Troubleshooting Siemens MM420 USS Bus Crash: BOP P---- Lockup

1. Overview: The Dual-MICROMASTER 420 USS Lockup Problem

Two Siemens MICROMASTER 420 (MM420) drives, frame sizes 3 kW (6SE6420-2UD17-5AA1 / 2UC15-5AA1 class) and 7.5 kW (6SE6420-2UD23-0BA1 / 2UC23-0BA1 class) installed side by side in the same electrical enclosure, drive two pumps. Individually, both drives run cleanly. When the larger 7.5 kW MM420 is energized, the smaller 3 kW MM420 occasionally locks up:

• The Basic Operator Panel (BOP) freezes on P----.
• USS (Universal Serial Interface Protocol) RS-485 communication goes silent — no response from the affected drive.
• Clearing the fault requires a hard power-off / power-on cycle of the affected drive; the fault is not latched, not stored in the fault buffer, and is not visible through the BOP fault history.

Because the BOP shows no error code, no fault number, and no alarm string, the symptom is misleading. The drive is not reporting a hardware fault — it is reporting that its parameter/state machine has been corrupted by an external event, almost always a high-energy electrical disturbance on the USS bus or on the control wiring. This article is a field-tested diagnostic and remediation guide for this exact class of failure, with concrete checks, parameter references, and an EMC-aware installation procedure.

2. What "P----" on the BOP Actually Means on a MM420

The MM420 BOP has a limited set of fixed display patterns. Knowing the pattern matters because it tells you which subsystem is wedged.

The P---- pattern is not a documented fault code in the MM420 operating instructions; it is an undocumented state that appears when the parameter index register or the BOP handler pointer is corrupted by an electrical event. When this pattern shows, the BOP key handler stops accepting input, and the USS port stops accepting new requests because the firmware is no longer servicing interrupts. A power cycle clears it because the corrupted register is re-initialized from the parameter image in EEPROM on next boot.

3. Why a Second MM420 on the Same USS Bus Causes the Smaller Drive to Crash

Three mutually reinforcing mechanisms produce the symptom. The smallest drive (3 kW MM420) is the victim in essentially every reported case because its 230 V class or 400 V class control board has lower noise margin than the larger 7.5 kW unit's stiffer gate-drive and bulk DC bus.

3.1 Inrush and DC-Bus Charge on the Larger Drive

When the 7.5 kW MM420 is energized, its rectifier stage charges the bulk capacitors. For the first 100–500 ms after switch-on, the input rectifier draws high peak current through the same line-side conductors feeding the 3 kW drive. The resulting common-mode voltage step appears on the protective earth (PE) of the cabinet and, more importantly, on the 0 V reference of any low-level signal cable that is referenced to the cabinet PE at the PLC end and the drive end. The USS bus uses a 0 V reference conductor; a step on that conductor is interpreted as a bit transition by the RS-485 receivers.

3.2 Radiated Switching Noise from the IGBT Stage

Once the 7.5 kW MM420 runs, its IGBTs switch at typically 4–16 kHz (default 8 kHz on MM420, parameter P1800). The fast dv/dt at the motor terminals radiates and conducts high-frequency noise back along the motor cable. If the motor cable is unshielded, not properly terminated at the gland, or run in the same tray as the USS cable, the noise is capacitively and inductively coupled into the USS pair.

3.3 RS-485 Bus Fault Tolerance

USS is a half-duplex RS-485 multi-drop protocol. The MM420 USS interface is implemented on the control board's RS-485 transceiver (typically a half-duplex 75176-class device). RS-485 has excellent common-mode rejection within spec, but the spec assumes a properly terminated, properly biased, single-segment bus. Once you add the 7.5 kW drive's noise environment to a bus that is missing one or more of:

• 120 Ω termination at both ends
• Fail-safe bias resistors on the master end
• Shield bonded to PE at one end (or both ends with care)
• Adequate physical separation from power cables

the receiver at the 3 kW drive intermittently sees a string of garbage bytes. The MM420 firmware is designed to discard noise on the USS port, but a sustained burst during a parameter write — especially a write initiated by the BOP hand-shake or by the master PLC — can corrupt the parameter index in volatile RAM. Once corrupted, the firmware loops waiting for an index that does not exist, and the BOP renders that as P----.

