ABB ACS880 vs Delta MS300: Sizing by Real Watts – Where the Ratio Breaks
How much real motor power can a VFD actually deliver without nuisance tripping or overheating—and when does the nameplate ampacity become the wrong number to use? That’s the sizing question that separates a drive that runs for a decade from one that cuts out every Tuesday afternoon.
I’m going to tear down the sizing logic for two drives: the ABB ACS880 (0.55–1300 kW, Direct Torque Control) and the Delta VFD MS300 (compact to ~5.5 kW, sensorless vector). Both are solid products, but the ratio of their rated output to real motor watts tells a very different story depending on load type and overload duty. The argument here is about magnitude proportion—how many percentage points of margin you actually need versus what each drive leaves you.
1. Overload Capability – The Ratio That Decides The Rest
The MS300 is dual-rated: 120% for 60 s in Normal Duty (ND) and 150% for 60 s in Heavy Duty (HD). The ACS580/880 platform (we use the ACS880 for industrial parity) offers 110% for 1 min every 5 min as standard, with a 150% starting torque capability via DTC. That looks close until you map it to real watts.
Mechanism: A centrifugal pump at full speed draws roughly its nameplate power. But a pump that’s partially clogged or running slightly above its BEP can pull 115–125% of nameplate current for extended periods. The MS300’s 120% ND rating means it can sustain that overload for 60 seconds, then it must drop back to 100% or trip. If the clog persists for 90 seconds—common in sludge handling—you get a fault. The ACS880’s 110% overload is lower, but its DTC loop can deliver 150% torque momentarily for starting; for continuous overload, the drive relies on the I²t model and a larger thermal mass in the IGBT module. In practice, the ACS880 at 110% can sustain a 10% overload indefinitely if ambient stays below 40 °C (illustrative, ~roughly), because the thermal time constant of the frame is longer than the MS300’s compact chassis.
Worked consequence: For a 3.7 kW (5 hp) pump motor with actual absorbed power of 4.0 kW (about 108% of nameplate), the MS300 would need to be upsized to at least the next frame (5.5 kW / 7.5 hp) to keep the overload ratio below its 120% ceiling for continuous operation. The ACS880 4 kW frame handles 4.0 kW at 110% load = 4.4 kW thermal limit—so the same motor runs without upsizing. That’s roughly a 15–20% cost difference on the drive alone, plus panel space.
When this reverses: If your load never exceeds 100%—clean water pump, well-maintained conveyor—the MS300’s 120% headroom is more than enough, and its compact size and lower base cost win. The ACS880’s higher continuous margin is wasted.
2. Real Watts vs. Apparent Power – The Deception of Motor Nameplate Amps
Motor nameplates list full-load amps (FLA), but VFDs are rated for output amps at a given voltage. The real watts delivered = √3 × V × A × power factor × efficiency. A typical NEMA Design B motor at full load has a PF of about 0.82–0.87. A VFD can improve that PF at the line side, but at the motor terminals the PF stays roughly the same unless you enable flux optimization.
Mechanism: The ACS880’s DTC can actively adjust the magnetizing current to maintain optimal rotor flux, raising motor PF to ~0.92–0.95 under load (illustrative, based on DTC literature). The MS300’s sensorless vector control also improves PF, but without direct torque feedback the correction is less aggressive—typical improvement to about 0.88–0.90. For a 3.7 kW motor drawing 8.5 A at 460 V, the real power at PF 0.85 is 3.7 kW. At PF 0.93, the same amps deliver 4.05 kW—about 9% more real work from the same current.
Worked consequence: A 5 hp pump that actually needs 4.2 kW would require 9.7 A at PF 0.85, but only 8.9 A at PF 0.93. The MS300, rated for 9.5 A continuous (roughly, 3.7 kW frame), would be at 102% of its rated current—over the edge. The ACS880, with the same current rating and DTC-driven PF improvement, can deliver 4.2 kW within its 9.5 A limit. That’s the difference between a drive that runs at 98% load and one that trips on overcurrent.
