Resolving Siemens SINUMERIK 802S Alarms 700029, 700016, and 3000 on CNC Milling Machines

When a SINUMERIK 802S base line controller on a CNC milling machine — for example, the ZK7640 built by Jiangsu Douleng CNC Machine Tool Co. — presents three simultaneous alarms (700029, 700016, and 3000), the symptom points to a single cascaded fault rather than three independent problems. The 3000 Emergency Stop is the visible manifestation, while 700016 (MCP Drive Enable) and 700029 (Spindle Inverter Alarm) are user-defined PLC messages generated by the machine builder's ladder program to describe why the controller refused to release the spindle drive.

This reference documents how to interpret these alarms on the 802S base line, isolate the underlying cause on a non-Siemens-built machine, and apply the verification sequence that returns the mill to a serviceable state without removing the spindle drive or replacing the controller.

1. Problem Details

The reported fault on the ZK7640 mill appears as three concurrent messages on the 802S base line HMI:

The machine had been running normally prior to the fault. No new tooling, no recent program edit, no parameter change. That fact alone removes roughly 60 % of probable causes and tells the troubleshooter to focus on hardware interlocks rather than part program or NCK parameter corruption.

2. Alarm Numbering Convention on the SINUMERIK 802S base line

SINUMERIK controllers separate alarm authorship into two distinct bands:

• Standard Siemens alarms (1 to 6xxxx): Documented in the SINUMERIK 802S base line Lists Manual (6FC5597-3AA10-0BP0). These cover NCK, drive, communication, and PLC runtime issues for which Siemens ships a published explanation and clearance procedure.
• OEM / machine builder alarms (700000 and above): Defined entirely inside the machine builder's compiled PLC program. Siemens neither documents nor recognises the meaning of these numbers. Only the builder's ladder source, the builder's parameter sheet, or a contact at the builder's service desk can authoritatively decode them.

Both 700016 and 700029 fall in the OEM band. That is why Siemens support — and any third party — can only guess at the meaning. The guess below is the most likely interpretation given how 802S base line builders conventionally name their messages; treat it as a working hypothesis until confirmed by the original ladder dump.

3. Alarm 3000 — Emergency Stop: Cause Analysis

Alarm 3000 on the 802S base line is the standard Siemens message raised when the PLC's E_STOP input bit transitions low. The PLC bit is fed from a single hardware chain, which on a typical 3-axis milling machine reads as follows:

1. Front panel Emergency Stop mushroom (NC contact)
2. Rear cabinet Emergency Stop mushroom (NC contact)
3. Door interlocks on each access door (NC contacts in series)
4. Drive enable contactors' normally-closed feedback (NC contacts)
5. External safety relay output (e.g., Pilz PNOZ, Sick UE10)

All five elements must be closed (conducting) for the PLC to consider the chain healthy. If any one opens — even momentarily — the PLC latches E_STOP active, drops the drive enable relay, and posts alarm 3000 in the alarm bar.

Because 3000 is a latched condition, the operator must clear it explicitly with the Reset key on the MCP once the chain is electrically closed again. Cycling power will not, by itself, clear 3000 if any NC contact in the chain remains open.

4. Alarm 700016 — MCP Drive Enable Key: Cause Analysis

MCP stands for Machine Control Panel — the operator keypad with jog, rapid, feed-hold, cycle-start and reset keys. On the 802S base line, the MCP is connected via a dedicated ribbon to the I/O module and exposes each key as a discrete input bit to the PLC.

The most common builder interpretation of alarm 700016 ("Key MCP Drive Enable key") is one of two scenarios:

• Stuck or shorted key: The drive enable key contact is closed in hardware while the PLC expects it open. This is the more probable cause when 3000 (E-Stop) and 700029 (Spindle Inverter Alarm) are also present, because most builders wire the drive enable key into the same permissive chain.
• Required-action key: The builder's PLC is asking the operator to physically press and release the drive enable key as part of the re-arm sequence after E-Stop recovery.

To distinguish the two, watch the MCP's drive enable LED while the controller is powered and no key is being touched. If the LED is lit without anyone pressing the key, the contact is shorted. If the LED stays dark until you press it, the PLC is requesting the action.

5. Alarm 700029 — Spindle Inverter Alarm: Cause Analysis

The 802S base line does not, by itself, decode spindle drive faults in detail. It relies on a single discrete input from the spindle drive (or its interface card) wired into the PLC's SPINDLE_READY or SPINDLE_FAULT input. When the PLC sees the fault bit active, it raises the OEM alarm 700029 with whatever text the builder chose — in this case "Spindle Inverter Alarm".

