Here's an opinion that has made me unpopular in more than one maintenance meeting: most Danfoss VFD alarm 14 faults are cooling fan faults — not drive faults. If you order a replacement drive before you check that fan, you are not fixing a failure. You're gambling with the production schedule.
Since 2021, my role has involved coordinating emergency Danfoss VFD repairs, rush parts, and field support for plants whose lines were down. I've handled over 200 urgent VFD shutdowns in that time, from an FC 102 on a small conveyor to large-frame drives running critical pumps. Alarm 14 comes up constantly. And in a surprising number of those cases, the drive that gets blamed is perfectly fine except for one thing: the cooling fan stopped doing its job.
Alarm 14 doesn't mean the drive is dead. It means the drive got too hot and shut itself down to protect the power components. This article is about what to check before you spend money, how to use a multimeter to verify the fan, and why paying extra for a guaranteed replacement is often the cheapest decision you'll make.
What a Danfoss VFD Alarm 14 Actually Tells You
On the Danfoss VFD fault list — for FC 102, FC 202, FC 302, and other VLT drives — alarm 14 is over-temperature. The thermistor on the heat sink has reached the drive's shutdown limit. The drive will not run again until that heat sink cools below the reset threshold.
That is the full extent of what the code tells you. It does not tell you the drive is broken. It tells you heat is building up faster than it can be removed. The usual causes are:
- A cooling fan that stopped, slowed down, or seized.
- A heat sink or air filter that is blocked with dust and debris.
- Ambient temperature higher than the drive's rated operating range.
- A power module that is failing internally and generating more heat than it should.
My own notes from the last few years are not scientific, but they are consistent. Of the alarm 14 emergency calls I've attended since 2022, roughly eight out of ten were traced back to a failed or weak cooling fan. Most of the remaining cases were blocked airflow or extreme ambient heat. A failed power module was rare, and usually it had been damaged by earlier overheating.
The frustrating part is that a fan rarely announces its own failure. It doesn't blow a fuse or throw a code. It just stops moving the air that keeps the IGBTs cool, and the first warning you get is alarm 14 on the display. That's why the diagnostic habit matters more than the part number.
Before You Order a Danfoss VFD Cooling Fan, Run These Multimeter Checks
When production is stopped, the urge to order a Danfoss VFD cooling fan immediately is strong. I understand it. But the smart order — and the one with the shortest total downtime — comes after ten minutes of verification, not before.
The good news: you do not need expensive test gear. A standard digital multimeter is enough. Here are the three checks I run.
1. Check the Supply Voltage at the Fan Connector
Find the cooling fan inside the drive and read its nameplate. Many of the fans used in Danfoss VFDs are 24 V DC units with a small plug-in connector, though some are AC types depending on the drive frame and model. Set your multimeter to the correct voltage range and measure across the fan supply pins while the drive is powered and the fan should be running.
If you have no voltage at the connector, a new fan won't help — the problem is upstream in the control or power supply circuit. If you have proper voltage and the fan still does not spin, move to the next check.
2. How to Check Current With a Multimeter on the Fan Circuit
Voltage tells you power is reaching the fan. Current tells you if the fan actually uses it. This is the step most people skip because they are not sure how to check current with a multimeter. It is not complicated.
For a DC fan, set the multimeter to DC amps and start on the highest current range, usually 10 A. Disconnect the positive fan lead and connect the meter in series. The current should flow from the supply positive lead through the meter and then into the fan positive wire. If you have a DC-capable clamp meter, clamp it around the positive wire instead. Then compare the reading with the current printed on the fan label.
If the current reads 0 A, the fan winding or its internal driver is open. If it is far below the nameplate rating, the fan is weak even if it spins freely by hand. If it is well above nameplate, the bearing may be tight, or the fan is mechanically stuck while the motor tries to turn it. Any of those is a valid reason to replace the fan.
Current tells the truth in a way that voltage alone cannot. I used the same approach on a Lippert control panel in an RV slide-out system, where the panel was shutting down under load. The current reading showed a binding motor drawing nearly double its rated amps. The control panel was doing exactly what it should have done. The meter saved an unnecessary control panel replacement.
3. Use Multimeter Diode Mode to Check the Drive's Power Section
If the cooling fan checks out, or a new fan does not stop the trips, the trouble may be inside the drive. That is when I switch my meter to multimeter diode mode — the symbol that looks like an arrow pointing at a vertical line.
In diode mode, the meter sends a small current through a semiconductor junction and displays the forward voltage drop. A healthy power diode reads roughly 0.4 to 0.6 V in one direction and open in the other. If it reads like a dead short in both directions, the component is damaged.
With the drive locked out and the DC bus fully discharged, use diode mode to check between the DC bus and each motor output terminal — U, V, and W. Compare the readings phase to phase. They should look similar and behave like diodes. If one phase is clearly different, especially a short in both directions, the power module is suspect. That may be why the drive keeps overheating.
This check is worth doing even when the fan looks guilty. I have seen alarm 14 return two weeks after a fan replacement because a degrading IGBT was dumping extra heat into the heat sink. The module needed to be replaced, and the meter showed it before the drive started tripping again.
If the Fan Is Confirmed Bad, Buy Certainty — Not Hope
Now we get to the point where I annoy people. When the fan is confirmed bad, I do not recommend the cheapest fan from the slowest warehouse. In an emergency, I recommend the part that comes with a date. The extra money for expedited or guaranteed supply is buying certainty, and certainty has a measured value when a line is down.
Here is a real example. In March 2024, a packaging plant called me on a Thursday afternoon. Their FC 102 had tripped on alarm 14, and the cooling fan bearing was seized. I ordered the matching Danfoss fan kit from an authorized distributor and paid for next-morning delivery. Total with freight was around $190. The alternative was a compatible fan from an online marketplace that would probably arrive by Monday or Tuesday. The plant's line was running 24/7, and the finance manager told me every hour of unplanned downtime was roughly $2,200.
The fan was installed and the line was running by Friday noon. The $190 option looked expensive until you multiplied $2,200 by however many hours 'probably Monday' could have cost. A dated guarantee is not the same as speed for its own sake. It is the difference between a promise and a guess.
So Should You Always Buy the Genuine Fan? No.
To be clear, this is not a 'genuine or nothing' article. I have specified good-quality replacement fans in applications where the operating conditions were mild, the lead time was flexible, and the specification matched. If downtime is not at stake, the lowest total cost of ownership may be a quality generic fan.
But two conditions have to be met. First, the electrical and mechanical specification must match: rated voltage, current, airflow, static pressure, speed, connector type, and maximum operating temperature. A fan that fits the bracket but moves less air is not a spare part; it's a future alarm 14. Second, the delivery date has to be reliable — not an estimate. In a true emergency, an estimated arrival date is not a solution.
One maintenance manager summed it up better than I ever could: 'A fan that probably gets here Tuesday is not the same as a fan that will be here Tuesday.' The price difference is small. The production risk difference is enormous.
Here is my bottom line, and I will not soften it to sound balanced. Treat alarm 14 as a cooling system problem first. Confirm the cooling fan with voltage and current checks. Suspect the power module only after diode-mode tests show something wrong. And when the fan is bad and the line is waiting, pay for a guaranteed arrival date, not a maybe.
The multimeter in your tool bag is the cheapest diagnostic tool you own. The most expensive part in this story is not the drive. It's the downtime caused by guessing.
