Alarm 13 Is an Overcurrent Trip—Start at the Load, Not the Drive
On a Danfoss VLT drive, Alarm 13 is an overcurrent fault. The drive saw current above its internal limit and tripped to protect the IGBTs. On fan applications—rooftop exhaust, process cooling, combustion air—the drive is almost never the root cause.
In the fan-application Alarm 13 calls I've coordinated since 2022, the majority traced back to the load side: motor insulation, cable damage, or a fan that was mechanically harder to turn than it looked. The drive itself was the problem in a small minority of cases.
So here's the short version before the long one: check the fan load, the motor insulation, and the cable terminations before you order a replacement VFD. That order is cheap-to-expensive for a reason.
If you're already under a production deadline—which is usually why anyone reads this at 11 p.m.—skip to the check order below. The case study further down explains why I stopped trusting the phrase "the fan is fine."
Why Fan Applications Trip Differently
Why does a VFD fan trip on overcurrent more often than a pump of similar rating? Inertia, mostly.
A centrifugal fan has a large rotating mass and a load curve that rises with the square of speed. On startup, the drive has to accelerate that mass through the breakaway region, and the current can spike well above nameplate before the fan reaches operating speed. If the ramp time is set too short, or the current limit is set too tight, the drive trips at 13 even though nothing is electrically wrong.
A few things make this worse, and they're all mechanical:
- Dirty or unbalanced fan blades—extra mass, extra vibration, extra current
- Bearing wear—small at first, then the fan no longer freewheels
- Belt tension on belt-driven units—too tight is as bad as too loose
- Damper or louver linkage that sticks partway open
None of those show up on a drive display. All of them push Alarm 13.
The Order I Check Things (Cheap to Expensive)
Three things: safety, insulation, cable. In that order.
1. Safety first—and this means more than flipping the disconnect
Before I touch anything in a drive cabinet, I verify dead with a non-contact voltage tester and then confirm with a multimeter. Not one or the other. Both.
How does a non-contact voltage tester work? The sensor inside detects the electric field radiating from an energized conductor through capacitive coupling. It doesn't need to touch metal. That's convenient, and it's also why it can mislead you. A shielded cable or a grounded metal raceway can damp the field enough that the tester reads dead on a live conductor. Induced "ghost" voltage from parallel runs can make it read live on a dead one. It's a screening tool, not a confirmation tool.
For confirmation I use a CAT III or CAT IV rated multimeter—the CAT rating is defined in IEC 61010-1, and a CAT II meter has no business inside a 480 V drive cabinet. If you're looking for a reference for the symbols on the dial, a multimeter symbols PDF is fine for a quick refresher, but the meter's own manual is the only version you should trust for your specific model. The symbols that matter in this context: V~ for AC volts, V⎓ for DC volts, Ω for resistance, and the CAT rating printed near the input jacks.
NFPA 70E and OSHA 29 CFR 1910.333 both apply here. Lockout/tagout, verified absence of voltage, appropriate PPE. If that's not your daily work, get someone who does it daily.
2. Motor insulation—megger before you theorize
Disconnect the motor from the drive. Yes, physically. A megger test with the drive still connected can damage the drive's output stage, and it won't give you a clean reading anyway.
Test phase-to-phase and phase-to-ground. IEEE 43-2013 gives the minimum acceptable insulation resistance as roughly 1 MΩ per kV of rated voltage plus 1 MΩ—so a 480 V motor wants at least around 1.5 MΩ, and I don't like seeing anything under 5 MΩ on a motor that's supposed to run another five years.
Moisture in the windings, cracked insulation, or a motor that's been sitting in a humid penthouse for two seasons will all show up here, and all of them will trip Alarm 13 under load.
3. Cable and terminations
Look at the motor terminal box before you look at the drive parameters. Loose lugs, corroded ring terminals, cable jacket damage where it enters the conduit—those are all intermittent overcurrent sources, and they're free to fix.