The fault bit can be triggered by any of the following at the spindle drive:

Because the original poster reports "no error light lit at inverter", either the drive is in a latched-clear state (most modern drives blank the LED when the fault input is acknowledged but the drive is still in a faulted state), or the drive is signalling the fault through a relay contact rather than an LED — that is normal for drives wired to a controller's discrete input.

6. Initial Field Diagnostics — Power Cycle Procedure

Before opening the cabinet, perform the controlled power cycle that 802S base line service technicians use as the first triage step:

1. Press Reset on the MCP. If 3000 does not clear, the E-Stop chain is still open electrically — go to Section 9.
2. Turn the main disconnect to OFF. Wait 30 seconds for the controller's 24 VDC rail to decay.
3. Open the cabinet and confirm zero potential at the 400 VAC bus with a CAT III meter.
4. Press and hold the cabinet-mounted Emergency Stop while powering back on. This forces the controller into a safe state on boot.
5. Restore the main disconnect. Wait for the 802S base line HMI to fully boot (typical boot time on a 802S base line is 45–90 seconds; older PCU panels can take up to 3 minutes).
6. Release the Emergency Stop mushroom.
7. Press Reset on the MCP. Observe the alarm bar.

If only 3000 clears but 700016 and 700029 remain, the OEM ladder is correctly reporting that one or both of its monitored interlocks is still false. If all three clear and re-appear within five seconds, the ladder is detecting a steady-state hardware fault and the controller must be observed live with the door open to read the spindle drive LED before the boot sequence erases it.

7. Spindle Drive Inspection Procedure

Perform the following visual and electrical checks at the spindle drive:

1. Read the LED or 7-segment display on the drive front. If blank, power the drive off, wait the discharge time, then back on without enabling — many drives will re-display the last fault on power-up even when cleared.
2. Check the drive's fault output relay (terminals typically labelled RA-RB-RC or 30A-30B-30C). With the drive in fault state, the relay should be in its alarm position. Measure with a meter on the appropriate terminals.
3. Check 3-phase input voltage at the drive's L1/L2/L3 terminals. Reading should be within ±10 % of the drive's nameplate voltage (typically 380–400 VAC line-to-line for Chinese-built machines).
4. Check the DC bus voltage (terminals P and N) on drives that expose it. On a healthy 400 VAC input, expect ~540 VDC. If the DC bus is below 470 VDC, the input line is sagging — investigate upstream contactor, breaker or supply transformer tap.
5. Check spindle motor insulation resistance with a 500 V megohmmeter between any U/V/W terminal and earth. Reading should be >100 MΩ. Lower readings indicate moisture ingress or winding damage, both of which will produce intermittent overcurrent faults even when the drive looks "OK" at rest.
6. Check the encoder feedback cable (if fitted) for continuity between the drive end and the motor end. A broken shield or a single broken conductor is the single most common cause of "spindle inverter alarm" on first start-up of a machine that has been idle for weeks.

8. MCP Signal Tracing Procedure

With the controller powered and the door interlock defeated (using the documented maintenance key, never by bridging a safety switch), trace the drive enable key signal:

1. Locate the MCP wiring diagram in the machine's electrical drawing set (typically in the back of the cabinet or in the builder's documentation pack).
2. Identify the drive enable key's two terminals on the MCP.
3. With no key pressed, measure resistance across the contacts. Expect >1 MΩ (open).
4. Press and hold the key. Expect <1 Ω (closed).
5. If the contact reads shorted with no key pressed, the key switch is mechanically failed and must be replaced as a unit — repair is not field-expedient.
6. If the contact is healthy, the builder's ladder is requesting the operator action. Acknowledge by pressing the drive enable key once on power-up, exactly as the builder's commissioning procedure specifies.

Most 802S base line builders use the drive enable key as part of a deliberate re-arm sequence after E-Stop, and they will latch 700016 until they see a clean rising edge on that input. A momentary press and release is normally sufficient.

9. Emergency Stop Chain Verification

Alarm 3000 will not clear while any NC contact in the E-Stop chain is open. Use a meter on each section to localise the break:

1. Pull the upstream safety relay (Pilz, Sick, or equivalent) and bench-test its output with a jumper between the two safety input channels — only if you are authorised to defeat the safety circuit and only for the duration of the test.
2. Manually depress and twist-release each Emergency Stop mushroom to ensure none is mechanically latched in the depressed position.
3. Open and close each cabinet door, listening for the door interlock to click. Sticky interlocks (common in dusty shops) can leave the contact half-open.
4. Measure the voltage at the PLC's E_STOP input. On a 24 VDC system, expect 24 V when the chain is closed and 0 V when any NC contact is open.
5. Trace the chain section by section, isolating each NC contact with a temporary jumper until the PLC's E_STOP bit transitions high again.

