ABB ACS355 machinery drive (needs FMBA-01 RS485 adapter)

Register map(s) for this device, with a note of how far each has been checked. Addresses are 0-based, as on the wire. Scale and byte order are confirmed only at the hardware-verified rung.

coherence-checked

Round-trips through a real Modbus stack with no overlapping addresses and a word-order-sensitive codec — the map is internally coherent and wire-decodable. Does NOT prove byte-order or scaling match a real device: the emulator is seeded from the map's own types, so a uniformly wrong byte-order still round-trips. Only hardware-verified confirms byte-order/scale on the wire.

point	address (0-based)	type	scale
control_word	0	U16	1
status_word	3	U16	1
speed_and_dir	100	I16	1
speed	101	I16	1
output_frequency	102	I16	0.1
current	103	U16	0.1
torque	104	I16	0.1
power	105	I16	0.1
dc_bus_voltage	106	U16	1
output_voltage	108	U16	1
drive_temp	109	I16	0.1
kwh_counter	114	U16	1
analog_input_1	119	I16	0.1
analog_input_2	120	I16	0.1
analog_output_1	123	U16	0.1
motor_temp	144	I16	1
control_board_temp	149	I16	0.1

link settings as documented (unverified): RTU; default 9600 8N1; unit ID 1; FC 03,06,08,16,23; Supports Modbus RTU protocol. EIA-485 or RS-232 interface.

⚠ According to IEC 60664 and IEC 61800-5-1, the connection of the motor thermistor to the digital input requires double or reinforced insulation. — Starkest reads only; we never write a control register.

bench facts as documented (unverified)

terminals	X1 on FMBA-01 Modbus adapter or X2 on control panel
connector	RJ-45 for RS-232, Terminal block for EIA-485
shield/ground	Ground the shield in the cabling between the activation switch and the control board at the control board.
max registers/read	120
broadcast	supported
protocol notes	Parameter writes through standard Modbus are always volatile unless saved with parameter 1607 PARAM SAVE.
power	1-phase 200...240 V AC, 3-phase 200...240 V AC, or 3-phase 380...480 V AC
environment	Enclosure degree of protection: IP20
certifications	CE, C-UL US, RoHS, TÜV NORD, C-Tick
doc revision	3AUA0000066143 Rev E, EFFECTIVE: 2020-11-02

field reports — community-sourced, unverified (checked 2026-06-11)

Users frequently encounter off-by-one addressing errors when using standard Modbus libraries (such as ArduinoModbus or Unitronics) because the drive's documentation lists 1-based register addresses (e.g., 40101 for Speed/Direction), whereas protocol-layer clients require 0-based offsets (e.g., 100). [source]
If the RS-485 communication lines (A and B) are swapped, the drive's EFB OK MESSAGES parameter (5306) will not increment, while EFB CRC ERRORS (5307) and EFB UART ERRORS (5308) will continuously increase. [source]
confirmed: doc_revision: 3AUA0000066143 Rev E, EFFECTIVE: 2020-11-02 — confirmed by ABB Library document index [1.1.1]
confirmed: comms.baud_rates: [1200, 2400, 4800, 9600, 19200, 38400, 57600, 115200] — confirmed by parameter 5202 (BAUD RATE) in the ACS355 User's Manual
confirmed: comms.default_baud: 9600 — confirmed by parameter 5202 default setting of 9.6 kb/s
confirmed: comms.default_unit_id: 1 — confirmed by parameter 5201 (STATION ID) default setting of 1
confirmed: comms.framing: ['RTU'] — confirmed by FMBA-01 Modbus Adapter Module User's Manual stating it supports RTU protocol only
confirmed: comms.function_codes: [3, 6, 8, 16, 23] — confirmed by the 'Function Codes' section in the ACS355 User's Manual
confirmed: comms.parity: 'N' and stop_bits: 1 — confirmed by parameter 5203 (PARITY) default setting of '8 NONE 1'
confirmed: facts.certifications: 'CE, C-UL US, RoHS, TÜV NORD, C-Tick' — confirmed by the ABB and Baldor Drives Catalog

The trust ladder

Each rung states what it proved and what it did not. Gray until proven; green is earned by hardware only.

generated — A prompt produced it. Unproven — anyone can regenerate this.
structure-verified — Decode path is structurally sound (addresses, widths, finite values). Does NOT prove word order or scaling.
coherence-checked — Round-trips through a real Modbus stack with no overlapping addresses and a word-order-sensitive codec — the map is internally coherent and wire-decodable. Does NOT prove byte-order or scaling match a real device: the emulator is seeded from the map's own types, so a uniformly wrong byte-order still round-trips. Only hardware-verified confirms byte-order/scale on the wire.
hardware-verified (software-key) x1 — Matched a real device with operator-attested readings as ground truth — the only rung that confirms word order + scaling on the wire.

Device not here? Upload its manual and get a checked map back. It's free.

Want it confirmed against your own device? Order hardware verification. If it doesn't check out, you don't pay.

Building on a lot of devices? License the library — machine-readable maps, the firmware matrix, and API access.