Short answer: the 50 A breaker in a servo drive manual is sized for capacitor-charging inrush, not for running current. A 1.8 kW servo drive on single-phase 240 V draws about 10 A at full load, then peaks at 10-30 times that for under 10 ms at power-up. Keep the manual's breaker and contactor ratings, or downsize by one rule: a D-curve breaker holds the drive only when 10 times its rating clears the drive's documented inrush peak. Work the numbers below once and your parts list is right the first time.
Why a 1.8 kW drive shows a 50 A breaker
A servo drive rectifies the 240 V line onto an internal DC bus, and that bus sits behind capacitors that arrive empty every time power is off. At power-up they look close to a short circuit until charged, so the input spikes to 10-30 times its running value and is over inside one half-cycle, 8.3 ms at 60 Hz. The charge held behind a rectifier is the same quantity behind how power supply bypass capacitors get correctly sized, and it is what your breaker sees at every power-up.
Running current for this drive is about 10 A, with the arithmetic in the example below, so the manual's rating is nearly five times the load. That margin is for the pulse, not the load: it keeps the magnetic trip above the inrush for every breaker the maker's tolerance allows. Treat the manual's figure as inrush-driven until the datasheet prints a peak you can plot.
What the trip curve does with a 10 ms pulse
A thermal-magnetic breaker carries two elements on different clocks. The bimetal needs seconds at overload, and a sub-half-cycle pulse never warms it. The magnetic element acts in milliseconds, and at power-up it is the whole game.
IEC 60898-1 defines each curve as a magnetic pickup band, not a single line:
| Curve | Magnetic pickup, x rating | Band on a 20 A breaker | Loads it fits |
|---|---|---|---|
| B | 3-5 | 60-100 A | Long cable runs, sensitive electronics |
| C | 5-10 | 100-200 A | Mixed loads, gently started motors |
| D | 10-20 | 200-400 A | Transformers, drive inputs, capacitive loads |
Below the lower edge the standard requires no magnetic trip; above the upper edge it requires a trip within a tenth of a second; in between, the individual breaker decides. So set the rating from the lower edge: 10x the rating on a D curve must sit above the drive's documented peak, and then no breaker in the tolerance band can nuisance-trip at power-up. One more rating belongs on the same datasheet pass: interrupting capacity. The 14 kA printed on a 35 mm DIN-rail miniature breaker must exceed the available fault current at your cabinet; on a small single-phase service that figure is often under 10 kA, so get it from the utility rather than guessing.
Checking a 20 A D-curve breaker against a 1.8 kW drive at 240 V
Inputs, and where each comes from:
- Mechanical power 1.8 kW, from the drive and motor rating.
- Supply 240 V single phase, the case the manual's breaker line covers.
- Drive efficiency 0.9 and input power factor 0.8 at full load, typical for this class; a printed datasheet current outranks the estimate.
- Peak inrush 10-30x input current for under 10 ms, the range small-drive makers publish; use the printed figure when you have it.
- Candidate breaker: 20 A, two-pole, D curve, 14 kA interrupting rating.
The rule in words: a pulse shorter than a half-cycle can only act on the magnetic element, so the guaranteed non-trip level, 10x the rating on a D curve, must sit above the peak.
The arithmetic:
- Input current: 1800 W / (240 V x 0.9 x 0.8) = 1800 / 172.8 = 10.4 A.
- Peak inrush: 10 x 10.4 = 104 A at the bottom of the range; 30 x 10.4 = 312 A at the top.
- Breaker band on a 20 A D: lower edge 10 x 20 = 200 A; upper edge 20 x 20 = 400 A.
- Compare: 104 A sits below 200 A, clean. 312 A lands inside 200-400 A, the tolerance band where some breakers trip and some hold.
- Smallest D rating whose lower edge clears 312 A: 312 / 10 = 31.2 A, so the next size, 32 A, with 10 x 32 = 320 A above the peak.
Decision, with its threshold: a documented peak at or below 200 A makes the 20 A D-curve breaker sound, protecting a 12 AWG copper branch. A peak near 300 A moves you to the 32 A D curve with 10 AWG copper, or to the manual's printed breaker with conductors sized to it. Pulses shorter than a half-cycle trip magnetic elements less readily than steady current, which is why a 30 A C curve, band 150-300 A, survives in some working installs, but the 10x floor holds at any pulse length, so design to it.
The contactor closes onto the inrush
A contactor is chosen by what it makes on closing, not only by what it carries. IEC 60947-4-1 sets the utilization categories: AC-1 for resistive loads, AC-3 for motors started direct-on-line, where the contactor is rated to make 5-7 times its operating current. A drive input is harder duty than a motor, because you close onto empty capacitors pulling 10-30x with no back-EMF to soften it.
Run the same arithmetic on a contactor quoted for the running load. An 18 A AC-3 unit on a 10.4 A drive makes 5 x 18 = 90 A to 7 x 18 = 126 A by category, against a pulse that reaches 312 A. Contacts bounce on that pulse, arc, and can weld shut, and a welded contactor stops answering its coil.
Give the drive the maker's minimum making rating for its supply: the class the manual shows on single-phase, with 32 A the proven floor on three-phase installs of the same drive, where the charge current divides across poles. Power the coil from the control side and keep the two supplies apart, as covered in isolating two power sources in one cabinet. The contactor is still not a disconnecting means: it can be re-energized remotely, and it is not rated to open a fault.