4. EMC: The Dominant Root Cause in Cabinets with Mixed MM420 Sizes

Siemens publishes an EMC-compliant installation guideline for the MICROMASTER 4 family. The relevant rules are summarized below. The user's installation fails one or more of them in essentially every documented case of dual-Micromaster USS lockup.

For a deeper look at the Siemens EMC installation guideline applicable to the MICROMASTER 4 family, see the Siemens Industry Online Support portal and the operating instructions document set for 6SE6400-series MM420. The full EMC installation guide is bundled in the MM420 documentation DVD and is also available as the document "MICROMASTER 4 EMC Guidelines" referenced historically in Siemens field notes.

5. Cable Routing Topologies That Change Emissions

How power cables are physically laid in the cabinet has a first-order effect on common-mode noise. The two most common configurations and their field-measured emissions differ by approximately an order of magnitude.

Cabinet builders commonly pull L1, L2, L3 in single cores from a top trunking channel to each drive's input terminals. When the three phases are laid parallel and not twisted, the magnetic field of the return currents does not cancel, and the cable bundle radiates like a single large loop. The same conductors laid in a trefoil (three cores repeatedly crossed over each other, not unlike a hair plait) cancel the magnetic field along the run, and common-mode radiation drops to roughly one-tenth of the flat-laid configuration. If you cannot refit the cabinet, the practical compromise is to use a three-core + PE cable that comes pre-laid in a near-trefoil geometry from the factory.

6. USS Bus Topology, Termination, and Biasing Requirements

USS is a single-master, multi-drop RS-485 bus. The MM420 implements USS on its control terminals as follows (refer to the MM420 operating instructions for the specific 6SE6400-series control terminal diagram, the standard pinout is reproduced below):

Mandatory rules for a stable USS multi-drop bus with two MM420 slaves:

1. Single bus, single segment, daisy chain only. No stubs. The bus enters each drive and continues to the next. Stubs of any length reflect energy and corrupt the bits.
2. 120 Ω termination at both physical ends of the bus only. For a two-drive bus with the master at one end, the master end and the far drive are the two ends. Do not terminate a drive in the middle of the bus. The MM420 USS interface does not include an internal termination resistor that can be enabled by parameter; the resistor must be added externally across terminals 29 and 30, or you must use a cable that includes a 120 Ω built-in termination at the ends.
3. Fail-safe bias. RS-485 needs the line to be in a defined idle state. The master (PLC, S7-200, S7-1200, or external controller) should provide a bias network: typically a 680 Ω pull-up from P+ to +5 V and a 680 Ω pull-down from N- to 0 V, with the 120 Ω terminator at the master end. Many Siemens S7 USS master blocks (e.g. USS_INIT for S7-200, USS_MASTER for S7-1200 with CM1241 RS-485) include this bias internally; check your CM module's documentation.
4. Single common-mode reference. The 0 V reference (terminal 31) must be continuous along the bus and must be connected to the cabinet ground at one point only — typically the master PLC end. A second bond at the far drive end creates a ground loop.
5. Maximum nodes. USS supports up to 32 nodes (including the master) on a single segment. Two MM420 slaves is well within this, but the practical node count drops with cable length and baud rate.
6. Baud rate selection. The MM420 USS port is set by parameter P2010 (USS baud rate): 4 = 2400, 5 = 4800, 6 = 9600, 7 = 19200, 8 = 38400, 9 = 57600, 10 = 115200. The slave address is P2011 (1…247). The PKW (parameter ID-value) length is set by P2013 (set to 4 for full PKW with parameter value access; 0 disables PKW). The PZD (process data) length is P2012. The telegram monitoring timeout is P2014 (in ms; 0 disables monitoring). Each drive must have a unique P2011. Verify both drives are on a different address.

7. Step-by-Step Diagnostic Procedure

Follow this sequence. Do not skip steps; each rule-out narrows the failure mode.