Also check cable length against the drive's recommendations. Danfoss publishes maximum motor cable lengths in the FC 102 Operating Guide, and long unshielded runs on a fan application are a common cause of nuisance trips. If the run is marginal, shielded VFD cable and a proper 360° gland termination solve more problems than any parameter change.
4. Parameters—but only after the hardware checks out
If insulation and cable are clean, then look at the drive. On an FC 102, the parameters I actually touch on a fan Alarm 13 call are the ramp time (parameter 3-41), the current limit (4-18), and the motor data (1-20 through 1-25). If the motor nameplate data was entered wrong—or never entered at all—the drive's internal model of the motor is wrong, and its current limit calculations will be wrong with it.
One caution: raising the current limit to stop a trip is treating a symptom. If the fan is drawing more current than it should, something downstream is loading it up. Find that first.
5. The fan itself
Spin it by hand with the power off and the drive disconnected. It should turn freely and coast for a while. If it stops almost immediately, you've found your overcurrent source. If it grinds or wobbles, same answer.
The Communication Failure That Cost Us a Weekend
In March 2024, a food-processing plant called at about 11 p.m. on a Friday. A 22 kW Danfoss VLT FC 102 driving a rooftop exhaust fan had tripped on Alarm 13. Production was scheduled to restart Monday morning. Normal lead time for a replacement drive in that frame size was three to five business days, so the clock was already against us.
I asked the maintenance tech on the phone what the fan was doing. He said: "The fan is fine. It's the drive."
We were using the same words but meaning different things. He meant the fan blades were intact and the fan was spinning when he watched it. I heard "the mechanical load is normal." What he hadn't checked was the fan bearing—and it was already rough enough that the motor was pulling about 15% above its normal running current before the drive tripped.
We replaced a drive that didn't need replacing. The new unit tripped on Alarm 13 within forty minutes of startup. Then we pulled the fan, found the bearing, and had it back together by Sunday afternoon. We paid a weekend premium on the bearing and ate the cost of restocking the drive. Call it a lesson learned the hard way.
Looking back, I should have asked him to spin the fan by hand before we shipped anything. At the time, I was focused on the drive because that's what the display was showing. But the display was reporting a symptom, not a cause.
We didn't have a formal trip-log process then. We do now. Every Alarm 13 call gets a one-page record: fan model, bearing condition, megger reading, cable length, parameter snapshot. The third time I had to ask a customer the same questions from memory, I finally built the form. Should have done it after the first time.
When to Stop and Call Someone Else
I'm not a motor design engineer, so I can't speak to winding tolerances or reclosing duty on a specific motor frame. What I can tell you is the field check order that gets a fan application back online fastest.
This gets into territory outside my experience when:
- The drive is still under Danfoss warranty—open it and you may void it
- The installation is in a classified (hazardous) location—different rules entirely
- The motor is medium voltage or the cabinet is above 600 V
- You've meggered clean, checked the fan, replaced the cable, and it still trips—that's a drive-level fault and it needs a service tech, not a parts swap
My experience is based on about sixty fan-application VFD fault calls, most of them in food processing and light manufacturing. If you're running combustion air on a boiler or a wind-tunnel fan, your load profile is different enough that some of this order may not apply.
One last thing, and it's the part I care about more than the technical detail. A 7.5 kW exhaust fan on a small bakery's oven line doesn't sound like a big deal until the oven shuts down and the whole production day is gone. I've taken weekend calls from two-person shops and from plants with a hundred employees, and the two-person shop has less margin for a Monday morning surprise, not more. Small doesn't mean unimportant—it means less room for error. That's why I ask the boring questions before I quote a replacement drive.
If you take one thing from this: on a Danfoss VLT fan application, Alarm 13 is a load-side conversation until proven otherwise. Megger the motor, spin the fan, look at the cable—then talk about the drive.
Leave a Reply