Resolving Siemens MM420 USS Parity Error 3 on S7-200 Port 0

1. Overview: MM420 USS Communication with S7-200 Port 0

The Siemens MICROMASTER 420 (MM420) is a legacy 0.12 kW to 11 kW variable frequency drive that communicates over the Universal Serial Interface Protocol (USS) using a master-slave model. The SIMATIC S7-200 family, while predating the S7-1200, is still widely deployed and offers native USS support through Port 0 (RS485) and Port 1 (RS485/RS232), driven by the USS instruction library (USS_INIT, USS_CTRL, USS_RPM_x, USS_WPM_x).

A recurring field problem is the appearance of USS Error Code 3 for one or two PLC scans while the drive is running. The error coincides with the drive running bit clearing and the speed setpoint latching to zero, then both restoring on the next scan. From the operator's perspective this looks like a flicker, but on the wire it represents corrupted USS frames returning from the drive. The most common root cause in installations with cable runs below 3 metres is RS485 termination mismatch, combined with the default P2014[0] = 0 telegram-off-time behaviour and elevated 9600 baud signalling that exposes termination-induced reflections.

2. USS Protocol Technical Foundation

USS is a Siemens-proprietary serial protocol documented in the Siemens USS Protocol Reference. A USS telegram consists of:

• STX (0x02) start byte
• LGE length byte (telegram length excluding STX and BCC)
• ADR slave address (bit 5 = broadcast, bit 6 = mirror, bits 0-4 = address 0-31)
• PKW parameter channel (parameter ID, index, value) — variable or fixed length set by P2013
• PZD process data (control word, status word, setpoint, actual value) — length set by P2012
• BCC block check character (XOR of all bytes from ADR to last data byte)

Each transmitted byte carries 1 start bit, 8 data bits, an optional parity bit (USS default is even parity), and 1 stop bit, giving a 10-bit or 11-bit character. At 9600 baud one character takes 10 bits / 9600 bps = 1.0417 ms. A typical 14-byte telegram with default PZD = 2 and PKW = 4 occupies roughly 14 chars × 1.0417 ms ≈ 14.6 ms on the wire.

USS telegram time calculation

For an installation using N drives polled round-robin at baud B with average telegram length L bytes:

T_cycle (ms) = (L × N × (10 + parity)) / B × 1000

With L = 14, N = 1, B = 9600, parity enabled: T_cycle ≈ 14.58 ms. The drive will tolerate a telegram-off time (P2014) anywhere between zero (no monitoring) and a value larger than the worst observed inter-telegram gap. Setting P2014 below this value produces intermittent F0072 faults.

3. MM420 Hardware Interfaces and Wiring

The MM420 exposes two relevant serial paths:

• RS485 terminals 29 and 30 on the control board — terminals 29 = P+ (data+), 30 = N− (data−). This is the recommended path for USS integration with the S7-200.
• BOP / RS232 link via the front-panel port using the PC-to-drive RS232 cable. Used for commissioning and for single-master ASCII control, but is not the recommended path for S7-200 USS at 9600 baud with multi-drop because the RS232 driver is single-ended.

Wiring topology to S7-200 Port 0

The S7-200 Port 0 is a half-duplex RS485 port on the DB9 connector at pins 3 (B / data+) and 8 (A / data−), with pin 5 as logic ground and pin 1 as chassis/shield. The standard Siemens PC/PPI cable (6ES7 901-3CB30-0XA0) is not suitable for USS at 9600 baud; instead use a shielded two-wire twisted pair with the shield bonded to chassis ground at one end only (typically the PLC end).

Per the Siemens USS example configuration in TIA Portal, the shield must be tied to chassis ground or to pin 1 of the 9-pin connector, and the 2-0 V reference of the drive terminal block must also be bonded to chassis ground. Failure to bond both ends of the cable reference introduces common-mode voltages that the RS485 receiver interprets as parity errors.

