 HP with continuous regen |
| Common DC bus | Multiple drives share a bus; braking axes feed motoring axes | Multi-axis machines where some axes brake while others accelerate | Design-intensive; requires matched family and careful bus sizing |
Two numbers matter and they're often confused. The resistance determines peak braking torque and must not fall below the drive's stated minimum, or the chopper transistor is destroyed. The wattage determines how long you can brake for and is set by duty cycle, not by peak.
The peak braking power is roughly the mechanical power you're absorbing. The continuous wattage rating you need is that peak multiplied by the fraction of each cycle spent braking:
Average braking power ≈ kinetic energy absorbed ÷ deceleration time
= (0.5 × J × (ω₁² − ω₂²)) ÷ t_decel
Peak braking power ≈ 2 × average, for a linear deceleration ramp
Resistor W rating ≈ Average braking power × duty cycle fraction
(e.g. braking 6 s in every 60 s → × 0.10)
Resistance value ≥ the drive's minimum permitted ohms (never below)
≈ V_bus_trip² ÷ Peak braking power
Note that the resistance value comes from the peak and the wattage from the average — confusing the two is how chopper transistors die. For an overhauling load — a hoist lowering continuously, or a decline conveyor — the duty cycle fraction is effectively 1.0 and the resistor must be rated for full continuous braking power. This is usually the point where a regenerative unit becomes the cheaper and cooler option. It's also where installers get caught out: a resistor sized for 10% duty on a hoist that spends its shift lowering will glow, then open, then leave the drive with nowhere to put the energy.
On a large fan wheel, a centrifuge, or a flywheel, the acceleration torque swamps the load torque. The drive sits at current limit for the whole ramp, and the sizing question becomes a race between the ramp time and the drive's overload timer. A drive rated 150% for 60 seconds that needs 150% for 90 seconds will trip on overload every single start.
You have two levers. Lengthen the ramp so the current demand falls below the overload threshold — free, and usually acceptable on equipment that starts twice a day. Or size the drive up so its continuous rating covers the accelerating current — necessary when the process demands a fast start, or when the machine cycles frequently enough that thermal accumulation matters. Where starts are frequent and inertia is high, budget for both a larger frame and a braking solution, and expect to specify a motor with a separately-powered cooling blower.
A correctly sized drive can still cause problems on both sides of itself. On the input side it's a non-linear load that distorts the supply. On the output side it produces a PWM waveform with rise times fast enough to damage motor insulation. Neither shows up in a horsepower calculation, and both are cheaper to solve at the quote stage than after commissioning.
A standard six-pulse VFD draws current in pulses rather than sinusoidally, generating characteristic 5th, 7th, 11th and 13th harmonics. IEEE 519 is the reference standard for acceptable limits at the point of common coupling. For buses at or below 1 kV it sets a voltage limit of 8% THD, with 5% on any individual harmonic; from 1 kV to 69 kV the THD limit tightens to 5%. Allowable current distortion (TDD) is not a single number — it depends on the ratio of available short-circuit current to maximum demand load current at the PCC, so it has to be calculated for your specific service. Utilities and large facility owners increasingly write it into specifications, and it's the standard your commissioning engineer will be measuring against.
| Mitigation | Typical line current THD | When it's the right answer |
|---|---|---|
| Six-pulse drive, no reactor or DC choke | ~60–80% | Only defensible on very small drives on a stiff supply |
| 3% line reactor | ~30–40% | Baseline good practice on almost any drive; also buys transient protection |
| 5% line reactor | ~25–35% | Where notching or capacitor switching is a known problem |
| DC link choke | ~30–40% | Built into many drives already — check before adding a reactor |
| Passive harmonic filter | ~5–8% | Single large drive that must meet a spec; cheaper than AFE |
| 12- or 18-pulse drive | ~10% / ~5% | Large drives, dedicated transformer available |
| Active harmonic filter | <5% | Multiple drives on one bus; corrects the whole bus, not one drive |
| Active front end drive | <5% | Harmonic compliance and regeneration needed together |
A useful screening rule: if the total connected drive load exceeds roughly 20–25% of the supply transformer's kVA, harmonics deserve a real study rather than a rule of thumb. Below that, a 3% line reactor per drive is usually sufficient and worth fitting regardless — it also protects the drive from line transients and reduces DC bus capacitor stress, which extends the drive's life.
The drive's output is a train of steep-edged voltage pulses. A long motor cable behaves as a transmission line, and where the cable's surge impedance doesn't match the motor's, the pulse reflects at the motor terminals and superimposes on the incoming pulse. The result can be a peak voltage approaching twice the DC bus voltage — on a 480 V system, transient peaks approaching 1,360 V at the motor terminals.