When this reverses: If you size the drive one frame up (e.g., 5.5 kW MS300 for a 3.7 kW motor), the current headroom is so large that the PF difference becomes noise. The MS300 in a larger frame is still cheaper than the ACS880 in the same size.
3. Starting Torque Ratio – Not Just “150%” But How Long
Both drives claim high starting torque: the ACS880 delivers up to ~150% starting torque and full torque at zero speed via DTC; the MS300 in Heavy Duty mode can supply 150% for 60 s. But these are different ratios applied to different time windows.
Mechanism: DTC calculates torque 40,000 times per second and can hold full rated torque at 0 Hz indefinitely (within thermal limits). That matters for applications like extruders or mixers that need breakaway torque from a dead stop. The MS300’s sensorless vector control can achieve 150% torque at low speed, but the control loop updates slower and relies on a current model—above 3–5 Hz it’s fine, but at 0 Hz the torque drops off rapidly (illustrative, typical of sensorless vector without encoder).
Worked consequence: A conveyor with a sticky belt that requires 140% torque to break free: the ACS880 delivers that torque at zero speed and holds it while the belt starts moving. The MS300 might deliver 140% at 2–3 Hz, but the belt doesn’t move until the motor reaches that speed—so the drive may fault on current limit trying to accelerate into the load. The ratio of real starting torque to rated current is better on the DTC drive by about 20–30% at near-zero speed (roughly, based on published torque curves).
When this reverses: Most centrifugal pumps and fans need only 20–40% starting torque. The MS300’s 150% for 60 s is overkill; the DTC advantage adds no value. For a pump start, you pay for what you don’t use.
The ACS880’s larger frame (IP21/IP55, chassis up to 1300 kW) has a higher thermal time constant than the MS300’s compact molded case. That means the ABB VFD drive can absorb a 120% overload for 2–3 minutes before the IGBT junction hits the limit, even though the datasheet says 110% for 1 minute. The MS300’s compact design sheds heat faster but also saturates its thermal mass in about 60 seconds. So the real overload capability ratio is not 110% vs 120%—it’s about 110% for 3 minutes vs 120% for 1 minute. For loads that exceed nameplate for 90–120 seconds (clogged pump, cold oil startup), the ABB drive wins even though its nominal overload number is lower.
Consider a 3.7 kW fan in a dust-laden environment with a dirty filter. The motor current gradually rises to 110% over 2 minutes. The MS300 (120% ND) holds for 60 s, then trips at 110%—right when the filter is about to clear. The ACS880 (110% continuous margin) never trips; the thermal mass absorbs the slow ramp. But if the ambient temperature is 50 °C, the ACS880’s thermal margin shrinks to about 105% (derate ~1%/°C above 40 °C, illustrative). Now both drives are marginal. The MS300 can be moved to a cooler location more easily because of its small size; the ACS880’s larger chassis may not fit. So the failure mode flips at high ambient.
| Dimension | ABB ACS880 (host) | Delta MS300 (rival) | What changes at the motor |
|---|---|---|---|
| Continuous overload margin (thermal) | 110% for 1 min; effectively ~110% for 2–3 min due to larger frame | 120% ND / 150% HD for 60 s; compact thermal mass saturates | Clogged pump: ACS880 runs, MS300 may trip after 70 s |
| Real watts from same current (PF improvement) | ~0.93 PF under DTC (illustrative) → 9% more real power for same amps | ~0.88 PF (illustrative) → 3.5% more real power | 4.2 kW motor: ACS880 fits 3.7 kW frame; MS300 needs 5.5 kW frame |
| Stalled torque at 0 Hz | Full torque at 0 Hz, up to 150% | Torque drops below ~3 Hz; 150% for 60 s only at speed | Sticky conveyor: ACS880 breaks away; MS300 may stall |
| Thermal mass time constant (relative) | High – larger chassis, longer I²t absorption | Low – compact, quick thermal saturation | Slow ramp load: ACS880 absorbs; MS300 trips sooner |
Topology/standards per the cited standards; all product ratings are manufacturer-stated values from the cited datasheets, current to 2026-06; derived/illustrative figures are labelled as such. This is not an independent head-to-head test. ABB is a brand affiliated with this site; competitor names are used for identification only.
