What a Bad Motor Taught Me About ABB Motors, VFDs, and Total Cost

Last May, I found out what a bad motor actually costs

The motor was humming. Not the smooth hum of a healthy induction motor, but the rough, grinding kind that tells you something is wrong. I was standing in our warehouse with a torque wrench in one hand and a universal joint socket in the other, trying to reach a coupling bolt hidden behind a guard. The motor was an ABB ERH motor from the original line, and at that moment I was pulling it off to replace it with one of the "lower-cost equivalent" units we had approved.

It was the third failure in seven weeks.

I'm a quality and compliance manager at an industrial equipment firm. I review motors, drives, and control components before they ship—roughly 200+ unique items per year. In 2022, I implemented a verification protocol that cut our field failures by a third. In Q1 2024 alone, I rejected 14% of first deliveries because dimensions, documentation, or performance data didn't match the contract.

So I know how painful this story is. I lived it.

Why we swapped the ABB motor for a cheaper one

It wasn't the ABB motor that failed. The original motor had been running for 11,000 hours and was still within tolerance. But our purchasing manager had a budget target. A supplier offered a "compatible" motor at 28% lower unit cost. The brochure said the frame size was right, the IP rating was right, the voltage was right. The CFO asked: "Why are we paying ABB motor company prices for something that's the same?"

Here's the thing: it wasn't the same. The mounting bolt pattern looked right, but the pilot diameter was 0.020" off. There was a note in the technical drawing about "optional" shims, which nobody noticed. And the motor's published torque curve didn't match the actual performance on our test rig.

Why does 0.020" matter? Because a coupling doesn't know it's a small number. It knows the shaft centerline is off. It amplifies the error at speed. What starts as a barely visible offset ends as a bearing failure.

I had two days to decide. The PO deadline for the next production run was approaching. Normally, I'd have requested a sample motor, run it through our acceptance test, and compared it against the original in a blind test. That would have taken three weeks. Instead, I relied on the supplier's data sheet.

Before that, I thought of ABB motor company as a brand for projects with slack in the budget. Not exactly luxury, but not the default either. I treated the extra cost as overhead. This incident changed my view.

What VFD stands for—and what it doesn't do

If you've ever wondered what VFD stands for, the short answer is variable frequency drive. It controls motor speed by changing the frequency and voltage of the power going to the motor. On paper, the cheaper motor was "VFD compatible." In reality, that phrase meant something different to the vendor than it did to us.

We were using the same words but meaning different things. They heard "the motor can be used with a VFD." I meant "the motor can hold torque across the entire speed range in this duty cycle." Found out when we ran it at 30% speed for two hours. The bearings ran hot and the cooling fan wasn't effective at low speed. The VFD threw an overtemp fault because the motor current stayed high.

A VFD can mask a bad motor for a while. It can ramp current, limit torque, and hide a misalignment. But the motor still generates heat. At low RPM, you lose cooling. The drive doesn't fix a mismatch.

Not exactly what you'd expect from a "drop-in replacement."

The shaft, the socket, and the actuator decision

While we were sorting out the motor mess, I was also reviewing two quotes for the diverter gates in our bulk handling line. One was based on pneumatic cylinders. The other used electric rotary actuators. The pneumatic option was cheaper per unit, but the compressor had no spare air capacity. To use pneumatics, we would have needed a $12,000 compressor upgrade. The electric rotary actuator cost more up front and avoided the compressor entirely.

That's the same TCO logic I use for motors. You don't compare prices without asking what the component is supposed to do. You compare outcomes, risks, and the cost of failure.

The $22,000 lesson

The first cheap motor failed during final testing. It was a bearing issue—the misalignment caused by the pilot diameter and the thermal expansion at low speed. The motor was toast. The coupling was scored. The driven shaft was misaligned. We had to rework the entire drive assembly. The redo cost us $22,000 and pushed a customer launch out by a month.

When I pulled the original ABB ERH motor from storage and ran it on the same rig, it passed every check. That motor was still within its original tolerance after years of service. The "savings" from switching to a cheaper motor disappeared the moment it failed. Actually, they went negative.

I don't blame the supplier completely. We didn't ask the right questions. We treated a motor as a commodity part, and we paid for that assumption.

What I changed afterward

It's tempting to think you can just compare unit prices. But identical specs from different vendors can produce wildly different outcomes. The $1,200 motor quote turned into $6,000 after freight, machining a new adapter plate, VFD tuning, testing, and overtime labor. The $1,800 all-inclusive quote would have been cheaper. That's what total cost of ownership means.

The question everyone asks is "what's the price?" The question they should ask is "what's included in that price?"

Since then, I've added three questions to every motor or actuator quote: What test data supports this claim? How does it perform at low speed? What happens if it fails? Per FTC guidelines (ftc.gov), marketing claims need substantiation. I stopped accepting adjectives as evidence. I want curves, test reports, and tolerances.

Per FTC Business Guidance, claims must be truthful, not misleading, and substantiated with evidence. That's a good rule for procurement, too.

We also created a standard verification process that includes a sample motor test for any new supplier. Is it slower? Yes. Does it cost money? Yes. But compared to a $22,000 redo, the cost of testing is small.

Did we go back to ABB for every motor? No. But for the duty cycle that failed, I still write an ABB ERH motor into the spec. The test data exists, and I can verify it. Some applications don't need that level of data or duty cycle. But now we choose based on risk, not just price. And we don't use "VFD compatible" without agreeing on what that means. If you're still wondering what VFD stands for, keep reading—but also ask what the whole system costs, because that's the number that matters.