NEMA MG 1 Part 31 defines the inverter-duty motor insulation standard. The standard expresses the requirement as 3.1 x rated line-to-line volts with a rise time no faster than 0.1 µs — about 1,430 V for a 460 V motor and 1,490 V for a 480 V one, which is why vendor literature usually rounds it to a 1,600 V class limit. A motor built to that standard has real margin. An older motor, or a rewound one, or any general-purpose motor without that rating, may not — and the failure mode is insulation breakdown in the first turns of the winding, months or years after commissioning, presenting as a phase-to-phase or turn-to-turn short that looks like a random motor failure.
| Motor lead length (480 V class) | Typical requirement |
|---|---|
| Under ~50 ft / 15 m | Generally no filter needed |
| ~50–100 ft / 15–30 m | Usually fine with an inverter-duty motor; add a filter for general-purpose or rewound motors |
| ~100–300 ft / 30–90 m | dV/dt filter or load reactor recommended on essentially all installations |
| ~300–1,000 ft / 90–300 m | Sine-wave output filter strongly recommended; also check for drive overcurrent trips from cable charging current |
| Over ~1,000 ft / 300 m | Engineered solution required; consider relocating the drive closer to the motor |
Always check the specific drive's published maximum cable length, because it varies widely by family and by carrier frequency — higher carrier frequencies shorten the permissible run considerably. Two secondary effects also bite on long runs: the cable's distributed capacitance draws charging current that the drive counts as output current (so a long run effectively derates the drive), and the same fast edges drive shaft currents through the motor bearings, causing fluting and premature bearing failure. On motors above roughly 100 HP, or on any long-lead installation, an insulated bearing or a shaft grounding ring is cheap compared to a bearing replacement.
Use shielded VFD cable, terminate the shield 360° at both ends, keep motor cables physically separated from control and signal wiring, and cross them at right angles where you must cross. Where the installation must meet an EMC standard — IEC 61800-3 is the drive-specific one — confirm whether the drive has an integral EMC filter and which environment category (industrial vs residential/commercial) it satisfies. Retrofitting an external EMC filter after a failed emissions test is significantly more expensive than specifying the right internal one.
| Rating | Protects against | Typical VFD use |
|---|---|---|
| Open / chassis (IP20) | Nothing — for mounting inside another enclosure | Multi-drive control panels, MCC buckets |
| NEMA 1 / IP21 | Falling dirt, incidental contact | Clean indoor electrical rooms |
| NEMA 12 / IP54–55 | Dust, dripping non-corrosive liquid | General plant floor, sealed against airborne dust |
| NEMA 3R / IP24 | Rain, sleet, external ice | Outdoor, non-washdown |
| NEMA 4 / IP56 | Hose-directed water | Washdown areas, food and beverage |
| NEMA 4X / IP66 | Hose-directed water plus corrosion | Wastewater, chemical, coastal, marine |
Enclosure choice feeds straight back into sizing, and in two directions at once. A sealed NEMA 12 or 4X box cannot exchange air with the room, so all the drive's losses accumulate inside it — which raises the drive's real ambient and therefore its derate. Meanwhile the drive itself is usually rated to a lower ambient in a NEMA 1 kit than as an open chassis, because the vent kit restricts airflow. It's common for a drive to need a frame size up purely because of how it's being enclosed.
A VFD is roughly 95–97% efficient, so budget 3–5% of rated output power as heat. Use 4% for planning.
Drive heat output (W) ≈ HP × 746 × 0.04
Worked example — 50 HP drive in a sealed NEMA 12 enclosure:
Heat load = 50 × 746 × 0.04 ≈ 1,490 W
Plus the contactor, PLC, power supply
and control transformer in the box ≈ 200 W
Total in-box heat load ≈ 1,690 W
Rise above room ambient with no active cooling depends on the
enclosure's radiating surface area. A 60 × 36 × 12 in box has
4.27 m² of total surface area — and only about 2.9 m² of that
radiates usefully, because the back face sits flat against a wall.