Cable specification

4. S7-200 Port 0 and USS Library Architecture

The S7-200 USS library (instruction library version 2.0 and later) provides five sub-blocks. Only USS_INIT and USS_CTRL are required for closed-loop control; USS_RPM_W and USS_RPM_D are used for parameter read-back, USS_WPM_W and USS_WPM_D for parameter writes.

USS_INIT inputs

USS_CTRL inputs and outputs

The Error byte from USS_CTRL is the field that displays the transient 3 in the reported problem. When this byte becomes non-zero, the block internally latches the running bit and speed setpoint to 0 for one scan, which matches the user's symptom.

5. MM420 USS Parameter Map (P2010 – P2014)

The MM420 parameter set is firmware-dependent; the values below apply to firmware V1.5x through V3.2x (the range typically encountered in service). All parameters have an index [0] for the COM link on terminals 29/30.

6. USS_INIT and USS_CTRL Error Code Reference

The two error fields originate from different states of the communication stack:

7. Error 3 (Parity) Deep Dive: Causes and Symptoms

A parity error detected by the S7-200 serial port means that at least one of the ten or eleven bits per received character did not satisfy the parity rule (USS defaults to even parity). The MM420 generates an even-parity character; if the receiver sees an odd number of 1-bits, bit 8 of the character register triggers the parity interrupt. Common root causes:

1. RS485 termination mismatch. The cable impedance does not match the 120 Ω terminator at the receiver, causing the rising/falling edge of the stop bit to ring and cross the sampling threshold of the UART. RS485 UARTS sample the stop bit near the centre; ringing crosses that threshold and produces parity errors even though the data is otherwise intact.
2. Common-mode voltage exceeding RS485 receiver range. RS485 receivers tolerate ±7 V common mode; if the drive's 0V reference is not bonded to the PLC's 0V (pin 5 of DB9), and the cable shield is floating or bonded at both ends creating a ground loop, the differential signal rides on a DC offset that pushes the receiver input out of spec.
3. EMI from VFD output cables. The MM420 output cable carries dV/dt up to several kV/µs. If the USS cable runs parallel to motor cable without 200 mm separation, capacitive coupling injects common-mode noise that flips parity bits.
4. Excessive stub length or T-tap. USS daisy-chain stubs should be ≤ 1 m. Long stubs create impedance discontinuities.
5. Incorrect baud rate mismatch. P2010 = 9600 in the drive and USS_INIT Baud = 19200 in the PLC. Sampling at the wrong baud produces systematic parity errors, usually constant rather than one-scan.
6. Wrong parity setting. MM420 default is even parity; if the PLC is set to odd or no parity, every character fails. Typically constant, not transient.

8. The Transient Error Mechanism: One-Scan Anomaly

The error-3 flicker that drops the running bit and speed reference for exactly one scan is caused by the USS_CTRL block's internal handling of Error ≠ 0:

1. PLC polls drive N at scan T.
2. Drive response arrives but its parity bit is corrupted.
3. UART flags parity error, the PLC's USS driver increments the drive's Error byte to 3 and freezes Resp_R = 0, RUN/STOP internal echo = 0, Speed_SP internal echo = 0 for this scan.
4. The next scan polls drive N again; the corrupted bit was a transient reflection; this response is valid; Error clears to 0, Resp_R = 1, running bit and speed setpoint restore.

Because P2014[0] = 0 in the MM420 default, the drive never sees a telegram-off-time violation, so it does not trip F0072. This produces the benign-looking flicker rather than a hard fault, but it is still a physical-layer integrity problem that should be eliminated.

9. Cable Length, Termination, and Signal Integrity

RS485 termination follows the rule: install 120 Ω across the differential pair at the two physical ends of the bus, and nowhere else. The S7-200 CPU 224 / CPU 226 Port 0 has a built-in 120 Ω terminator activated by switch SW2 (DIP position 1 for Port 0, position 2 for Port 1). The MM420 RS485 terminals 29/30 have an internal pull-up/pull-down bias network but no integrated 120 Ω; termination on the drive side is provided either by an external resistor across 29–30 or by setting the bus terminator DIP switch on certain MM420 control board variants (early V1.x boards have no switch — install a discrete 120 Ω 1/4 W across 29/30).

