Before You Buy Another Boston Gear Reducer or Motor, Read This 6-Step Checklist

If you typed "boston celtics fan gear" looking for a Tatum jersey: wrong aisle, wrong building, wrong sport. But if you typed "boston gear" and mean the motion control company—the one with 700 series reducers, AC motors, and servo drives—you're in the right place.

Here's who I am, since you should know whose advice this is: I review every component order before it ships to customers. Roughly 200+ unique line items cross my desk every year—worm gears, gearboxes, stepper motors, bearings, linear actuators. I reject about 8% of first-pass submissions. Almost never because the parts are defective. Almost always because the specifications are wrong.

This is the checklist I run through before approving any order. It's for maintenance supervisors replacing a worn reducer, engineers building a new conveyor line, and buyers handed a list of part numbers with no context. Six steps. Each one has caught a real mistake on my watch.

Step 1: Read the catalog as a contract, not a menu

Boston Gear's worm gear catalog—the printed reference and the online PDF—is organized by center distance and ratio. The 700 series is what you'll see most often in standard industrial applications. It's a solid catalog. But it's not a menu.

When I open a catalog page, I check three things, in this order:

• Ratio. A 10:1 and a 20:1 worm gear may share the same frame size. They are not interchangeable.
• Service factor. This is not a safety factor. It's an overload allowance based on the application class—uniform, moderate shock, or heavy shock.
• Input speed range. A reducer rated at 1,750 RPM input may or may not be rated for 3,600 RPM. The table shows it, if you look.

The part number tells you what's included, but it doesn't tell you what's excluded. That distinction has cost us about $3,000 per mistake, in my experience. Maybe $2,800, I'd have to check the RMA log.

Step 2: Check the duty cycle before you check the torque rating

The classic mistake is sizing a speed reducer by torque alone. That single catalog number looks so authoritative. It's also only half the story.

People think a higher torque rating means a longer-lasting reducer. The reality: a reducer's rated torque assumes a duty cycle. Run a 500 lb-in rated worm gear at 500 lb-in with no downtime and you'll bake the lubricant. The torque rating isn't wrong. The context is missing.

The causation runs the other way. A reducer that outlasts its application earns its torque rating. The spec follows the duty cycle, not the other way around.

Our plant runs three shifts, roughly six days a week—a continuous duty curve by anyone's definition. One shift would have much more thermal headroom. Same part number, same torque, completely different lifespan. We installed a batch of 700 series reducers that ran hot for a full quarter because we sized for our standard duty cycle and then increased production. Our fault, not the equipment's.

Step 3: Match the motor frame to the reducer input—then verify it again

This one sounds too obvious to be on a checklist. It isn't.

I still kick myself for a NEMA frame mismatch back in 2022. We ordered AC motors with a 56C frame for a reducer that required a 143T input. Shaft diameters were different. Mounting bolts didn't line up. Twenty-four units, roughly seven thousand dollars, on pallets waiting for a reorder.

Why? The part number on the quote looked right. The sales order looked right. Nobody walked to the rack and compared the reducer input face to the motor face. That's my job, and I missed it.

Per NEMA MG 1 (updated 2021), frame designations are standardized—56C, 143T, 182T, 213T. The number encodes shaft height, shaft diameter, and mounting dimensions. There's no excuse for guessing. Look it up before you approve, not after the truck arrives.

We caught that one at the dock instead of the production floor. Lucky. An autocomplete mixup on an $18,000 project almost slipped through last year—one character different, completely different mounting. Almost.

Step 4: What's a stepper motor? (And when you actually need a DC servo motor)

"What's a stepper motor?" I get this question more often than you'd think, and not from hobbyists—from buyers with purchase orders in hand. Short answer:

A stepper motor rotates in discrete steps. Send one electrical pulse and it turns a fixed angle, typically 1.8 degrees per step for a standard 200-step motor. No encoder, no feedback loop. That's why steppers are inexpensive, repeatable, and popular in 3D printers, CNC machines, and positioning tables.

A DC servo motor, by contrast, runs on a closed loop. An encoder reads the actual position; the drive corrects errors in real time. More torque at higher speeds, but you pay for the feedback system. Think of it this way: a servo is like a contractor who checks every measurement with a laser level. A stepper trusts the print and only finds out later if the door doesn't close.

Which one do you need? If your axis moves slowly and end-of-travel accuracy matters more than dynamic response, a stepper holds its own. If you need high torque at high speed, go servo. But when you spec the servo, don't stop at torque—check encoder resolution and feedback type, incremental vs absolute. Wrong feedback protocol means the drive and motor don't talk. Non-trivial change order.

Step 5: Write the part number down. Then read it back.

This one feels like it belongs in a training slide for interns. That's exactly why everyone skips it.

In the past year alone, I've caught three part number errors on purchase orders:

• A 60:1 ratio entered as 6:1
• A left-hand output entered as right-hand
• A catalog number missing its leading digit entirely

Not one of those errors would have surfaced if somebody didn't read the number aloud, character by character, against the approved quote. No software catches that. It's a habit. Not glamorous. But it costs nothing.

Also, don't trust autocomplete. Modern purchasing systems fill in product descriptions for you. Convenient. Confidently wrong about 2% of the time. Our own system almost substituted a different Boston Gear model—one character apart and completely incompatible. On an $18,000 project. Almost.

Step 6: Ask "what's NOT included" before you ask "what's the price"

My favorite part of the checklist, because it's the most overlooked. When a quote arrives for a gearbox or a motor, the line item price almost never covers everything the application needs.

I've learned to ask "what's not included" first. The vendor who itemizes every fee upfront—even if the total looks higher—usually costs less in the end. The vendor who quotes a bare-bones number and adds extras later isn't a bargain. They're a headache.

What's usually missing? Shipping. Mounting hardware. Keyway cutting. Oil fill. Couplings. Sometimes the bolts for the motor face. None of these are conspiracy-level hidden fees; they're standard assumptions in the industry. But an assumption is exactly where a budget goes to die.

In our Q1 2024 audit, we compared total landed cost across four vendors for identical reducer specs. Sticker prices varied by about 12%. Final costs varied by over 40% once freight and add-ons entered the picture. The lowest sticker price came in second-most-expensive overall. Good data. Expensive lesson.

Three installation mistakes that still catch people

Say you get the order right. These are the next set of traps, after the parts land on the floor.

• Ignoring ambient temperature derating. A worm gear's rating is typically based on a 25°C ambient. At 40°C—common in plants where the gearbox sits near ovens with no wall in between—the catalog rating shrinks. Boston Gear's worm gear catalog includes derating factors in the engineering data section. Check them before you size, not after.
• Misreading backlash specs. Standard reducers have a published backlash range. If your application can't tolerate a few arc-minutes of play, you don't need a lucky sample—you need a different part number with special backlash. It's not a quality lottery.
• Trusting the ratio on the label. A "10:1" worm gear may actually be 10.416:1. Depending on input RPM, that's a measurable error at the output. Manually rotate the input and count the output revolutions before you set machine parameters.

One last thing

This checklist has worked well for us, but our operation is a mid-size B2B facility with predictable ordering patterns. If you're in a seasonal business with demand spikes, or you're an OEM building custom machinery, your mileage may vary. I can only speak to my own context.

If you take one thing from this: slow down. Spec first. Purchase second.

Parts don't fail as often as assumptions do.