I'm going to say something that will annoy a few people: a bad contactor gets blamed for a "PLC fault" far more often than the PLC itself. I've seen it in control cabinets with every major brand, and I don't blame the processor. It doesn't matter if you run an Omron PLC, an older relay system, or a body control module in a truck. If the contactor is tired, you'll get intermittent faults that point everywhere but the actual part.
I'm a quality/compliance manager at an industrial automation company. I review every unit that comes back through our returns door—roughly 200 items a year (or rather, 180 to 200, depending on season). In our Q1 2024 audit, 23% of "defective PLC" returns had no PLC fault at all. The real issue was usually a contactor on the load side, a loose terminal, or a wire that had turned green inside the insulation.
One case that stuck with me was a Freightliner Columbia AC control panel. The customer said the blower didn't run reliably. Another shop had already replaced the HVAC control module and was about to replace the body control computer. When the panel reached me, the blower motor contactor had high-resistance main contacts. The coil pulled in fine, but under a 12 A load the voltage dropped almost 2 volts across the poles. It wasn't the controller. It wasn't the PLC. It was a $28 contactor with worn silver tips.
The first tool I grabbed was a Micro Ferrets multimeter. Not because it's fancy—because it's the one I trust for a quick coil resistance check and a loaded voltage-drop test. The old habit of staring at a contactor and tapping it with a screwdriver doesn't cut it anymore.
How to Check a Contactor in Six Steps
Here's the process I use before I even think about ordering a replacement PLC. It doesn't take long, and it saves a lot of downtime.
- Lock out and tag out. This is not optional. Per NFPA 70E, always disconnect and verify zero energy before opening a panel. A contactor can be held in by a stuck coil or a shorted output.
- Look at it. Check for blackened terminals, melted plastic, arc tracks, and loose lugs. A contactor can look fine on the outside and still be dead inside, so don't stop here.
- Check the coil. Measure resistance between A1 and A2. A 24 VDC coil might read a few hundred ohms; an open coil reads OL. Compare with the datasheet. In my first year, I made the classic beginner error: I checked a coil with the multimeter on the wrong range, read OL, and nearly blamed a PLC that was fine. On an Omron PLC CP1H, if you're switching the coil directly, also verify you have a flyback diode across the coil. Without it, the inductive kick can eventually kill the output.
- Check the contacts without power. Put the meter on low ohms and check the NO contacts; they should read near 0.0 ohm. If the contactor has a NC auxiliary contact, verify it reads open when the coil is de-energized. But an unpowered contact can read 0.1 ohm and still fail under load. This is where the conventional wisdom starts to fall apart.
- Do a loaded voltage-drop test. This step changes everything. Everything I read as a beginner said a good contact is just "closed." In practice, I found a contactor that read 0.0 ohm on the bench, then dropped 1.8 volts under a 12 A load. That contact was carrying the load through about a square millimeter of good surface; the rest was carbon. With the load running, measure across each pole. If you see more than 0.1 volt per pole (check manufacturer spec), the contact is bad.
- Cycle it. Energize the coil at least ten times. Listen for a clean pull-in. If it chatters, hesitates, or gets hot after cycling, replace it.
What the PLC Really Tells You
I like the Omron PLC CP1H for small machines. It's affordable, reliable, and the programming software is easy enough for a mechanic to follow. But it's not magic. The transistor outputs are rated 0.5 A per point at 24 VDC, according to Omron's CP1H datasheet (omron.com, accessed February 2025). If you're switching a contactor coil directly, you need a flyback diode or a proper suppressor. I've seen CP1H outputs die because someone skipped that.
That said, I've also seen "dead" PLC outputs that were actually a bad contactor coil pulling too much current and welding the output transistor closed. If the output LED is on but the load doesn't run, the issue is usually somewhere between the PLC terminal and the load. The PLC only tells you what it's trying to do, not what actually reaches the coil.
The Way We Check Contactors Has Changed
What was best practice in 2020 doesn't always cover 2025. Five years ago, a lot of field techs would check a contactor by listening for a click and calling it done. That was already borderline, but today's compact contactors run closer to their thermal limits, and the tolerances are tighter. The old screwdriver-tap method misses the failure modes that actually happen: worn silver tips, cracked springs, and corrosion inside the arc chamber.
The fundamentals haven't changed, though. You still need the right coil voltage, a clean mechanical connection, and contacts rated for the load per IEC 60947-4-1. The difference is in the tools and the willingness to do a load test. That's a change in execution, not in physics.
The "But the PLC Output Is On" Excuse
I hear it constantly: "The PLC output LED is on, so the PLC must be fine." That only tells you the processor is trying. It doesn't tell you if voltage is reaching the coil through the terminal block, fuse, relay, and wire. I've lost count of how many times I've opened a panel and found the output LED on with 0 V at the coil because the wire had backed out of a cage clamp.
Am I saying you should never replace a PLC? No. I've rejected PLCs with swollen capacitors and dead outputs. An Omron PLC CP1H is a good little controller, but it's not indestructible. I just think the contactor is the easier thing to eliminate first. Replacing a processor is expensive, and it doesn't fix a pitted contact.
Stop Blaming the PLC
After we replaced that contactor in the Freightliner Columbia AC control panel, I kept second-guessing myself. What if the body controller also had a latent fault? I didn't relax until we cycled the blower forty times and the panel passed a full load test. The upside of swapping a $28 contactor was small. The risk of not swapping it was a return truck and a customer who'd already burned one shop. That math is easy.
Here's my final position, and I won't soften it: before you touch the PLC program, before you order a new processor, check the contactor. If it's bad, replace it. If it's good, move on. Because I've seen the same pattern over and over—a "dead PLC" that was really a tired contactor—I'll keep saying it until the industry catches up.
