Resolving Sinumerik 840C MMC Error 8: Powered Without NC

1. Problem Overview

The SINUMERIK 840C is a multi-CPU CNC platform that combines a Man-Machine Communication CPU (MMC-CPU), a Numerical Control CPU (NC-CPU), a Programmable Logic Controller CPU (PLC-CPU), and a Communications Board CPU (CSB-CPU) in a single rack. When any of these CPUs fails to handshake correctly at power-up, the operator panel and individual CPU diagnostic displays surface the fault.

The fault pattern reported in the field source is characteristic:

• MMC-CPU 7-segment display: Code 8 (steady)
• NC-CPU 7-segment display: Code 0 then flashes and changes to 8
• PLC-CPU red LED: Blinks 11 times in a repeating pattern
• CSB-CPU: Battery-backed; status not specified in the fault

Per the official SINUMERIK 840C / SIMODRIVE 611-D Diagnostics Guide (PDF), these codes are decoded in the Error Display on CPU section. The combination reported here points to a coordinated failure of the inter-CPU startup handshake, dominated by the NC-CPU not becoming ready in time for the MMC to complete its bootstrap.

2. SINUMERIK 840C Multi-CPU Architecture

The 840C rack is laid out, from right to left in the standard 19-inch subrack, as follows:

Inter-CPU communication is performed over a parallel backplane bus. At power-on, the CPUs execute a synchronized bring-up sequence:

1. CSB-CPU initializes and posts ready.
2. NC-CPU performs POST, loads its system software from the MMC's floppy or hard disk, and signals ready on the backplane.
3. PLC-CPU brings the S5 runtime online, then signals ready.
4. MMC-CPU waits for the NC-CPU ready flag, then loads the HMI images and operator panels.

Any break in this dependency chain — particularly steps 2, 3, or 4 — results in the MMC being powered without a fully initialized NC, which the MMC firmware reports as code 8.

3. Decoding the MMC-CPU Code 8

The MMC 7-segment display is the operator's primary visible fault indicator. Per the 840C Diagnostics Guide § 2 ("Error Display on CPU"), the MMC codes include:

Code 8 is a hard fault: the MMC is energized, the +5 V / +15 V supplies are healthy, but the NC-CPU has not established any handshake with the MMC. This rules out MMC-side software faults (which would present as code 0 or 1) and points the investigation toward the NC-CPU, the backplane, or the power distribution feeding the NC slot.

4. Decoding the NC-CPU Code 0 → 8 Transition

The NC-CPU 7-segment display first shows 0, then blinks and changes to 8. This two-stage pattern is informative:

• Initial 0: The NC POST passed and the CPU entered its idle / kernel state without any active error condition. Memory test, processor test, and bus test all completed cleanly.
• Subsequent 8: The NC reached the point at which it should have loaded its system software (NCK, interpolator, axis modules) from the MMC's mass storage. Because the MMC did not present the expected handshake response — either because the MMC is itself stuck in code 8, or because the backplane coupling is degraded — the NC times out and posts 8 ("system software not loaded").

On its own, NC code 8 confirms that the NC-CPU hardware is healthy and the POST is good; the problem is the failure of the NC ↔ MMC coupling sequence.

5. Decoding the PLC-CPU 11-Blink Code

The PLC-CPU on the 840C is an S5-115U-compatible module. Its front-panel red LED is the S5 "BA" / "SF" fault indicator. The blink count is a fault class pointer; it must be read in conjunction with the ISTACK (interrupt stack) and the error code byte in the S5 diagnostic buffer.

For the 840C PLC subsystem, the Diagnostics Guide § 1 ("Diagnostics on the PLC") maps the LED blink counts to fault classes. The 11-blink code, in the context of a multi-CPU startup stall, is typically a coupling fault or bus access fault:

6. Combined Fault Interpretation

Triangulating the three indicators, the fault chain is:

1. Power is applied to the rack; CSB initializes.
2. NC-CPU passes POST (initial 0) but stalls loading its system software because the MMC's NC coupling handshake never completes.
3. PLC-CPU attempts to establish its coupling with the NC at startup, times out, and posts the 11-blink bus/coupling fault.
4. MMC never receives a "NC ready" signal and falls through to its terminal fault code, 8 ("MMC powered without NC").

The MMC code 8 is a symptom, not the root cause. The investigation must be directed at the NC-CPU bring-up, the backplane coupling, and the power rails feeding the NC slot.

7. Root Cause Matrix

8. Required Tools and Prerequisites

• Documentation: SINUMERIK 840C / SIMODRIVE 611-D Diagnostics Guide (PDF)
• PG/PC with S5 programming cable (15-pin D-sub to PC) or PG 685/750
• Digital multimeter (CAT III, ≥600 V)
• Oscilloscope (for +5 V rail ripple check on NC slot)
• ESD-safe work surface and wrist strap
• Spare 840C modules (NC-CPU, PLC-CPU, backplane) if available
• Bootable NC system disk (matched to machine version)

9. Step-by-Step Diagnostic and Resolution Procedure

Step 1 — Lockout and Discharge

1. Open the main cabinet disconnect.
2. Verify 0 V on the 611-D DC bus with a CAT III meter.
3. Wait 5+ minutes; re-verify.