The 3-metre rule

For cable lengths ≤ 3 m, the round-trip propagation delay of a typical PVC-jacketed 24 AWG twisted pair is roughly 2 × 3 m / (2 × 10^8 m/s × 0.66 c) ≈ 15 ns. At 9600 baud, the bit period is 104 µs. The reflection energy has decayed to under 5 % well before the UART samples the stop bit, so termination is functionally optional. However, both ends being terminated (S7-200 SW2 ON + drive-end 120 Ω) loads the differential pair with 60 Ω DC, which lowers the high-state voltage margin on some MM420 RS485 driver outputs. Removing one terminator — typically the drive-end one on cables ≤ 3 m — eliminates the loading while leaving the small reflection attenuated naturally by cable loss.

Termination rule of thumb

Inline topology diagram

10. Step-by-Step Diagnostic Procedure

1. Capture the scan-time log. In Micro/WIN, add a VW that latches the value of USS_CTRL Error at every scan with a 0.1 s timestamp. Confirm the error is 3 and that it appears for exactly one or two scans before clearing.
2. Verify parity and baud settings match. PLC USS_INIT Parity = 1 (even). Drive P2010[0] = 6 (9600) or 7 (19200). Document the values on the drive label.
3. Inspect the cable shield bonding. Shield at PLC DB9 pin 1 only; drive-end terminal 2-0 V bonded to chassis ground at one point. Floating shields, double-bonded shields, and un-bonded drive reference are the three most common defects.
4. Measure cable length. If < 3 m and both ends are terminated, remove the drive-end 120 Ω (open the MM420 control board DIP or extract the external resistor). This is the documented field cure for the reported fault.
5. Check stub length. If wiring is daisy-chained with T-taps, shorten each tap to < 1 m and ensure the main bus runs point-to-point between PLC and the last drive.
6. Check separation from motor cable. Minimum 200 mm parallel separation, cross at 90°. Use shielded VFD output cable with both ends of the motor-cable shield bonded to motor frame and drive PE.
7. Configure P2014[0] to a safe value. Set P2014[0] = 5 × T_cycle. For 9600 baud with one drive, P2014 = 75 ms. This produces an F0072 if the physical layer ever fails hard, converting a silent flicker into a visible fault.
8. Re-arm P0700[0] and P1000[0]. Set both to 5 (USS on COM link) and execute a power cycle on the drive so the parameters are non-volatile-saved.
9. Monitor continuously for 24 hours. Add a counter that increments every time Error is non-zero. Target: zero increments.

11. Telegram Timing and P2014 Off-Time Optimization

Setting P2014[0] = 0 (default) means the drive never times out the USS link, masking intermittent errors. Setting it too low causes nuisance F0072 faults. The minimum safe value depends on:

• Baud rate B (bps)
• Number of polled drives N
• Telegram length L bytes (PKW + PZD)
• PLC scan time T_scan (s)

T_cycle = (L × 10 × N) / B + 2 × T_scan

Recommended: P2014 = 5 × T_cycle, rounded up to the next 10 ms

Worked example for a 1-drive system at 9600 baud with PKW = 4, PZD = 2 (L = 14 bytes) and T_scan = 30 ms:

T_cycle = (14 × 10 × 1) / 9600 + 2 × 0.030 = 0.0146 + 0.060 = 0.0746 s ≈ 75 ms

P2014 = 5 × 75 ms = 375 ms, round up to 380 ms

For a 4-drive system at 9600 baud:

T_cycle = (14 × 10 × 4) / 9600 + 0.060 = 0.0583 + 0.060 = 0.118 s ≈ 120 ms

P2014 = 5 × 120 = 600 ms

12. Verification and Commissioning Checklist

13. Field-Tested Troubleshooting Matrix

14. Frequently Asked Questions