Three ABB VFD Breakdowns Cost Us $18,000 — The Drives Weren't the Problem

Nine forty-seven on a Wednesday morning. The bagging line stopped. Third stoppage in fourteen months where an ABB VFD was at the center of it, and I was already mentally drafting the report to plant manager: replace every ABB drive in the building with a different brand.

I had maintenance logs, fault codes, timestamps. What I did not have was a root cause for any of the stops.

Then our senior controls tech asked a question I did not want to hear: 'Did these drives cause the stops, or did they detect the problems?'

I didn't have an answer. That was the problem. It ended up costing us roughly $18,000 to learn what I'm about to put in this checklist.

The review: three stops, zero failed drives

Let me be precise. The drives tripped. All three stops were real. But in each case, the ABB VFD was doing exactly what it was designed to do. It detected a condition that should have been caught long before installation.

Mixer, April 2024. The motor nameplate said 30 kW, so I ordered a 30 kW ABB VFD, standard duty. No one asked if the load was constant torque or variable torque. No one asked how often the batch got loaded, or by how much the motor current spiked. It ran about two months before it began tripping on overcurrent during heavy batches. The drive wasn't bad. It was undersized. The overload rating of the frame we bought was too low for the real load profile. We replaced it with a correctly sized unit after a conversation with an applications engineer. The original drive has been running a simple fan in another building since last September. Same drive. Lighter duty. Zero faults.

Centrifuge, September 2024. We asked for a quote on an ABB VFD for a high-inertia centrifuge and sent the supplier only the motor power and the part number. During every deceleration, the drive tripped on DC bus overvoltage. Our first reaction was to blame the product. The real issue is that a centrifuge stores kinetic energy. When you decelerate it faster than the mechanical system can absorb, that energy feeds back into the drive's DC bus, the voltage rises, and the drive trips to protect itself. The supplier never asked about stopping time, inertia, or whether we needed a braking resistor. One phone call with someone who understood the application would have caught it before we ever ordered anything.

Bagging line, January 2025. The drive displayed an input phase loss fault and stopped the line. We replaced it with a spare. The new drive displayed the same fault. Only then did someone check upstream and find a worn contactor dropping one phase under load. The original drive had detected the condition correctly. We reinstalled the original drive after replacing the contactor. It has run fine since. The spare sits on a shelf, waiting for a failure that never happened.

What surprised me wasn't that we made three separate mistakes. It was that none of the suppliers who quoted those drives asked a single question that would have prevented them.

The real cost of buying VFDs as commodity parts

When I went back through the invoices, the total landed at roughly $18,000. That includes the larger drive for the mixer, an unnecessary spare, an emergency service call, freight, and about 14 hours of line downtime. It does not include the overtime our own crew worked, the meeting where the plant manager asked me why we keep replacing 'failed' drives, or the credibility I lost with operations.

The hardware failure rate was zero. The process failure rate was three out of three.

Somewhere in that review, the phrase that stuck with me was: if you buy a VFD like it's a circuit breaker, you get supplier support like it's a circuit breaker. You can't expect someone to help you solve an application problem if they were only given a part number and a request for the lowest price.

What we do before ordering now

This is the checklist I keep in the maintenance office now. It is short, but it would have caught all three problems:

1. Send motor data, not just power. FLA, voltage, service factor, and cable run length matter.
2. Describe the load: constant torque or variable torque, starts per hour, stopping time, ambient temperature.
3. List the physical requirements: panel or wall-mount, enclosure rating, keypad, fieldbus, braking options.
4. Ask the supplier what they know about the application. If they quote instantly without questions, that tells you exactly how much support you'll get after the sale.

We don't use this for every single order. If I'm replacing a drive on a small fan in a clean electrical room and my controls engineer is on site, a quote-only supplier can work fine. But for critical process lines, I now look for an ABB VFD supplier who asks questions before they ask for a purchase order. That difference is worth more than the price difference between quotes.

The way I see it, a VFD is not a disposable part. It's a control system that happens to have a power stage. Treating it like a commodity is how you end up with spare drives that are not needed, and with three line stops that never had to happen.

I'm not claiming ABB industrial VFD drives are bulletproof. No industrial electronics are. But I stopped blaming the hardware for problems that our own spec process and supplier choice caused. Take this with a grain of salt: I'm not a drive application engineer. I'm the guy who broke things first.