At a typical 5.5 W/m²/°C for painted steel in still air:
Counting all six faces:
Rise ≈ 1,690 ÷ (4.27 × 5.5) ≈ 72 °C
Excluding the wall-mounted back (the realistic case):
Rise ≈ 1,690 ÷ (2.90 × 5.5) ≈ 106 °C
Room at 30 °C → in-box ambient of 100 °C or more
That drive will not survive. Options, cheapest first: vent the enclosure with filtered fans if the environment allows it (loses the NEMA 12 rating unless the fans are rated too), fit a closed-loop air conditioner or air-to-air heat exchanger sized to the calculated watt load, move to a larger enclosure to increase radiating area, or mount the drive's heatsink through the back panel so the losses dissipate outside the box — a through-the-wall or "flange mount" option many families offer, and often the most elegant answer.
The panel's Short-Circuit Current Rating must equal or exceed the available fault current at its point of installation, and it is determined by the weakest component in the power circuit, not by the drive alone. Many drives have a modest standalone SCCR (5 kA is common on small frames) that rises substantially — often to 65 kA or 100 kA — only when protected by a specific class and rating of fuse or circuit breaker named in the manual. Get the branch protection wrong and the panel's SCCR collapses to the drive's bare figure, which is a documented inspection failure and a genuine arc-flash hazard.
Ask two questions on every job: what is the available fault current at this panel, and exactly which fuse or breaker does the drive manual require to reach the SCCR I need?
VFD circuits are covered by NEC Article 430, Part X (Adjustable-Speed Drive Systems). The essentials:
Respect the manufacturer's clearances above, below and beside the drive — they exist because the heatsink is convection-assisted and blocking the flow triggers the same derate as a hot room. Mount vertically unless the datasheet specifically permits otherwise. Where drives are stacked vertically in a panel, the upper drive lives in the lower drive's exhaust, and its ambient must be calculated accordingly or a baffle fitted.
Most drives that "fail" were never faulty. They were asked to do something the rating didn't cover. This table works backward from what you're seeing on the keypad to the sizing decision that caused it.
| What you're seeing | Likely sizing root cause | What to check | Fix |
|---|---|---|---|
| Overcurrent trip on start, every start | Duty class wrong — sized on Normal Duty for a constant-torque load | Compare load type against duty class; check the drive's overload rating | Re-size on Heavy Duty, or extend the accel ramp if the process allows |
| Overload trip after 30–60 s of running | Continuous rating too low, or motor FLA parameter set below actual | Clamp the running amps and compare to the drive's continuous rating | Frame up, or correct the FLA parameter if it was mis-entered |
| Drive overtemperature, worse in summer or afternoons | Ambient derate not applied, or enclosure cooling undersized | Measure air temperature at the drive's own intake, not the room | Add enclosure cooling, or frame up to absorb the derate (derating, enclosure) |
| Overtemperature at low speed only, motor also hot | Constant-torque load run below the motor's self-cooling range | Check running frequency; a shaft-mounted fan is useless below ~20 Hz | Add a separately-powered motor blower, or an inverter-duty motor |
| DC bus overvoltage on deceleration or lowering | No braking path for regenerated energy | Check whether the load overhauls or has high inertia | Add DB resistor, or a regen unit for continuous overhauling (braking) |
| Nuisance trips only at certain times of day | Supply-side issue: capacitor bank switching, voltage imbalance, or a big load sharing the bus | Log voltage and imbalance across a full shift | Add a 3% line reactor; investigate the supply |
| Motor runs but won't produce rated torque; undervoltage warnings | Wrong voltage class, or motor wired for the other voltage on a dual-rated nameplate | Verify actual supply voltage and how the motor is connected | Correct the motor connection or replace with the correct voltage class |
| Ground fault trips with no insulation fault found | Long motor leads; cable charging current read as ground current | Measure lead length; check for shielded cable terminated at one end only | Add output reactor or dV/dt filter; terminate shield 360° both ends (cable length) |
| Random motor winding failures 1–3 years in | Reflected-wave overvoltage on a non-inverter-duty motor | Check lead length and whether the motor meets NEMA MG 1 Part 31 | Add dV/dt or sine filter; specify inverter-duty on replacement |
| Bearing failure, fluting visible on races | Shaft currents from PWM common-mode voltage | Common above ~100 HP and on long leads | Insulated bearing or shaft grounding ring; improve bonding |
| Output current higher than expected at same load | Carrier frequency raised without derating, or motor is not the design assumed | Check carrier parameter against the derate table | Return carrier to default, or frame up (derating) |
| Drive was fine for years, now trips regularly | The application changed, not the drive | Compare current running amps to the original commissioning record | Re-size for the load as it exists today, not as it was specified |
| Won't start at all, no fault | Usually not sizing — control wiring, enable, or parameter lock | Work through the control chain before condemning the drive | See our field guide: VFD Won't Start? 9 Causes to Check Before You Replace the Drive |
A 30 HP motor gets a 30 HP drive and the job is considered done. It works often enough to feel safe and fails often enough to be expensive.
