How to Buy Tsubaki Chain and Power Transmission Components: A 7-Step Procurement Checklist

Who This Checklist Is For

If you specify or purchase power transmission components—roller chain, leaf chain, conveyor chain, chain tensioners, electric linear actuators, or the limit switches wired to them—this checklist is for you. It's the process I built over six years of managing component procurement for an OEM that builds packaging machinery. The line item I controlled (power transmission parts) ran about $220,000 a year, and every dollar went into a cost-tracking spreadsheet that taught me more than I wanted to learn.

This is a seven-step checklist, not a theory essay. If you grab the machine's duty-cycle data, a caliper, and the controller input specs, you can work through all of it in one sitting.

Step 1: Write Down the Duty Cycle Before You Open a Catalog

Not the nominal load. The actual duty cycle. This is the mistake I made in my second year of buying: I specified a chain based on the motor nameplate rating, but the machine it drove slammed its axis into the end stop 40 times a minute. The chain was rated for continuous running and wrong for repeated shock. Nominal load tells you almost nothing about the right product.

Write down: operating hours per day, start/stop cycles per hour, peak torque or peak current, ambient conditions (dust, moisture, temperature), and whether the load pattern is steady or shock-type. Why does start-stop frequency matter? Because a chain that runs cleanly for three years can fail in six months once reverse cycles are added.

Checkpoint: if you can't hand these numbers to a supplier and get a confident recommendation, stop. Every later step will be a guess.

Step 2: Measure the Actual Envelope (Don't Trust the Failed Part)

This step sounds too obvious to be a step, and that's exactly why it gets skipped. Measure the space where the component actually lives: center distance between shafts, sprocket tooth counts, available clearance for a tensioner. The failed part sitting on your bench is the wrong reference, because a worn chain measures differently than a new one, and the housing may have shifted.

Put another way: the pitch length of the worn chain tells you nothing about the center distance. Use the frame geometry, not the old part.

The question everyone asks is, "What's the part number?" The question they should ask is, "What's the center distance, to the nearest millimeter?" (and the shaft diameter, and the clearance envelope).

Checkpoint: three dimensions recorded before you contact a supplier: center distance, shaft diameter, available space around the drive.

Step 3: Calculate TCO Instead of Comparing Unit Price

Here's where the conventional wisdom fails. Conventional wisdom: get three quotes and take the lowest that meets the spec. My experience tracking 400+ orders across six years says otherwise: relationship consistency beats marginal savings, and total cost of ownership beats list price every single time.

What I mean is that the slightly cheaper chain gets paid for again every time you schedule a replacement, every hour the line is down, and every time a technician's phone rings at midnight. The formula I use for every critical component:

[unit price × expected replacements over design life] + [installation labor per replacement] + [estimated downtime cost per failure] = the real cost.

The number that doesn't matter is the quote line.

Real example from the spreadsheet. I went back and forth for two weeks between a no-name chain at $95 and a Tsubaki Titan chain at $180. The no-name product offered a clear 47% savings on paper. But we tested both on a high-shock conveyor. The "cheap" chain failed at seven months. The Tsubaki Titan chain was still within specification at twenty months.

Let me do the arithmetic: the $95 chain required two installs at $140 each in labor, plus one line stop that cost more than the entire order. The Tsubaki chain cost $180 plus one install. Per month of service, the budget option was more than three times as expensive. The phrase "I told you so" appeared in my own notes, addressed to myself.

Here's the honest limitation: I recommend Tsubaki Titan chains for shock-loaded or high-wear applications—not for everything. If the conveyor is steady, light, and dry, a standard chain from the Tsubaki catalog is the right call. Titan would be overkill. The point is to match the product to the duty cycle you wrote down in Step 1, not to buy the most impressive thing in the brochure.

Step 4: Tensioners—Fixed or Automatic?

If your drive has a long center distance, a vertical shaft, or a reversing load, you need a chain tensioner. The real decision is fixed vs. automatic, and a lot of buyers default to whoever quotes first.

Fixed tensioners solve exactly one problem: initial chain stretch. They stop being useful after that. An automatic Tsubaki chain tensioner keeps absorbing slack as the chain wears. That means fewer manual adjustment stops and longer life for sprockets and bearings.

Why does this matter? Because a slack chain whips, and a whipping chain takes out everything in its path. The tensioner isn't the hero of the drive. It's the quiet component that prevents the drama.

