The PLC is obsolete.
The plant is not.
Like-for-like spares, partial migration or a full rebuild: how to tell which one a plant needs, what to read from the nameplate, and how to keep the outage short while the controller changes.
Ask about a PLC replacement or email your part numbers- When a PLC is really obsolete
- The three options
- Reading the nameplate
- I/O and program
- Risk controls
- Decision checklist
- Sourcing spares
When a PLC is really obsolete
“Obsolete” covers three situations, and each calls for a different action.
- Phase-out announced. An end date is published, but the product is still made and sold. This is the time to plan.
- Product cancelled. New units are no longer made. Spares and repairs usually continue for a defined period, but every purchase competes with other plants on the same controller.
- End of service. No manufacturer repair, no firmware, no supply except the secondary market. Each failure becomes a sourcing job.
Siemens is the best-known example: the SIMATIC S7-300 and S7-400 families have a published phase-out announcement, a product-cancellation date and a spare-parts period that follows; check the current dates in Siemens’ product phase-out notices for S7-300 and S7-400.
The practical test is not the manufacturer’s date but the days between a CPU failure and a working controller, against the days the plant can stand still. A PLC that still runs is not a problem; one that cannot be replaced inside that window is.
The three options
1. Like-for-like spares
Keep the installed system and hold spares for what fails: CPU, power supply, critical I/O, memory card. No program change, no rewiring, a changeover measured in hours. The cost: spares get rarer and dearer every year, and secondary-market units vary, so condition and warranty must be stated before you buy. Right for a plant with a few years left.
2. Partial migration
Replace the CPU and communication, keep the I/O wiring and field devices. Several manufacturers offer migration paths that keep the rack footprint or fieldbus, so the terminals stay put. Shorter outage than a rebuild. The risk sits in the converted program: a mixed system needs a complete functional test.
3. Full migration
New controller, I/O, panel and program, built and tested off-site and swapped during a planned shutdown. Highest upfront cost, lowest long-term risk, and the only option that also fixes tired wiring and missing drawings. Right when the panel itself is due, or the process is changing anyway.
Reading the nameplate, and what a supplier needs
Every module carries a label with the manufacturer’s order number, a product or hardware version and a serial number; the firmware version is read from the engineering tool. Siemens order numbers (MLFB) look like 6ES7 315-2AH14-0AB0. The order number identifies the module; version and firmware decide whether a replacement drops in or needs a project update. To quote an exact replacement, a supplier needs:
- a photo of each label: CPU, power supply, every I/O and communication module, memory card;
- the quantity per reference, and which ones are critical;
- the firmware version, if the engineering tool shows it;
- a hardware-configuration export or a rack-and-slot list, if one exists;
- whether a program backup exists and who holds it.
With that list, a supplier can confirm the exact reference, or the manufacturer’s successor with the cross-reference when the original is gone.
I/O and program considerations
I/O. Count channels per type (digital and analog, in and out) and note the signal levels: 24 V DC, relay outputs, 4–20 mA, RTD or thermocouple. List special functions such as high-speed counters, and every communication link: PROFIBUS, PROFINET, Modbus, serial links to drives, HMI and SCADA. Connector type decides how much rewiring a migration really costs.
Program. Note the language and structure, system functions and libraries that may not exist on the new platform, HMI tag names, recipe and setpoint data, passwords and licences. Conversion tools help, but a converted program needs a line-by-line review and a full functional test.
Risk controls during a live changeover
- Spares on site before you start: the new CPU plus one spare, power supply, critical I/O, memory card, connectors, fuses. Keep the old CPU as a cold standby until the new system has run a full production cycle.
- Backup first. Upload the running program and hardware configuration before anything is touched, keep two copies in two places, record the passwords.
- Build and test off-site. Have the new panel assembled, wired, marked and tested in a workshop while the old system still runs; a factory acceptance test with as-built drawings, test reports and component certificates finds the errors before the outage does.
- Mark and photograph every wire before it is disconnected, and have the I/O list signed off.
- Plan the window with a rollback. Define the point of no return, loop-check each I/O, test interlocks and safety functions first, then run the process under supervision before handover.
Decision checklist
- Where is the controller in its lifecycle: announced, cancelled or out of service?
- How long can the plant stand still after a CPU failure?
- Is there a verified program backup, and is the engineering software still licensed?
- Are spares for the critical modules available, in which condition and warranty?
- Are the I/O wiring and the panel sound, or due for replacement anyway?
- Will the process change in the next few years?
- Who does the engineering: in-house, an integrator, or both?
- Which outage window can be planned?
Spares available, panel sound, no process change: option 1. Backup verified, CPU the only weak point: option 2. Panel tired, process changing or no usable backup: option 3.
Sourcing spares for a phased-out controller
Verified stock means more than a matching part number: the serial numbers and documentation are checked, the hardware revision is confirmed, and the condition is stated in writing before purchase — new, new old stock, refurbished or used — together with the warranty that follows that condition.
Every consignment should travel with a commercial invoice, packing list, certificates of origin and the manufacturer’s documentation, and the quotation should carry the HS code and the lead time per line, so the file is complete before the module ships.