Instead: size on nameplate amps at your duty class, and use the horsepower label only to check that your answer is in the right neighbourhood.
A 230/460 V motor lists two amp figures. Take the 460 V figure for a motor actually wired at 230 V and the drive is exactly half the size it needs to be.
Instead: confirm how the motor is physically connected before ordering. If you can't get into the peckerhead, clamp the running amps.
The nameplate says 1.15 and it feels like free capacity.
Instead: treat an inverter-fed motor as SF 1.0. Inverter harmonics add heating that the service factor was never defined to cover.
The datasheet's headline HP figure is almost always the Normal Duty rating, and it's the number that ends up on the quote.
Instead: find the Heavy Duty column for any load that can stall, jam, or must break away under load.
Room temperature gets used as ambient, and the enclosure is forgotten entirely.
Instead: derate from the air temperature at the drive's own intake, which in a sealed enclosure is the room plus the drive's own losses. Run the calculation in enclosure and panel heat.
The drive is sized correctly and installed 250 feet from the motor with no output filter.
Instead: check the lead length against the drive's published limit at your carrier frequency, and budget for a dV/dt or sine filter above roughly 100 ft on general-purpose motors.
A hoist, a downhill conveyor, or a big fan wheel gets a correctly-sized drive and no braking path.
Instead: any load that can overhaul the motor or that stores significant kinetic energy needs a DB resistor or a regenerative front end, sized on duty cycle rather than on peak.
Two frames up feels prudent. It isn't free: a heavily oversized drive has poor current resolution at low load, degrades the motor overload protection's accuracy, draws more magnetising current than necessary, costs more, and takes more panel space.
Instead: one frame of deliberate margin, justified by a number. Three frames of vague caution is a different mistake, not the absence of one.
We stock both new and tested refurbished drives across the major platforms, with a 24-month warranty on every part and repair and same-day shipping on in-stock items from Raleigh, NC.
Fans, pumps, cooling towers, AHUs and rooftop units — variable-torque loads where Normal Duty sizing applies and the firmware macros save commissioning time.
ABB, Invertek, LS Electric, Schneider Electric, Yaskawa
Deep stock across the ACH and ACS families, including the ACH580 and ACH550 for HVAC and the ACS355 for machinery. Well-documented rating tables with clear ND/HD columns.
Three tiers that map cleanly onto the load classes in this page: Eco for variable-torque building services, E3 for general machinery, P2 for high-performance and high-overload duty.
The Altivar range across machine, process and building applications. Common on OEM machinery and in retrofit work where the PLC already speaks Modbus.
For continuous overhauling loads and harmonic-sensitive sites: energy goes back onto the line instead of into a resistor bank.
Where the application needs precise position or torque control rather than speed control, the sizing method in this page doesn't apply — servo sizing works from a load inertia and torque profile.
Servo Drives & Amplifiers · Spindle Drives · Yaskawa Servo Drives Field Guide
Obsolete, discontinued or end-of-life? We hold surplus and refurbished inventory for legacy platforms and have technicians in-house who test and repair. If the drive you need hasn't been manufactured in a decade, that's a normal Tuesday for us. → New, Refurbished, or Repaired?
Send us the motor nameplate, the load, and the ambient — a photo of the nameplate is enough — and one of our application engineers will confirm the frame size and quote from stock. There's no charge and nothing goes to sales unless you ask for a quote.
Sometimes, and that's what makes it dangerous. On a clean variable-torque load — a centrifugal fan or pump in a temperate indoor space at low altitude — matching horsepower usually lands you in the right frame. On a constant-torque load, in a hot or high location, or with a motor whose amps run above the typical figure for its rating, matching horsepower produces an undersized drive that passes commissioning and fails later. Take the extra two minutes and size on amps.
Almost always one of four things: you're comparing against the Normal Duty rating when your load needs Heavy Duty; the drive's ambient is higher than you think because it's in a sealed enclosure; the carrier frequency was raised without derating; or the motor draws more amps than its horsepower label suggests. Symptom to root cause walks the fault code back to the cause.
Measure if you can. The nameplate tells you what the motor can draw at full load; a clamp meter on the heaviest normal cycle tells you what your process actually demands. Where the measured figure is well below nameplate you may be able to size down a frame, and where it's at or above nameplate you've just found a problem worth knowing about before you buy anything.