Step 1 — Reproduce the symptom in a controlled way

1. Power on the 3 kW MM420 alone. Verify normal operation, normal USS polling, normal BOP response.
2. Power on the 7.5 kW MM420 (do not start the motor, just energize the drive). Wait 10 minutes. Record whether the 3 kW MM420 BOP shows P---- during this time.
3. If the lockup reproduces without the 7.5 kW motor running, the root cause is inrush / DC-bus charge noise, not motor switching. This dramatically narrows the suspect list.
4. If the lockup only occurs when the 7.5 kW motor is actually running, the root cause is motor cable radiation.

Step 2 — Isolate the USS bus

1. With both drives powered and the 7.5 kW running, disconnect the USS cable from the 3 kW MM420 only (terminals 29, 30, 31).
2. Observe the 3 kW MM420 for 30 minutes. If the lockup disappears, the USS bus is in the causal chain.
3. Reconnect the USS cable. Re-introduce the 7.5 kW drive. Confirm the lockup returns.

Step 3 — Measure common-mode voltage on the USS reference

1. Connect an oscilloscope between the cabinet PE bar and the 0 V of the USS bus (terminal 31 of the 3 kW MM420, isolated from the drive).
2. Use a differential probe or verify the scope ground is isolated.
3. Trigger on the moment the 7.5 kW drive is switched on. Look for a step of more than ±5 V common-mode shift. This is the canonical signature of inadequate PE bonding.

Step 4 — Inspect termination

1. With the USS cable disconnected from the master, measure resistance across terminals 29 and 30 of the cable at each end.
2. You should read 60 Ω (two 120 Ω terminators in parallel) at each end if both ends are terminated. If you read 120 Ω, only one end is terminated. If you read open, no termination.

Step 5 — Verify unique slave addresses

1. Read parameter P2011 on both drives. They must differ.

Step 6 — Audit cable routing

1. Walk the cable run. Note the proximity to motor cables, DC bus cables, and any switched-mode power supply wiring.
2. Flag any parallel run longer than 1 m. Flag any bundled run.

8. Remediation: From Quick Fix to Hardened Installation

The reported case in the field resolved by actions 1, 3, and 5. After remediation the system ran for an extended period without lockup, confirming USS noise was the proximate cause.

9. Verification Tests After the Fix

1. Power-cycle both drives. Confirm both BOPs show 0.00.
2. Start the 7.5 kW motor. Run for 30 minutes. Verify the 3 kW MM420 BOP remains responsive (try toggling parameter view with the BOP jog dial). Repeat three times.
3. Start both motors. Run for 8 hours. Verify no P---- lockup.
4. Use the USS master to read parameter r0000 from both drives at the configured polling rate. Verify zero response timeouts over a 1-hour test.
5. Re-measure common-mode voltage between PE and USS 0 V during the 7.5 kW inrush. Confirm step is below ±2 V.

10. Field-Commissioning Checklist for Multi-MM420 Installations

• ☐ Each MM420 has its own short, flat PE strap to the cabinet PE bar.
• ☐ Cabinet PE bar bonded to building ground at one point.
• ☐ Motor cables are shielded, terminated with EMC glands at both ends (drive end and motor junction box end).
• ☐ Motor cable shield bonded to the cabinet backplane with 360° gland contact.
• ☐ USS cable is shielded twisted pair, shield bonded at one end only (master end).
• ☐ USS cable routed ≥ 200 mm from any power cable; crosses at 90°.
• ☐ 120 Ω terminator installed at both physical ends of the USS bus.
• ☐ Fail-safe bias confirmed at the master end (or built into the master RS-485 port).
• ☐ Unique P2011 slave address on every MM420.
• ☐ USS baud rate (P2010) identical on master and slaves.
• ☐ USS telegram timeout (P2014) set to a non-zero value (typically 1000 ms) for production; temporarily set to 0 only for diagnostics.
• ☐ 8-hour dual-drive soak test with USS polling active.

11. Extended Troubleshooting Matrix

12. FAQ