Siemens SINUMERIK Alarm 2120 NCK Fan: Diagnostic Guide

Alarm 2120 NCK Fan Alarm — System Identification

Siemens SINUMERIK numerical controls monitor cooling airflow at the NC module (NCU) and raise alarm 2120 whenever the on-board tachometer feedback at the NCK fan falls outside the expected envelope. According to the official Siemens Industry Online Support entry for this alarm — ID 109817112 — SINUMERIK ONE / SINAMICS S120 alarms — the alarm text reads 2120 NCK fan alarm type %1 with the explanation: "The fan speed monitoring at the NC module has responded. If 'Type 1' is output, then it involves a…" The truncated %1 placeholder is the type identifier supplied by the NCK firmware at the moment the speed monitor trips, and is shown on the HMI alarm line together with the clearing condition.

The alarm originates in the NCK (Numerical Control Kernel) software, not the SINAMICS S120 drive firmware, although both subsystems are mounted in the same NCU box on most SINUMERIK ONE and SINUMERIK 840D sl machines. The speed monitor is a hardware watchdog on the NCU carrier board: a Hall-effect or two-pulse tach input tied to the NCK fan connector reports the rotational frequency of the impeller. If the frequency drops below a threshold, or if the tach signal is stuck high or stuck low, the NCK posts alarm 2120 and triggers the configured stop reaction.

Affected Hardware and Configuration

The field case that this guide is built around involves an indexMS32 multi-spindle turning machine from INDEX-Werke configured with two NCU modules. Multi-spindle machines that combine a main and a sub-spindle machining unit, or a left and a right machining unit, are routinely delivered with a 2-NCU topology in which:

• NCU_1 hosts the primary channel and the first SINAMICS S120 line module / motor module stack.
• NCU_2 hosts the secondary channel and the second SINAMICS S120 stack.
• A PROFIBUS or PROFINET link ties the two NCUs, and one NCK is usually designated the link master.

Each NCU box houses its own NCK fan, its own battery-backed SRAM retention circuit, and its own connection to the machine's 24 V control supply. The 2-NCU configuration therefore requires two independent NCK cooling paths, and alarm 2120 is reported per NCU — the HMI will show the alarm against NCU_1 NCK fan alarm or NCU_2 NCK fan alarm depending on which module's speed monitor has tripped.

Fan Architecture: Locating the Three Fan Types

Replacing the wrong fan is the most common reason the alarm persists after a service intervention. On a 2-NCU indexMS32 there are three distinct cooling components, and the Siemens documentation referenced by the service team only names one of them.

1. Fan + battery module (top slot) — slides into a guide directly beneath the DC-link busbars. It is released by pressing a small clip on the module and pulling the assembly towards the operator. This is the module mentioned in the operator documentation, and it is the one most engineers replace first.
2. Bottom NCU cooling fans (2 fans per NCU) — these sit on the underside of the NCU carrier assembly. On a 2-NCU machine the total count is 2 × 2 = 4 bottom fans. They are not labelled in the operator's handbook as separate spare parts, which leads to field surprises when they fail.
3. Internal NCU box fan sensor / tach input — not a service-replaceable fan at all, but the input that the NCK reads to decide whether the fans are turning. A failure here looks identical to a stalled fan to the NCK.

Alarm 2120 Parameters and Type Identifiers

The %1 placeholder in the alarm line is a numeric type code. The Siemens Industry Online Support entry for alarm 2120 names at least the Type 1 condition, which corresponds to a sustained under-speed of the monitored impeller. The exact threshold values are firmware-version dependent and must be confirmed against the active NCK build. The interpretation of the type field can be cross-checked with the following heuristics, which match the published behavior of the NCK fan monitor:

The alarm has the standard NCK response stack: it is displayed, written to the alarm log, and — depending on MD configuration in the active commissioning archive — assigned a stop reaction that can be as harsh as a NCK-CLEAR-with-Reset (i.e. the controller will not restart until the alarm is cleared and a power-on is performed). The clearing bit is the same bit that the HMI "Reset" key drives, but the underlying fan condition must also be gone; the alarm re-arms if the tach input is still out of range.

Pre-Troubleshooting: Backup, Battery, and Safety

Before any NCU module is opened, capture the controller's data so that the machine can be re-commissioned if a board is damaged. The indexMS32 is battery-backed, and any disturbance to the battery circuit can cause SRAM loss.

