Here’s something I’ve learned the hard way: ordering a gearbox is not like ordering a box of bolts. For the past five years, I’ve been the office administrator who handles purchasing for a 180-person manufacturing plant. That means I manage roughly $350,000 in MRO and transmission component orders across 12 vendors. I report to operations on Mondays and finance on Fridays—and neither one lets me forget the time I ordered the wrong Winsmith gearbox part.

Let me back up. The conventional wisdom in purchasing is that if you have the part number, you’re safe. My experience with 60–80 orders a year suggests otherwise.

The part number is a starting point, not a conclusion.

When I took over purchasing in 2020, I thought the hardest part of this job was getting vendors to answer emails. I was wrong. The hardest part is making sure the component you ordered is actually compatible with the system it’s going into. That lesson came with a repair bill.

The surface problem: “I just need a Winsmith gearbox part”

Searching for a “winsmith gearbox supplier” doesn’t feel dangerous. You get a list of distributors, some with photos, some with chat widgets. You click, type the part number, and expect a brown box at the loading dock.

The first time I ordered replacement parts for a Winsmith gearbox, I was trying to fix a worn input shaft on a small conveyor drive. I assumed—there’s that word—that “Winsmith 920” was all I needed. I didn’t verify. Turned out the unit had been modified years earlier with a different input flange and a C-face adapter for a motor that had since been replaced. The part number in the old maintenance file was for a gearbox that had been retired in 2008.

Whether you need a 920, 917, or 926, the model only gets you to the family. It doesn’t tell you the ratio, the mounting position, the oil fill orientation, or whether the output shaft is solid, hollow, or keyed. A specialist can help, but only if you send enough information. I learned to ask for a nameplate photo before assuming the model number means what I think it means.

The deeper cause: compatibility is a system, not a component

The mistake that cost us actual money wasn’t a gearbox purchase. It was a motor.

Somebody in production asked me, “What motors are compatible with VFD?” The easy answer is that most induction motors can be connected to a variable frequency drive. But “can be connected” is not the same as “designed for it.” I learned the difference when a motor we bought from a general supplier failed after three weeks on a VFD.

Here’s the part that doesn’t make sense until you’ve seen it: a VFD doesn’t deliver a clean sine wave. It sends high-voltage pulses at a carrier frequency. If the motor’s insulation isn’t rated for that, the voltage spikes can stress the winding until it fails. Lead length and grounding also matter. According to the NEMA MG 1 standard (nema.org), Part 31 covers inverter-fed motor requirements. I checked that after we lost a shift.

So when someone asks “what motors are compatible with VFD?” the real answer is: it depends on the motor’s design, the drive’s configuration, the cable length, and the load profile. A motor marked “inverter duty” is a safer starting point. A motor that’s just an AC motor is a gamble.

Servo motor encoders and servo motor controllers are a different animal

Maybe the most frustrating part of my job is when someone asks me to source a “servo motor encoder” or “servo motor controllers” as if they’re interchangeable. They are not.

A servo motor encoder is not just a sensor that counts rotations. In modern servo systems, the encoder is part of a feedback loop with the controller. It tells the drive where the rotor is, how fast it’s going, and sometimes the absolute position in a multi-turn system. The servo motor controllers send power based on that feedback. If the encoder’s output format or resolution doesn’t match what the controller expects, the axis can vibrate, fault, or slam to a hard stop.

I don’t fully understand encoder protocols. I’ve learned to say so. When our maintenance manager asked me to buy a replacement encoder, I searched online, found a similar-looking part, and created the PO. It looked right. Actually, it looked close. The connector was right. The price was $140 less than the OEM part. It wasn’t right. The controller didn’t recognize the communication frame, and the axis homed to the wrong position twice. The fix cost us a weekend and a service call from the integrator.

The OEM part cost something like $600—don’t quote me on that; our purchasing system has the exact number. The point is that saving $140 turned into a four-figure lesson.

The cost of getting it wrong is not the price of the part

In March 2023, we had a line down for seven hours waiting for a motor that matched the drive and the gearbox. The downtime cost was roughly $1,800 an hour—that number came from our operations manager’s spreadsheet, not a sales brochure. The motor itself was $900. The freight to expedite it was $320. The real cost was $12,600 in missed production, plus overtime for the night crew.

This may sound like an engineering problem, but it’s also a purchasing problem. The vendor who couldn’t provide proper invoicing cost us $2,400 in rejected expenses. That unreliable supplier made me look bad to my VP when materials arrived late. I’ve learned that trust is a feature.

I had another situation where I had two hours to decide on a replacement Winsmith gearbox because the OEM lead time was eight weeks and we had a customer order due. Normally I’d get three quotes and check references. There was no time. I went with the only winsmith gearbox supplier who asked me to send a photo of the nameplate. That photo—along with the shaft dimensions we measured—caught a mismatch that would have turned a “simple” replacement into a $3,000 machining job.

That experience changed how I think about “winsmith gearbox parts.” The parts themselves aren’t the risk. The missing information is the risk.

What I do differently now

First, I stopped sending just a part number. I send a nameplate photo, the motor frame, the ratio from the nameplate, the direction of rotation, and the type of load if I know it. If I don’t know it, I say so. The words “I don’t know” have saved me more money than any discount code.

Second, I stopped asking “What motors are compatible with VFD?” and started asking “Is this motor listed for inverter duty per NEMA MG 1?” That one question filters out products that look okay but aren’t built for continuous inverter service.

Third, I work with a specialist for servo motor encoders and servo motor controllers. Not a general supplier. A controls specialist who can tell me which encoders match which drive. I’d rather work with a specialist who knows their limits than a generalist who overpromises. The vendor who said “this isn’t our strength—here’s who does it better” earned my trust for everything else.

And when I order winsmith gearbox parts, I use a supplier who verifies the data before confirming the price. That extra phone call is worth more than a fast quote.

I’m not an engineer. I just sign the POs. But I’ve learned that the buyers who cause the least downtime aren’t the ones with the most technical knowledge. They’re the ones who know when to ask for help. That’s true whether you’re buying a gearbox, a VFD motor, or a servo encoder.