Danfoss VFD Not Working in Auto Mode? Alarm 14? Here's What Actually Causes These Faults

I'm the quality compliance manager at a drives and motion control supplier. Every Danfoss VFD that goes out our door crosses my desk first—roughly 2,000 units a year across VLT drives, servo drives, and controllers. In 2025, I've already rejected about 4% of first shipments. The reason is almost never hardware. It's wrong labels, mismatched documentation, or spec discrepancies.

That same pattern shows up on the return side. When a customer's system stops working, the VFD gets blamed first. But after four years of reviewing warranty claims, about 60% of the "broken" drives we receive back test perfectly fine on our bench. The drive wasn't the problem. It was the canary.

Here's the thing: there's no single answer to "why is my Danfoss VFD not working?" The right fix depends entirely on your symptoms—and on what happened right before the fault. Let me walk you through the four situations I see most often.

When I first started reviewing returns, I assumed any alarm code meant the drive hardware was dead. I was wrong. A customer sent back an FC-51 with alarm 14 in the fault log, frustrated about a line that was down for a week. I bench-tested it: fully functional. The real problem was a motor cable that had chafed through against a cable tray. That lesson stuck with me.

Scenario 1: The Drive Won't Start in Auto Mode

The most common complaint I get is, "It works in Hand, but in Auto nothing happens."

That symptom alone tells me the power stage is almost certainly fine. The drive can run a motor; it's just not receiving a valid run command in Auto. The problem is usually in the control wiring or the terminal configuration.

On Danfoss VLT drives, Auto mode relies on the control terminals. Terminal 18 is the default start command on most FC-51 and FC-302 models. If it never sees 24V, the drive will sit there doing nothing forever. I've seen dispatch tickets where someone checked the PLC program, checked the cable, then blamed the drive—only to find the wire had come loose on terminal 18 itself.

The second place to look is the coast input. Many setups use terminal 27, which defaults to "coast inverse" on some models. That means when it's not receiving 24V, the drive is inhibited from running. If someone reconfigured a parameter on an earlier machine and that setup got copied over, you can end up with a drive that powers up, accepts settings, but ignores start commands.

Here's what I'd do, in this order:

1. Put the LCP in Auto and send a start command from your control system.
2. Measure 24V between the start terminal and the common terminal (20) with a multimeter. No voltage means the wiring or PLC has a problem.
3. Check the terminal parameter assignments (parameter group 5-1x on most VLT models). Make sure the terminal you're using is actually configured as "start."
4. If you're using fieldbus or serial control, verify the drive is configured for bus commands rather than hardwired inputs.

Replacing the drive doesn't fix any of those. But I've seen it done more times than I can count.

Scenario 2: Alarm 14 — the Ground Fault Nobody Believes

Danfoss alarm 14 is a ground fault. The drive detected current leaking to ground on one of the motor phases. It's one of the most common alarms I get asked about—and also the most misunderstood.

The old belief is, "Alarm 14 means the drive shorted out and killed the motor." That thinking comes from an era when drive output stages were more fragile and motor insulation was often worse. Modern drives are more sensitive, and that's a feature, not a defect. They catch insulation problems before those become catastrophic phase-to-phase shorts.

In most alarm 14 cases I've reviewed, the root cause is one of three things:

• Damaged motor cable insulation. The cable chafed through where it passes a metal edge, a strain relief, or a cable tray divider. Moisture seeps in and you get intermittent ground faults that come and go with humidity. This is the most common cause, and it's easy to miss because it never trips while the machine is idle.
• Moisture in the motor junction box or conduit. Washdown environments are a classic source. We had a customer in Q1 2024 whose FC-302 was tripping alarm 14 every week. They wanted to return the drive. Then they sent us a photo of the motor junction box: it was dripping water because a washdown nozzle was aimed right at it. Re-routing the nozzle fixed the "drive problem."
• Motor winding insulation breakdown. An induction motor running on a VFD for years, especially with repeated over-temperature, eventually degrades. The drive will report alarm 14 long before a conventional starter would ever notice.

The fix is an insulation test—not a drive replacement. Disconnect the motor cables from the drive. Megger the cable and the motor windings separately. According to IEEE 43, the recommended minimum insulation resistance for a 480V motor is roughly 1.5 megohms at operating temperature (the formula is kilovolts + 1 megohm). A healthy motor will show much more—20 megohms or higher. If both the cable and motor are clean, inspect the brake resistor and brake wiring before you suspect the drive itself.