10. PLC Ladder Dump — The Last-Resort Method

If the electrical cause cannot be localised by inspection and the machine builder is unreachable, the only authoritative source for the meaning of 700016 and 700029 is the compiled PLC program inside the controller. Extracting it requires a Siemens PG field programmer, the SINUMERIK 802S base line Toolbox (or the older S7-200 / 802S service tool), and a serial connection to the COM port on the controller's I/O module.

The rough procedure is:

1. Connect the PG to the controller's MPI/serial service port.
2. Start the SINUMERIK 802S service tool.
3. Upload the PLC project (.TEA and compiled binary) to the PG.
4. Open the ladder in the editor and search for the symbol or constant associated with 700016 and 700029.
5. Trace the rung backwards to the input conditions that raise the alarm.

For the alarm text "Key MCP Drive Enable key", search the project for the string literal that feeds the alarm display. For the alarm text "Spindle Inverter Alarm", search for the discrete input mapped to the spindle drive's fault relay. Both will reference the input byte/word and bit position that the builder chose.

This procedure is well documented in the SINUMERIK 802S base line service course materials available through the Siemens lifelong educational advantage program for industrial machinery, but it should only be attempted by personnel already trained on the platform.

11. Diagnostic Decision Matrix

12. Verification Sequence

After any corrective action, run the following return-to-service sequence:

1. Power on with E-Stop depressed.
2. Release E-Stop. Confirm no alarm 3000.
3. Press Reset on the MCP.
4. Press the drive enable key once if the builder's commissioning procedure requires it.
5. Confirm alarm bar is empty.
6. Jog X and Y at low feedrate (5 %) in manual mode for at least one full axis length each.
7. Run the spindle at 500 rpm for 30 seconds in M03. Listen for abnormal noise, watch the drive current display on the HMI.
8. Run a no-load part program (an air cut of the last proven program) for one full cycle.
9. Reset the machine's hour-meter counter if the fault interrupted a production job, and record the fault and root cause in the maintenance log.

If the alarms re-appear during the verification sequence, the fault is intermittent and the diagnostic cycle restarts from Section 6 with focus on intermittent wiring (door interlock flexing, vibration-loose terminal, thermal trip on a cooling fan).

13. Preventive Measures for Repetitive Occurrence

Three alarms appearing simultaneously on a machine that was running smoothly the day before is rarely a coincidence. Apply the following preventive actions:

• Tighten cabinet terminals annually: Vibration from the spindle and chip conveyor will loosen screw terminals on Phoenix, Wago and Weidmuller blocks. A loose 24 VDC return on a safety input is a classic intermittent cause of phantom 3000 alarms.
• Clean door interlocks quarterly: Coolant mist and chips accumulate on the actuator surface. Worn actuators produce partially-open contacts that look healthy with a meter but fail under vibration.
• Inspect spindle encoder cabling every six months: Drag-chain flexing fatigues encoder conductors long before the cable jacket shows damage. Replace the encoder cable as a wear item, not as a repair.
• Verify drive cooling fans annually: A failed heatsink fan on a SIMODRIVE or Frenic will produce an over-temperature fault that clears on cool-down, only to re-appear hours later under load.
• Archive the PLC project to a labelled USB stick stored in the cabinet: When the next 700000-series alarm appears, the ladder dump is one minute away rather than one support call away.

14. Escalation Path When the Builder Is Unreachable

When the machine builder (in this case Jiangsu Douleng CNC Machine Tool Co.) cannot be contacted and the in-house team has exhausted the inspection-based diagnostics above, escalate in the following order:

1. Local Siemens industrial automation distributor service desk. They will not decode the OEM alarm text but they can confirm the controller hardware is healthy.
2. SINUMERIK hotline through Siemens Industry Online Support for standard 802S base line platform faults.
3. Independent SINUMERIK service integrator listed on the Siemens industrial machinery partner program.
4. SINUMERIK 802S base line service training and refresher courses through the educational advantage program — these equip in-house maintenance staff to perform ladder dumps and symbol-table inspections in future.

For sites outside China, the closest Siemens CNC competence centre is typically the regional Siemens office; in Pakistan, the local Siemens office is reachable via the official Siemens country website (linked above). Cross-border support for OEM alarms is limited by the builder's IP, but the controller's hardware health and standard alarm band can always be confirmed.

FAQ