1. Back up the NC — series-commissioning archive via HMI Commissioning → Series commissioning → NC → Create series start-up file. Save to a USB stick or a network share.
2. Back up the PLC — STEP 7 / TIA Portal upload of the PLC project, plus an export of the safety program (F-CPU) if present. Confirm the F-CPU signature and the safety acceptance test status.
3. Back up the drives — SINAMICS S120 parameter sets for every Control Unit and every active line module / motor module on both NCUs.
4. Check the battery state — HMI menu Diagnostics → Battery shows the buffer voltage. Replace the battery on the module being swapped if it is below 3.0 V under load, or if the date-of-last-replacement sticker is older than the site's preventive-maintenance interval.
5. ESD precautions — wrist strap, ESD mat, ESD-safe tools. NCU carrier boards are not field-repairable and an ESD event is unrecoverable.
6. Authorise a controlled power-down — on the indexMS32 the safety chain is normally run from a separate 24 V path, but the NCU itself should be powered down (DC-link discharged) for any work beyond a hot-swap of the fan + battery module.

Diagnostic Procedure: Step-by-Step Isolation

The sequence below is the order that has been validated in the field on a 2-NCU indexMS32. Skipping steps — especially the bottom-fan check — is the single most common reason the alarm re-appears within hours of a service call.

Fan + Battery Module Removal Procedure

The fan + battery module is the only NCU cooling component that the operator documentation highlights, and it is the one most engineers replace first. The mechanical procedure is straightforward and is hot-swappable in principle, but the battery circuit must be respected.

1. Confirm that a current NC and PLC backup is on external media. See the backup procedure above.
2. Locate the module: it sits in a guide directly beneath the DC-link busbars, immediately behind the front bezel of the NCU.
3. Press the small release clip on the module — typically on the right-hand side of the module as viewed from the front. The module will partially eject.
4. Pull the module straight out toward the operator. Do not angle the module; the DC-link busbar edge is unforgiving of skewed insertion.
5. Inspect the battery contacts for corrosion or low-spring tension. If the battery is older than the preventive-maintenance interval, replace it on the bench, not in the machine.
6. Insert the replacement module until the clip re-engages. Confirm that the module is flush with the surrounding NCU housing.
7. Power-cycle the NCU if the battery circuit was disturbed. A single battery alarm on the HMI is expected if the module was removed with the 24 V supply still up; the alarm clears on the next power-on.
8. Reset alarm 2120 from the HMI and observe for at least 10 minutes. The NCK speed monitor has a debounce time; a fan that hovers just above the threshold can take several minutes to re-arm the alarm.

Bottom NCU Fan Replacement

The bottom fans are the surprise component. The operator documentation does not call them out as separate spare parts, and on the indexMS32 they are easy to overlook because they are mounted on the underside of the NCU carrier. On a 2-NCU machine the total population is 2 fans × 2 NCUs = 4 bottom fans.

1. De-energise the affected NCU and confirm that the DC-link is discharged (typically < 5 V on the SINAMICS line-up, wait at least 5 minutes after opening the DC link).
2. Access the underside of the NCU. On the indexMS32 the NCUs are mounted in a sub-cabinet; a partial drawer or swing-out panel may be required.
3. Identify the bottom fan assembly. It is a pair of 40 mm or 50 mm axial fans on a common carrier, with a single 3-pin or 4-pin connector on the NCU carrier board.
4. Disconnect the fan connector. Do not pull on the wires; use the connector body. The latch is on the same side as the keying tab.
5. Remove the carrier screws. Note the orientation of the airflow arrow — bottom fans on most NCU variants push air up through the carrier; reversing the arrow creates a dead-headed condition that overheats the board.
6. Install the replacement carrier, route the cable in the original strain-relief path, and re-connect the connector.
7. Re-energise the NCU, reset the alarm, and observe.

NCU Box Sensor and Tachometer Signal Verification

If the alarm survives a fan + battery swap and a bottom-fan swap, the next suspect is the sensor input on the NCU carrier board. The NCK reads the tach signal through a dedicated pin on the NCU connector strip; the pinout and the input circuit are documented in the NCU hardware manual for the specific variant, but the diagnostic method is the same across the family.

1. With the fans running and the NCU powered, measure the tach signal at the carrier-side connector with an oscilloscope. A healthy Hall-effect output is a 0–5 V square wave (or open-collector with pull-up) at the fan's nominal RPM.
2. If the waveform is present at the connector but the alarm persists, the fault is on the NCU carrier — the input is open, shorted, or has lost its pull-up. The carrier is a field-replaceable unit, but the new carrier must be re-licensed with the same CF card image.
3. If the waveform is absent at the connector, work back toward the fan. Inspect the cable for chafe, the connector for bent pins, and the fan itself for a stuck rotor.
4. Cross-check by swapping the fan connector with a known-good fan from a sister NCU. If the alarm follows the fan, the fan is the cause; if it stays with the NCU, the carrier is the cause.