In 2025, about 70% of the alarm 14 returns we get test fine here. I'm not saying the drive is never the problem. I'm saying check the path from the drive to the motor before you pull it.

Scenario 3: It Runs, but the Motion Is Wrong

Sometimes there's no alarm at all. The drive runs, the motor spins, but the machine jerks, loses position, or stalls. The instinct is to start tweaking VFD parameters. I always tell our support team: walk over and look at the mechanical side before you touch a parameter.

Timing belt wear: the silent culprit

A timing belt—also called a synchronous belt—is designed to move without slip. Its teeth lock into pulley grooves. But teeth wear, strip, and tension drifts, and once that happens the machine starts behaving bizarrely.

A worn timing belt produces symptoms that look exactly like an electrical problem:

• Jerky or oscillating motion, like a poorly tuned servo loop
• Overcurrent trips, because the motor is fighting a belt that's skipping
• Positioning drift, because the tool is no longer where the controller thinks it is

If you're doing a timing belt repair, don't just swap the belt and call it done. Check the pulleys for worn teeth—a pulley with rounded teeth will destroy a new belt within weeks. Check shaft alignment too. Too much misalignment adds lateral stress to the belt and to the motor bearings.

Tension matters as much as anything. Too loose, and the belt skips teeth under load. Too tight, and you're pulling motor bearings. The belt manufacturer's deflection spec is a good starting point; for critical axes, a tension gauge is a no-brainer compared to the cost of a second failure.

Motor and controller mismatch

There's another "motion is wrong" case that isn't mechanical: the motor doesn't match the job. This is where questions like "what's a stepper motor?" and "do I need a servo?" come in.

Scenario 4: You're Choosing a Replacement or New Controller

When you're buying, there's no universal answer. It depends on your speed, torque, duty cycle, and who's going to maintain the thing.

What's a stepper motor?

A stepper motor is a brushless DC motor that moves in discrete steps—usually 200 steps per revolution (1.8 degrees per step). Each pulse from the controller advances it one step. Because the controller counts steps, it knows where the motor should be, without needing an encoder.

That's the open-loop advantage: simple, repeatable positioning at low-to-moderate speed, and no feedback hardware. Steppers are great for many indexing, feeding, and pick-and-place axes.

But open-loop has a weakness. If the load exceeds the available torque, the motor stalls and the controller doesn't know it. It just keeps sending pulses, and your position drifts with no alarm. That's the deal-breaker in applications where you can't tolerate a missed step.

If you expect high acceleration, high speed, or variable loads, that's what servo drives are for. They use feedback to verify actual motion and correct on the fly. You pay more, and you have to set them up properly, but they're the right choice for dynamic or uncertain loads.

When a programmable linear actuator controller makes more sense

Here's the option people forget: if your motion is linear and the stroke is relatively short, a linear actuator with its own programmable controller is often simpler than a rotary motor plus belt and mechanical conversion. You program positions, speeds, and dwell times directly on the controller. Fewer components, fewer failure points, less maintenance.

I had a customer in 2023 debating between a stepper-driven belt system and a programmable linear actuator for a 6-inch indexing move. On paper, the belt system was cheaper. They went back and forth for two weeks. But their machine already had three critical alignments, and adding a belt meant a fourth. The linear actuator won because it removed a variable—not because it was the cheapest. They ended up doing one timing belt repair later anyway, on a different axis where a belt was unavoidable.

Which Scenario Are You In?

Quick triage:

1. Won't start in Auto but fine in Hand → control wiring or terminal configuration. Multimeter before credit card.
2. Trips alarm 14 → inspect the cable and motor insulation path first—megger and visual check. Drive last.
3. Runs but motion is wrong → inspect the mechanical path first: belts, pulleys, alignment. Then question the motor and controller pairing.
4. Buying something new → be honest about speed, torque, duty cycle, and who's going to maintain it.

The drive is the canary, not the coal mine. That's the phrase I repeat to our support team. It saves us a lot of wasted RMAs.

Bottom Line

The fundamentals of fault diagnosis haven't changed in recent years—you still trace the symptom back to the source. But the execution has transformed. Modern drives trip more often because they're more sensitive. They warn you about a damaged cable that an older drive would've ignored until the motor caught fire. That sensitivity isn't a defect. It's the product doing its job.

So if your Danfoss VFD is showing alarm 14 or refusing to run in auto mode, step back and check the wiring, the insulation, and the belt before you order a replacement. In my experience, most of what looks like drive failure in the field is something else entirely.