
On a ControlLogix or CompactLogix rack, a solid red OK indicator means an unrecoverable module fault, flashing red means a recoverable fault or a module that has lost its connection, and flashing green means the module is healthy but not owned or not communicating with its controller. That last one is the important distinction: a flashing green I/O module is almost never broken hardware — it is a configuration or network problem, and pulling it is wasted downtime.
This guide decodes the patterns in the order a maintenance technician meets them, ranks what actually causes each, and says which module is worth shipping and which is not.
The indicator table
| Indicator state | On an I/O module | On a controller | First move |
|---|---|---|---|
| Steady green OK | Owned, configured, operating | Running, no faults | Nothing — look elsewhere |
| Flashing green OK | Not owned / no connection to controller | Standby or no project loaded | Check ownership, config, path |
| Flashing red OK | Recoverable fault or connection timeout | Major recoverable fault | Read the fault, check the path |
| Steady red OK | Unrecoverable hardware fault | Major non-recoverable fault | Power-cycle once, then suspect hardware |
| No indicator at all | No backplane power reaching the module | No power | Suspect the rack power supply |
| Red I/O or field-power indicator | Field-side fault: fuse, short, overload | — | Field wiring, not the module |
Two rows in that table save more money than the rest combined.
No indicator at all is a power problem, not a module problem. A module that is completely dark, or a whole rack that is, points at the rack power supply long before it points at the modules. That is far and away the most common "dead PLC" call, and it is worked through in dead PLC rack? check the power supply first.
A red field-power or I/O indicator is pointing outward. Output modules with electronic protection report a fault when the field side is shorted or overloaded — a jammed solenoid, a chafed cable, a failed valve coil. The module is telling you about your machine, not about itself. Clear the field fault and the indicator clears.
Flashing green: the one to get right
A flashing green OK on a 1756 I/O module means the module is powered, healthy and running its self-tests — and has no connection to an owning controller. Causes, ranked by how often they actually occur:
- The controller is not in Run, or the project holding that module's configuration is not loaded.
- The module was replaced with a different catalog number or series, and the configured module does not match the installed one. Electronic keying will refuse the connection on a mismatch, which is the feature working as designed.
- The connection was never configured — a module added to the rack physically but not to the I/O tree.
- A network or path problem on a remote rack: the communication module, the path, or the network itself.
- Slot mismatch — the module is in a different slot than the configuration expects.
None of those five are repaired on a bench. Shipping a flashing-green module means paying freight to be told it tests good.
"We sent three I/O cards out over about a year before somebody noticed they all came back fine. Every one of them was flashing green because the processor was faulted, not because the card was bad. The rack was telling us the truth and we were reading it as a hardware failure."
— Controls technician, 11 years, plastics; name withheld by request
Flashing red and solid red
Flashing red on an I/O module most often means a connection that was previously established has timed out — the controller and the module knew each other and now do not. That is a network, path or controller problem far more often than a module problem. It can also indicate a recoverable module fault, which is why the controller's fault information is the next place to look rather than the module itself.
Solid red on an I/O module means the module has diagnosed an unrecoverable internal fault. Power-cycle the rack once, deliberately, and see whether it returns. If it comes back solid red, the module has genuinely failed its own self-test and is a repair candidate.
On a controller, the same logic applies with higher stakes. A flashing red OK on a Logix processor indicates a major recoverable fault — the kind you clear after reading the fault code and correcting the cause — or a controller with missing or corrupted firmware. Solid red indicates a major non-recoverable fault, and on the L7x generation the four-character status display spells out what the processor found, which is worth photographing before you do anything else.
Note one thing about processors that we would rather say up front than after taking your money: most CPU faults are not economically repairable. We provide honest test-and-verify service on controllers, and a large share of the time the verdict is that the module is not worth repairing. A shop that tells you every processor is repairable is telling you something else.
The sequence that actually diagnoses a rack
Work outside in. It takes about ten minutes and it is almost always faster than the alternative.
- Lock out and verify before opening the enclosure. Control cabinets carry sources that are not obvious from the door, and the control-of-hazardous-energy rules (osha.gov) exist because of racks exactly like this one. NFPA 79 (nfpa.org) governs the electrical practice around the machine itself.
- Photograph the whole rack with the fault live — every indicator, every module, in one frame. Ten seconds now; it is the single most useful artefact for any later conversation.
- Is anything lit at all? If the rack is dark, stop and go to the power supply.