On the indexMS32 the service team confirmed that an apparently-good fan from a sister machine can fail to clear the alarm when the underlying carrier input is faulty. The "swap with another machine" test, which is normally diagnostic, becomes misleading in this case unless the tach signal is measured directly.

Root Cause: Internal Debris and Connection Faults

Field experience with the indexMS32 alarm 2120 case produced an unusual root cause: loose debris inside the NCU box — described by the service team as "little chips bouncing inside the NCU box". Conductive swarf or severed strands from a wiring harness can land on the tach input pin, the fan connector, or the carrier's power-rail and intermittently short the speed-monitor circuit. The fault is not a stuck rotor, which is why fan swaps do not clear it.

Other connection faults that present identically:

• Backplane connector seating — the NCU carrier board mates with a backplane connector that is sensitive to torque on its mounting screws. An under-torqued or over-torqued screw changes the connector's normal force and the tach input can flicker.
• Cable chafe against the chassis — the bottom-fan cable runs close to the cabinet sheet metal. A vibration-induced chafe can wear through the insulation and short the tach line to chassis ground. The short is not always hard — it can be a 50 % duty-cycle partial short that the NCK interprets as a missing pulse.
• Battery leakage on the fan + battery module — rare on NiMH cells, but documented on early Li-primary cells. The leakage path can corrode the connector and the tach input together.

NCU Replacement Procedure

If steps 1 through 5 of the diagnostic flowchart above have been exhausted and the alarm still appears on the HMI, the carrier is the prime suspect. The NCU is treated as a field-replaceable unit; the carrier board itself is not service-repairable.

1. Re-confirm the NC, PLC, and SINAMICS backups. The replacement NCU will arrive with a default CF card; the site's commissioned image must be reloaded before the machine is put back into production.
2. Power down the affected NCU and discharge the DC link. Wait at least 5 minutes after opening the line contactor.
3. Label every connector on the back of the NCU. The connector count is high (PROFIBUS, PROFINET, DRIVE-CLiQ, I/O, HMI, 24 V, enable, etc.) and a misplaced connector on the new NCU can be a long debugging exercise.
4. Remove the NCU from its mounting plate. Most NCU variants use four M5 or M6 screws and a grounding stud. Do not lose the grounding hardware; the EMC performance of the replacement is not the same without it.
5. Move the CF card from the old NCU to the new NCU if the replacement ships without one. Note that license-bound options may need to be reactivated through the Siemens license manager.
6. Install the new NCU, re-torque the mounting screws to the specified value, and re-attach every connector in the order they were labelled.
7. Re-load the commissioned image, re-license if required, and run a controlled power-up. Confirm that the NCK and PLC start cleanly and that the SINAMICS line-up returns to operational state.
8. Run an axis-reference and a tool-measurement cycle before returning the machine to production.

Verification, Commissioning, and Alarm Clear

The verification step is what tells you whether the fix is real or whether the alarm is going to return in an hour.

1. After each intervention, reset alarm 2120 from the HMI and observe for a minimum of 30 minutes under normal load. The NCK speed monitor has a debounce; a marginal fan can take 10–15 minutes to trip the alarm.
2. Force a thermal load on the NCU by running an active part program with high axis acceleration. The bottom fans carry the heat from the SINAMICS motor modules back through the cabinet; a fan that survives idle can fail under load.
3. Check the NCK operating-hours counter for the fan, where the active commissioning archive exposes it. A fan that has run past its design hours is a candidate for preventive replacement even if it has not yet alarmed.
4. Log the fix in the machine's maintenance record with the alarm number, the type field, the intervention, the part numbers used, and the NCK firmware version. This is the data that the next service engineer will need.
5. If the alarm returns, escalate to a full NCU swap. Do not chase a third fan replacement; the field evidence on the indexMS32 is that two fan swaps and a working sensor path were still not enough — the carrier had to come out.

Field-Proven Diagnostic Matrix

Documentation and Escalation

For every alarm 2120 event, capture the following into the machine's maintenance log so that the next intervention starts from data, not from guesswork:

• Exact alarm text from the HMI, including the %1 type field.
• NCK firmware version (HMI Diagnostics → Version).
• NCU hardware variant and serial number (printed on the NCU label).
• Fan + battery module part number, if replaced.
• Bottom-fan part number, if replaced.
• Carrier / NCU part number, if replaced.
• Time-stamped observation log of the alarm for at least 30 minutes after the fix.

If the alarm persists after the steps above, escalate to Siemens Industry Online Support with the captured data and the active commissioning archive attached. The Siemens alarm 2120 reference page is the starting point: SINUMERIK ONE / SINAMICS S120 alarms — ID 109817112.