No. Service factor is defined for sinusoidal utility power. Inverter output introduces harmonic heating and, on a shaft-mounted fan, reduced cooling at low speed. Most motor manufacturers advise treating an inverter-fed motor as service factor 1.0. Take your headroom from a deliberate duty margin instead, where you can see it and justify it.
Two certified ratings for the same hardware. Normal Duty publishes a higher continuous current with a modest overload allowance, typically 110% for 60 seconds — appropriate for loads that never shock the drive. Heavy Duty publishes a lower continuous current with a much larger overload allowance, typically 150% for 60 seconds — appropriate for anything that must break away under load or can jam. The published horsepower figure on a marketing page is nearly always the Normal Duty one.
Roughly 1–2% of continuous current for every °C above the drive's rated ambient, which for a typical NEMA 1 unit is 40 °C. Ten degrees over costs you 10–20% of the drive's capacity — enough to move you a full frame size on a marginal selection. The number that matters is the air temperature at the drive's own intake, which inside a sealed panel can be dramatically higher than the room.
Yes. Derating starts above 3,300 ft (1,000 m) at roughly 1% of continuous current per additional 330 ft (100 m). Denver at about 5,300 ft costs you around 6%; a mountain site at 8,000 ft costs around 14% and may also restrict the voltage class you can use, because thin air is a worse insulator as well as a worse coolant.
Many families allow it, but at roughly 50% of the nameplate output current, usually with a mandatory line reactor and a parameter change to disable input phase-loss detection. That means a 7.5 HP job at 230 V needs roughly a 20 HP three-phase drive. If single-phase supply is permanent, a drive explicitly rated for single-phase input is cheaper and correct. Some families forbid single-phase input outright — check the datasheet, because doing it anyway voids the warranty.
Electrically yes, with the drive sized for the sum of all motor FLA plus about 25%. But the drive can only provide overload protection for the aggregate, not for each motor, so every motor needs its own overload relay. You also lose independent speed control and you can't use sensorless vector control — the drive has to run in V/Hz mode. It's a reasonable approach for identical parallel fans; it's a poor one for dissimilar loads.
The resistance value must not fall below the minimum ohms the drive's manual permits, or you destroy the chopper transistor. The wattage rating is set by duty cycle: peak braking power multiplied by the fraction of each cycle you spend braking. Brake for 6 seconds in every 60 and you need about 10% of peak as a continuous rating. For a continuously overhauling load like a hoist lowering, that fraction is effectively 1.0 and a regenerative unit is usually cheaper than the resistor bank it replaces.
It depends on the drive family and the carrier frequency, so check the datasheet — but as a planning guide on a 480 V system: under 50 ft is generally fine, 50–100 ft is fine with an inverter-duty motor, 100–300 ft wants a dV/dt filter or load reactor, and beyond 300 ft wants a sine-wave output filter. Long runs also draw cable charging current that counts against the drive's output rating, so a very long lead effectively derates the drive.
It can, if the motor's insulation isn't built for it. PWM output produces voltage peaks that can approach twice the DC bus voltage at the motor terminals, and on long leads the reflected wave makes it worse. Motors built to NEMA MG 1 Part 31 are designed for this; older, general-purpose, or rewound motors may not be. Above roughly 100 HP, or on long leads, also plan for shaft-current protection — an insulated bearing or a shaft grounding ring — to avoid bearing fluting.
No. One frame of deliberate, justified margin is good engineering. Two or three frames of vague caution degrades the accuracy of the drive's electronic motor overload protection, worsens current resolution at low load, costs more, and takes panel space you may not have. Oversizing is a different mistake from undersizing, not the absence of one.
Size from the old drive rather than the motor. Its rated continuous output amps are your floor, its motor-overload parameter tells you the motor's real FLA, and its fault history tells you whether the original was adequate. Then confirm the physical envelope and the fieldbus the PLC expects. Example 5 walks the full procedure — and it's the job our team handles most often.
No. Servo and spindle drives are sized from a load inertia ratio and a torque-versus-time profile, not from a continuous FLA figure — a completely different calculation. For those, start with our servo drive and spindle drive collections, or talk to an engineer.
On a centrifugal fan or pump, substantially — power varies roughly with the cube of speed, so running at 80% speed draws about half the power. On a constant-torque load like a conveyor, the savings are modest because the torque demand doesn't fall with speed; the benefits there are process control, soft starting, and reduced mechanical wear rather than energy. If someone quotes you affinity-law savings on a conveyor, ask them to show the working.
Always size and install to the edition of the code adopted in your jurisdiction and to the specific drive manufacturer's manual, which takes precedence over any general guidance on this page.