- Read the controller first. A faulted processor makes healthy I/O modules flash green all the way down the rack. Diagnose the top of the tree before the branches.
- Separate field faults from module faults. Red field-power or per-point indicators point at your wiring and loads.
- Power-cycle once — deliberately, not repeatedly. If a solid red returns after a clean power-cycle, you have a hardware candidate.
- Swap-test if you can. Moving a suspect module to a known-good slot, or a known-good module into the suspect slot, resolves module-versus-slot in two minutes.
Ship a module when it is solid red after a clean power-cycle, dark in a rack that is otherwise powered, or has visible or olfactory damage — discoloured components, a burnt smell, a bulged capacitor.
What it costs, and what is worth repairing
As of September 2026, PLC module repairs — digital and analog I/O, communication modules and specialty cards for SLC 500, ControlLogix, CompactLogix, PLC-5, MicroLogix and Siemens S7 — typically run $250–$650, with a free bench evaluation and a firm quote before any work. Rack power supplies are a flat $349. Rush service is $149 and emergency service $399, and every repair carries the 24-month warranty. Broader pricing sits in PLC repair cost, and the supply side in industrial power supply repair cost.
The economics are clearest on obsolete analog and specialty cards, where a replacement may not exist at any price and a repaired card is the only way to avoid re-engineering an I/O scheme. They are least clear on commodity discrete I/O, where we will sometimes tell you the module is not worth the freight — that is a legitimate answer and you should expect a shop to give it.
The broader case for repairing rather than replacing control hardware is not sentimental. Producer prices for machinery and equipment (bls.gov) have outrun the general index for years; manufacturing capital expenditure data (census.gov) shows how much of a plant budget is committed before an unplanned failure arrives; and control-system availability now sits inside the operational-resilience conversation that industrial cybersecurity guidance (cisa.gov) frames around keeping systems supportable rather than merely current.
While the rack is open
If the rack contains drives as well as I/O, a drive throwing its own fault at the same moment is worth checking before you conclude the PLC is at fault — a PowerFlex reporting F004, F005 or F012 on the same network produces symptoms that look like controller trouble from the operator station. Those codes are decoded in PowerFlex fault codes F004, F005 and F012.
When you are ready to send a module, the mail-in instructions cover labelling and the packing guide covers the rest. Ship the module only, with the fault description and what the indicators were doing — a module that arrives with "it was flashing red at 6 AM Tuesday, controller was faulted" attached gets evaluated considerably faster than one that arrives bare.
Frequently asked questions
What does a flashing green OK light mean on a ControlLogix module?
It means the module is powered and healthy but has no connection to an owning controller. The usual causes are a controller that is not in Run, a project that does not include the module, a catalog-number or series mismatch that electronic keying is refusing, a module in the wrong slot, or a network path problem on a remote rack. It is almost never a hardware failure, so the module should not be shipped for repair on that basis alone.
What does solid red mean on a ControlLogix I/O module?
Solid red on the OK indicator means the module has detected an unrecoverable internal fault. Power-cycle the rack once and observe. If solid red returns after a clean power-cycle, the module has genuinely failed its self-test and is a legitimate repair candidate.
My whole rack is dark. Is the processor dead?
Probably not. A rack with no indicators lit anywhere is a power problem, and the rack power supply is the overwhelmingly likely culprit — aged filter capacitors, a failed rectifier stage, or a dead cooling fan that let it cook. Rack power supplies repair well and are a flat $349.
Is a red output-module fault light the module or my wiring?
Usually your wiring. Output modules with electronic protection report a fault when the field side is shorted or overloaded — a jammed solenoid, a chafed cable, a failed coil. Clear the field-side fault and the indicator clears. If it persists into a disconnected, unloaded output, then suspect the module.
Can you repair a ControlLogix processor?
We test and verify them honestly, but most CPU faults are not economically repairable and we say so rather than taking the work. I/O modules, communication modules, specialty and analog cards and rack power supplies are a different story and repair well.
How much does PLC module repair cost?
Typically $250–$650 for I/O, communication and specialty modules, with a free evaluation and a firm quote before any work begins. Rack power supplies are a flat $349. Rush service is $149 and emergency service is $399.
Should I power-cycle before calling anyone?
Once, deliberately, after you have photographed the rack with the fault live. A single clean power-cycle distinguishes a recoverable fault from a hardware failure. Repeated power-cycling in the hope that it clears mainly destroys the evidence that would have identified the cause.
Have this exact problem?
Start a repair or get a free technician review — we'll tell you honestly whether it's worth fixing.