Step 2 — Visual and Thermal Inspection

1. Open the 840C subrack.
2. Verify both rack cooling fans are present and free-spinning. Replace any missing or seized fans before proceeding.
3. Inspect the four CPU modules from right to left: MMC, PLC, NC, CSB.
4. Check the CSB backup battery voltage (typically 3.6 V lithium). Replace if below 3.0 V — a dead CSB battery can prevent NC bring-up on older firmware revisions.
5. Inspect backplane connectors for dust, oxidation, or bent pins.

Step 3 — Reseat All CPUs

1. With ESD precautions, remove and reseat the NC-CPU, then PLC-CPU, then MMC-CPU.
2. Power on and observe the three displays.
3. If the MMC shows anything other than 8, the fault was a contact issue and is resolved.

Step 4 — Verify Power Rails on the NC Slot

1. With the rack powered, measure the +5 V and +15 V rails on the NC slot backplane pins (refer to the 840C hardware manual for pinout).
2. +5 V must be within ±5 % (4.75 V to 5.25 V) with ripple < 100 mV pk-pk.
3. If out of tolerance, replace the rack power supply before further diagnosis.

Step 5 — Read the PLC Diagnostic Buffer

1. Connect the PG to the PLC-CPU's serial port.
2. From STEP 5 (or equivalent), select PLC → Diagnostic Buffer or read the ISTACK.
3. Look for the 11-blink fault class entry. The ISTACK will identify whether the PLC was waiting on a coupling mailbox to the NC, or whether the backplane access itself failed.
4. Note the OB (organization block) and event byte; cross-reference to the Diagnostics Guide § 1.

Step 6 — Reinstall NC System Software

1. Insert the bootable NC system floppy / load the HDD image onto the MMC.
2. Power-cycle the rack.
3. Observe the NC display. It should boot through a sequence (typically 0 → 1 → 2 → …) before settling on a ready state.
4. If the NC still stalls on 8 with valid system media, proceed to Step 7.

Step 7 — Swap the NC-CPU

1. Replace the NC-CPU with a known-good spare of the same MLFB (Siemens part number, e.g., 6FM1 840-x for the NC-CPU family — confirm against the machine's hardware list).
2. Power on and observe.
3. If the new NC-CPU boots cleanly and the MMC transitions out of code 8, the original NC-CPU is faulty and must be replaced.
4. If the new NC-CPU shows the same fault, the backplane or power supply is the root cause.

Step 8 — Backplane and Power Supply Isolation

1. If both NC-CPUs fault, swap in a known-good backplane subassembly.
2. If the fault persists, replace the rack power supply.
3. As a last resort, escalate to a Siemens service engagement (the 840C has been a phased-out platform; confirm the regional support window before initiating).

10. Verification Procedure

After resolution, perform the following verification sequence before returning the machine to production:

1. Power off → on. Confirm MMC advances from 0 to a steady "ready" state and the operator panel shows the standard 840C HMI.
2. Confirm NC display shows no error (typically 0 with the green ready LED lit).
3. Confirm PLC red LED is extinguished (no blink code, no steady red).
4. From the MMC, navigate to Diagnostics → NC/PLC Status and confirm both subsystems report Ready.
5. Reference all axes; confirm the position display updates without Follow-up error or Servo enable missing alarms.
6. Execute a dry-run NC block (e.g., G0 X0 Y0 Z0) at low feedrate override (5 %).
7. Run a full M-code and auxiliary function test (tool change, coolant, spindle CW/CCW).
8. Log the resolution in the machine's maintenance file; record the NC-CPU MLFB if replaced.

11. Preventive Maintenance Recommendations

• Cooling fans: Replace rack fans every 4 years, regardless of audible condition. Heat is the leading degradation factor for 840C CPUs.
• CSB battery: Replace every 3 years; use only the Siemens-specified lithium cell.
• Backplane: Inspect gold-finger contacts annually; clean with isopropyl and a non-abrasive swab.
• System media: Maintain two verified bootable copies of the NC system software.
• Spare parts inventory: Stock at least one of each CPU module (MMC, PLC, NC, CSB) and one rack power supply. The 840C is in long-term support, and lead times can be long.
• Thermal monitoring: Install a small temperature logger inside the rack cabinet; alarm at 45 °C internal.

12. Related Alarms and Edge Cases

Several adjacent MMC and NC codes can be confused with code 8. The following table is a quick disambiguation:

If the PLC red LED shows a count other than 11, consult the Diagnostics Guide § 1 directly — the 11-blink interpretation in this article is the most common for the specific fault pattern reported, but other blink counts indicate I/O, memory, or bus faults that require different remedies.

13. When to Escalate

The 840C is a mature, phased-out platform. Escalate to a Siemens CNC service partner (or a qualified third-party retrofitter) when:

• The NC-CPU and PLC-CPU are both confirmed faulty and no spares are on hand.
• The backplane is damaged (replacement requires a full machine re-wiring in some installations).
• The MMC hard disk / floppy subsystem is failed and the boot media cannot be reconstructed.
• Repeated thermal shutdowns indicate a deeper cabinet ventilation problem.

In many plants the long-term remediation is a migration to a current-generation SINUMERIK (e.g., 840D sl or ONE) — but the 840C fault resolution procedure above will keep the legacy machine productive in the meantime.

Frequently Asked Questions