Feeder, stagger and the line filter
The breaker in your main panel, 20 cable feet away, can carry the drive branch protection when it meets the same rule: its 10x non-trip level above the drive's peak, and its interrupting rating above the available fault current. Cable length changes nothing in that physics. What distance changes is your disconnect, so fit a lockable disconnect at the cabinet or a plug and receptacle rated at the full branch current. Lock out the feeder and confirm zero voltage at the drive terminals before you land anything.
Several drives on one feeder follow the proven scheme in multi servo drive enclosure layout and sizing: a common feeder sized for the group, one D-curve breaker per drive, and the contactors closed one at a time with 0.5 s between them. Four drives closing together can stack four 300 A-class pulses on one feeder; 500 ms apart, about 30 line cycles, each pulse is finished before the next begins. Branching one feeder to each device is laid out in one power input for multiple devices inside an enclosure.
The line filter is the part that gets under-ordered. A 3 A EMI filter on a 10.4 A drive saturates on the inrush and runs hot at load. Order the filter by the drive's maximum input current, one standard size up, a 16 A class here, and leave the inrush duty to the breaker curve.
What to put on the order
| Part | Specify by | For the 1.8 kW, 240 V single-phase drive |
|---|---|---|
| Branch breaker | D curve, 10x rating above documented peak, interrupting rating above available fault current | 20 A D at a peak up to 200 A; 32 A D at 300 A; else the manual's printed size |
| Branch conductors | Copper sized to the breaker under the NEC ampacity rules | 12 AWG at 20 A; 10 AWG at 32 A |
| Input contactor | Maker's minimum making rating for the supply | The manual's single-phase class; 32 A floor on three-phase |
| EMI filter | Drive's maximum input current | 16 A class |
| Coil supply | 24 V DC from the control side | Kept separate from drive power |
| Enclosure | Rail layout, heat, door clearance | Type 12 indoor; 4X / IP66 for washdown |
The breaker, contactor and filter all mount on 35 mm DIN rail, so lay the rail before you order the box; the backplate and DIN rail layout guide walks the spacing. Budget the drive's own heat too: this class sheds 2-5 % of throughput, roughly 40-90 W at 1.8 kW, and the cabinet arithmetic is walked in single-drive enclosure cooling. On ratings: our Type 12 and 4X / IP66 enclosures are rated by the manufacturer, not UL listed, and if the panel needs a UL 508A listing, the 14 kA interrupting ratings of the branch breakers feed the panel's short-circuit current rating. Start with the enclosure selector for inside height and width, or send the cutout list for custom enclosures, built from one piece to your layout and delivered in 10 business days, counted from your approval of the drawing.
Enclosures for this job
Catalog sizes that suit the job this article answers, picked from its subject; open one for its full specifications.
| Enclosure | Material | Outside H x W x D / display | Rating |
|---|---|---|---|
| 60 × 36 × 16 in | 304 stainless | 60 x 36 x 16 in | Rated NEMA 4X / IP66 by the manufacturer; not UL listed |
| 48 × 36 × 16 in | 304 stainless | 48 x 36 x 16 in | Rated NEMA 4X / IP66 by the manufacturer; not UL listed |
| 60 × 36 × 12 in | 304 stainless | 60 x 36 x 12 in | Rated NEMA 4X / IP66 by the manufacturer; not UL listed |
Questions
Why does a 1.8 kW servo drive show a 50 A breaker in its manual?
Because inrush sets the number, not running current. The drive draws about 10 A at full load on 240 V single-phase, then charges its DC bus at 10-30 times that, 100 to 300 A, for under 10 ms. That printed rating keeps every breaker inside the maker's tolerance above the pulse, and many molded-case breakers also let you set the magnetic trip directly. Keep the printed size until the datasheet gives a peak you can plot.
Will a 20 A D-curve breaker hold a 1.8 kW servo drive at 240 V?
Yes, when the drive's documented peak sits at or below 200 A, which is 10 times the breaker rating and the guaranteed non-trip level of a D curve under IEC 60898-1. At a 300 A peak the pulse falls inside the 200-400 A band where individual breakers differ, so some trip at power-up. If it trips at power-up and never at load, the magnetic element is the culprit; the fix is a 32 A D curve, not thicker wire.
What contactor does a servo drive input need?
One that meets the drive maker's minimum making rating for its supply, the class the manual shows for this drive on single-phase. An 18 A AC-3 contactor sized to the 10 A running current makes about 90-126 A on closing by category, since IEC 60947-4-1 credits 5-7x, against a pulse that reaches 300 A; the contacts can weld. The contactor is also what lets you stagger power-up by 0.5 s per drive on a shared feeder.
Can the breaker in the main panel protect the drive instead of one in the cabinet?
Yes. Distance changes nothing in the curve rule: the main-panel breaker protects the drive branch when its 10x non-trip level clears the drive's documented peak and its interrupting rating clears the available fault current. You still need a local disconnect at the cabinet, either a lockable disconnect or a 250 V twist-lock plug and receptacle rated at the branch current, 20 A or 30 A classes for this drive.
How do I size the EMI line filter on a servo drive?
By the drive's maximum input current, never by the breaker size. This drive draws 10.4 A at full load, so a 3 A filter saturates on the inrush and runs hot at load; order a 16 A class, one standard size up. Mount it at the enclosure's line entry, ahead of the breaker and contactor, so the whole branch sits behind it.
Why stagger contactor closing when several drives share a feeder?
So the feeder sees one pulse at a time. Four drives closing together can stack four 300 A-class pulses on one feeder; closing each contactor 0.5 s apart, about 30 line cycles at 60 Hz, finishes each pulse before the next begins, and four drives reach full power about 1.5 s after the first closes.
