The Woolly Bear Problem: Why Our Interroll Conveyor Parts Kept Failing

If you grew up somewhere with cold winters, you've probably heard of the woolly bear caterpillar. The old folklore was simple: the wider the brown band in the middle, the milder the coming winter. We checked those stripes every fall when I was a kid. It was fun, but it didn't work. Because a caterpillar's stripe pattern isn't a forecast—it's mostly a reflection of how old the larva is and how much it ate.

I think about that every time a conveyor roller fails. We stare at one fuzzy symptom—a jammed sorter, a dead drum motor—and make a prediction. We replace the part and assume the next few months will be fine. Then it fails again, and we're back to checking stripes.

I'm an office administrator, not a maintenance engineer. My job is purchasing. For the last five years I've managed parts orders for a mid-sized conveyor integration company—about $1.2 million a year, maybe $1.3 with all the rush orders, across 30-odd vendors. I don't have a black belt in mechanics. What I do have is a lot of history with orders that looked right and were still wrong.

So let me walk through what I learned the hard way.

The Surface Problem: A Roller Died. We Replaced It. It Died Again.

Last April, one of our motorized rollers in a sorter stopped turning. The unit was an Interroll drum motor. We had a spare in the warehouse, same part number, so we swapped it in and got the sorter running. Six weeks later, the new motor started drawing the wrong current and the sort control started throwing errors. Then the next one. Over ten months, we burned through four replacement motors on that one section before anyone asked why.

I put $18,000 in purchase orders on that exercise. Not a giant number in the grand scheme, but it was the same part number every time. The same exact Interroll part number. If you're a buyer, you know that feeling: you ordered exactly what the system said, and you still got called because "the machine's down."

Here's where the woolly bear analogy fits. We were using the part number the way folklore uses caterpillar stripes. We assumed the label contained the truth. But the label was only a small piece of the story.

The Real Divide Wasn't Between Suppliers. It Was Between Specs and Reality.

The deeper issue wasn't the Interroll unit. It was the specification on the drawing.

Sometime in 2023, another vendor had retrofitted that sorter. The retrofit changed the control voltage and increased the speed at the divert point. But no one updated the parts list on our side. The drawing still showed the original motor series. So we kept ordering the old drum motor—right part number, wrong application.

The motor was fine. The gear ratio was wrong for the new speed. The voltage was in the acceptable range but not ideal for the new controller's waveform. And the shaft load had changed because the product mix got heavier. Any one of those factors alone might not have killed the roller. Together, they made it fail in six weeks.

One of the earliest clues was a miscommunication between me and the maintenance lead. He asked me to order "the same as before." I heard "reorder the same part number." What he meant was "the same physical configuration we installed after the retrofit." We were using the same words and meaning different things. Discovered that when the invoice matched the old drawing and the part didn't match the machine.

I don't have hard data on how often this happens across the industry. I wish I had tracked our own failures more carefully before 2024. What I can say anecdotally is that when we finally audited replacement parts over a three-month period, 14 out of 60 parts didn't match the actual installed equipment. Some had the wrong shaft diameter. Some had the wrong gear ratio. Some were fine for the old controls but not the new ones.

That audit was the first time I saw the divide clearly. "Same part number" is not the same as "same spec." The gap between what's on the digital drawing and what's physically on the line is a lot wider than most office people want to admit.

What That Divide Costs When You Ignore It

A wrong replacement is never just the price of the part. It's also the phone call that starts with "the line is down again." I do not want that phone call. Nobody in procurement wants it. But it happens when we skip the verification process.

From our purchase history, a spare Interroll drum motor can run $400 to $600 depending on configuration. But the $500 unit that fails after six weeks costs more than another $500. You add:

• Emergency freight to get a rush replacement
• Technician overtime for the second or third install
• Production downtime on the line
• The project manager's time to explain the delay to a customer
• The expense report that finance rejects because the shipping cost tripled the estimate

I stopped counting our true cost after the second failure, but I know one thing: the lost margin on that sorter section was probably three times the annual savings we'd negotiated on parts. You don't save money by buying cheap components. You save money by buying the right component once.

There's also a softer cost. When a supplier's part fails, it doesn't matter whose logo is on the motor. The operations manager remembers "the conveyor company that couldn't fix it." For an integrator, that kind of memory is expensive.

What Helped: Sourcing Locally and Verifying Before Ordering

Let me give credit where it's due. Once we actually reached out for help, the Interroll side was useful. For our Auckland line, we started ordering through Interroll NZ instead of a third-party importer. That shortened lead times by about three weeks and made it easier to get the correct revision information before buying. In the U.S., the Interroll Wilmington team helped confirm a control board compatibility question that would have been a shot in the dark otherwise. Having a real person who can pull up drawings makes a difference.

But the supplier can't fix a bad internal record. The change that mattered most was building a component verification system—our own CVS, no relation to the drugstore.

It's embarrassing to admit how simple it is. Before we order any replacement motor, someone has to answer four questions on a shared sheet:

1. What is physically printed on the existing motor nameplate?
2. What voltage does the actual controller supply?
3. What is the required speed and torque at the application point?
4. Has any nearby component—like a gearbox or control card—been changed since the drawing was made?

That's it. No magic software. We call it CVS because "component verification system" sounds more official than "checklist," and honestly, giving it an acronym made the maintenance team roll their eyes less.

The first time we used it, the list caught a mismatch that would have caused our third replacement of the same motor. The nameplate said 24 VDC, but the controller had been swapped to 48 VDC in the retrofit. We would have ordered the wrong unit again.

What I'd Tell Another Buyer

If you're buying Interroll gear—or any conveyor components, really—stop assuming the part number in your system is current. There is a good chance the old number still exists in the supplier's catalog, but it might be a previous generation with different compatibility. The brand isn't the problem. The process is.

Per FTC guidance on advertising claims (ftc.gov/business-guidance/advertising-marketing), benefits like "energy-efficient" or "long-lasting" have to be substantiated. You should apply the same rule to internal parts requests. If the person asking for a replacement can't confirm the actual installed specs, don't just process the order. Ask for a photo of the nameplate. Ask when the last drive change was. Ask if the product flow has changed.

I also learned to keep a time boundary on all this. What I described was accurate as of Q1 2025. Lead times and product families change, so before you plan around a specific Interroll part, verify current availability and revision status. The part number on an old drawing may still exist, but it may not be the one you should install.

The woolly bear caterpillar's stripes are only a sign of the caterpillar's age and environment, not a forecast. A part number on a dusty drawing is about the same. It tells you what used to be true, not what needs to be true.

Bottom line: we still buy Interroll. We just don't buy by label alone anymore. The fix was boring, but it worked: close the divide between the drawing and the floor, verify every spec, and only then place the order. That's saved us more in the last four months than any vendor discount ever did.