Danfoss VFD Alarm 16 Shut Our Line Down for 5 Days—the Fix Was a $20 Ball Bearing

The phone rang at 6:40 AM. I knew before I answered that it was the plant—the night shift had already called once at 4 AM, and I'd told them to try a reset. They'd tried it twice more since then. Line 2 was dead, the morning shift would be walking in within the hour, and the display on our Danfoss VFD was blinking the same code it had been blinking all night.

Alarm 16: What It Actually Means

The drive was an 11 kW Danfoss FC-202 running a conveyor motor that had been in service for six years. The code was Alarm 16—ground fault. I knew that from memory, though I still keep the Danfoss VFD fault codes PDF bookmarked on my phone for the drive models I don't work with as often.

According to the FC-202 operating guide (available on danfoss.com), Alarm 16 means current flowing from the output phases to ground. In practice, that's usually caused by a chafed cable, moisture in a junction box, or motor winding insulation that's broken down. I'd seen all three before, and each time the fix was straightforward: find the damaged part, replace it, reset the drive.

So that's what I set out to do. The problem was, everything looked fine.

Mistake #1: Assuming the Drive Was Telling Me Everything

I checked the output cable visually. Clean. No chafing, no pinched spots, no discoloration. I measured insulation resistance. Acceptable. I checked the DC link, the output terminals, the motor connection box. Nothing out of spec.

I reset the drive at 7:40 AM. The line ran for about twenty minutes, then tripped with the same alarm. I reset it again—it ran for fifteen minutes and tripped again. That was my mistake, resetting it that third time. I wasn't troubleshooting anymore; I was playing odds. What are the odds it trips immediately again? Well, it did. It always does.

Part of me knew I should do a mechanical check. Turn the shaft by hand. Listen to the motor while it runs. In my first year as a drives technician—that was 2017—I made the classic mistake of treating every drive alarm like an electrical problem. Old habits come back faster than you'd think. A failing bearing seemed unlikely. The motor ran fine yesterday. Didn't it?

Mistake #2: Replacing the Wrong Part

By mid-afternoon I'd convinced myself the forty-meter shielded cable running through the overhead tray had a chafed shield inside a buried conduit. It's a common root cause. In my experience, it's the one that shows up when nothing else obviously explains a ground fault.

I ordered a replacement cable at 4 PM. $470, plus $60 for expedited shipping. Actually, $495 total—I'm not going to forget that number. The cable arrived the next morning. I installed it myself, re-terminated both ends, double-checked the grounding braid. The drive ran for forty minutes after I reset it. Then Alarm 16 came back.

That's when I should have stopped guessing. Instead, I called tech support.

Mistake #3: Calling Tech Support Before Calling Ed

I spent most of the afternoon on the phone with Danfoss product support. The conversation went in circles:

"Alarm 16. Ground fault."
"Have you verified the output cable?"
"Replaced it yesterday."
"Insulation on the motor windings?"
"Tested. Fine."
"Ground connection at the drive?"
"Checked and re-torqued."

Every answer I gave was correct. And every answer led nowhere. Both sides kept saying the same words—"ground fault," "insulation," "output current"—but we were speaking past each other. I was so focused on ruling out electrical causes that I wasn't hearing anything else.

Near the end of the call, the tech asked: "Did the motor make any unusual noise during cold starts in the past few weeks?"

And I said no. Which was true, technically—I hadn't noticed it. The line operator had mentioned the conveyor sounding "rougher" in the mornings, but I'd dismissed it. She wasn't an engineer. I was. (That thought still stings, honestly.)

The Twist: A $20 Ball Bearing

On day three, Ed walked over to the line. Ed has been at this plant for 23 years and can diagnose machinery with a screwdriver pressed to his ear. He asked what was going on. I gave him the short version: Alarm 16, new cable, no root cause found.

He walked to the motor, put his hand on the housing for a second, then reached down and turned the shaft by hand. Slowly. Twice. He stopped at the same spot both times, frowning.

"Bearings," he said. "Rear one. It's gone."

I started to explain—ground fault, not a mechanical symptom, the drive is telling us there's an electrical path—but he cut me off. "Get the end bell open and you'll see."

He was right. In hindsight, it was obvious.

For anyone who's never had to think about this: a ball bearing is a simple mechanical component that does an essential job. It's a set of steel balls held in a cage between an inner and outer race, letting a shaft spin at thousands of RPM with rolling friction instead of sliding friction. Small, cheap, and one of the most failure-prone parts in any rotating machine.

If you look at a typical AC motor diagram, the bearings sit in the end bells—one at the drive end, one at the fan end—supporting the rotor shaft on both sides. When the rear bearing starts breaking down, it sheds metal particles into the motor housing. Those particles can bridge the winding insulation to the motor frame. That's the ground path the VFD detected.

The drive wasn't lying. Alarm 16 was real. But the root cause was a $20 bearing that had been wearing for months. Every day it looked like "fine" on the outside, and every day it was getting closer to the moment when it turned into an electrical fault.

Ed had the new bearing installed in three hours. Total cost: the bearing and less than half a day of his labor. The line ran through the night without a single trip. On my desk sat an $1,850 quote for a replacement motor that I'd been about to order.

The surprise wasn't that the bearing failed. The surprise was how easy it was to check once you knew to look—and how expensive my refusal to look turned out to be.

What I'd Do Differently (and Maybe You Should Too)

Let me be honest: I've made this kind of mistake more than once. In 2017, I blamed a bad cable when the real problem was moisture. In 2021, I blamed a motor when the coupling was misaligned. This time was the one that stuck.

Here's the short version of everything I've learned since:

• Read the whole fault code table, not just the first line. The Danfoss VFD fault codes PDF for your specific drive series lists likely causes for every alarm. Alarm 16 might say ground fault. It doesn't say "cable" or "winding." You still have to check every item, including the mechanical ones, before you decide which one it is.
• Check the mechanical side first. Turn the shaft by hand. Listen for bearing noise. Look at the coupling. The motor is part of the system, and the drive can only tell you what it measures electrically. I keep a printed AC motor diagram in the maintenance binder now—partly as a reminder that the motor is a physical machine, not just a box of windings.
• Think in total cost, not unit price. The $495 cable was a write-off. The $1,850 motor would have been worse. But the real number that matters is five days of line downtime. If you measure it that way, spending 20 minutes checking a bearing is the most cost-effective diagnostic step available.
• Listen to the people who run the machines. The operator heard the roughness in the morning start-ups. Ed felt it in the shaft. Neither of them needed a fault code table to know something was wrong. I needed to set my ego aside and hear it from both of them.

This discipline doesn't apply only to Danfoss VFDs. If you run servo motor controllers anywhere in your facility, the same thinking applies. A servo drive may throw a position error or a velocity deviation when a coupling is loose or a linear rail is contaminated—not because the drive is broken, but because the mechanics are feeding it bad information. Same root cause. Same trap.

The $20 lesson. I just wish it didn't cost me five days to learn it.