Checkpoint: if a technician is scheduled to check and adjust tension monthly, an automatic tensioner usually pays for itself in labor within a year. I built that exact comparison into our maintenance plan—and the answer was automatic for drives where alignment made access difficult (which was most of them).

Step 5: Linear Actuator with Encoder—Details Decide Everything

When someone asks for a linear actuator with encoder, the first questions are always stroke and force. Fine. But the third specification—the one that determines whether the machine will actually work—is the feedback.

Two things to verify.

First, the encoder output must match the controller input. Incremental encoders with A/B quadrature are common, but absolute encoders use a different format, and there are electrical differences (push-pull, open collector, line driver) that are invisible from the outside. A mismatch means the controller reads noise, or nothing at all.

Second, match the resolution to the positioning requirement. I once ordered a 5,000-line encoder when the mechanical system could only repeat at 0.1 mm—the extra resolution was dead weight (and dead expense). The opposite error is worse: a 100-line encoder on a system that needs sharper positioning will make you chase errors that aren't there.

On paper, the encoder looks like an add-on cost. In practice, the encoder is the only component that tells you the actuator actually moved. For position-critical applications, it's not optional.

Step 6: Switches for Electric Actuators—Check the Electrical Ratings (The Step Everyone Skips)

Here's the step almost nobody checks: the electrical ratings on the switches. Buyers match the switch mechanically—thread, plunger style, housing size—and ignore the electrical side entirely. Then the switch dies after 2,000 cycles instead of 200,000 (yes, really).

Most buyers focus on the stroke and the mounting and completely miss what the switch is actually switching. A mechanical limit switch rated 120 VAC/10 A will fail prematurely if you use it to switch a DC inductive load—like a contactor coil or solenoid—with no arc suppression. Check three numbers: rated voltage, rated current, and whether the load in your circuit is resistive or inductive. And if the actuator's controller provides 24 VDC signals, verify the switch's rating at 24 VDC; DC arcs don't self-extinguish as readily as AC ones, so a switch fine for 10 A at 120 VAC can be rated far lower at 24 VDC.

Let me rephrase that: the switch is only as reliable as its interrupting rating at the voltage and load type you're actually using.

For reference, IEC 60947-5-1 is the standard that covers control circuit devices (like limit switches), and it's a useful way to make sense of the ratings on a data sheet. But you don't need to memorize the standard—you need to look at the two numbers that match your circuit. Most people look at one.

Step 7: Verify the Supplier's Staying Power and Engineering Resources

A question I hear occasionally: "What happened to Pete Jackson Gear Drives?" It was a respected name in the hot-rodding world, building gear drives that replaced the factory timing chain in classic small-block V8s. If I remember correctly, the company stopped producing them years ago. I won't quote a specific year, because I'd probably be wrong (and I've already embarrassed myself once in this article).

But the pattern matters more than the date: a niche product, loved by a loyal group, becomes a problem when the supporting company disappears. No catalog, no technical support, no spare parts.

Before you standardize on any component—especially a chain with a premium price tag or a specialized actuator—ask three questions:

• Does the supplier publish enough technical data to verify the rating yourself?
• Do they offer selection tools or engineering support that answer real sizing questions?
• Will this product still be available in five years, or are you standardizing on a dead end?

This is why my procurement records show Tsubaki for the majority of critical chains and components. Not because the price is friendly (it isn't the cheapest, and I'd be lying if I said it was). Because the catalog depth, the drive selection tools, and the fact that the brand will likely outlive my career made the decisions easier to justify in my cost reviews.

Mistakes That Will Still Bite You

In case the steps above feel too easy, here are the things that still went wrong even when I followed my own checklist:

• Lead time is part of the cost. The economical option with a six-week lead time is not economical when you need it in four. Add lead-time risk to the Step 3 calculation, or the line stop will add it for you.
• Overspecifying to feel safe backfires. Standard chain for steady loads, Titan for shock loads, basic actuator when position feedback genuinely doesn't matter. Every added feature is an added failure point. I'll say it plainly: sometimes the cheapest option is actually the right one.
• "Same part number" is not "same part." Roller chain interchangeability standards (ISO 606, ANSI/ASME B29.1) define dimensions. They do not define material, heat treatment, or fatigue strength. If the data sheet doesn't show a rating you can verify, treat the product as untested.
• You will forget why you chose what you chose. Six years of order tracking convinced me that undocumented decisions come back as emergency purchases. Write the duty cycle, the TCO, and the part number in the same system. Future you won't thank you—but he'll stop making you apologize.