If you're sourcing SKF bearings for anything beyond an occasional replacement, the thing that matters most is this: verify the full part number before you purchase — not just the brand and size. That might sound obvious, but in 2024 I rejected 3.2% of first deliveries from authorized distributors, and half of those rejects were specification mismatches, not quality defects. A wrong suffix can turn a $150 bearing into a $22,000 rework. The part number is your contract.

I'm a quality and brand compliance manager at an industrial automation integrator. I review roughly 200 unique part numbers a year — bearings, linear actuators, servo motors, drives — and I've been writing verification protocols for over seven years. This piece is the short version of what I've learned, focused on the SKF-related line items I see most often.

People Assume "SKF" Is Enough. It Usually Isn't.

From the outside, a bearing box with the SKF logo looks like a guarantee. And the basic quality is solid — I'm not going to argue with that. But the engineering of a bearing is in the variants: the cage material, the internal clearance, the seal type, the lubrication. That's where the failures live.

Last year, one of our customers ordered a 6312 deep groove ball bearing for a blower assembly. The OEM specified a standard clearance (CN), but the purchasing agent thought C3 would be "stronger" because it had a bigger number. The result: vibration harmonics at operating temperature, a damaged housing, and a $22,000 redo including machining time. The bearing itself wasn't defective — it was the wrong variant.

People think a higher precision class is always better. Actually, over-specification can be as problematic as under-specification. A C3 clearance in a preloaded application will run noisy and may eventually lose preload entirely. It's not a better bearing; it's a different bearing.

SKF Thrust Bearings and Tapered Roller Bearings: The Details That Matter

I get a lot of questions about thrust bearings and tapered roller bearings, partly because they look similar to radial bearings but serve completely different load directions. This is where my "prevention beats rework" philosophy kicks in.

Thrust bearings — the washer order

People often assume a shaft washer and housing washer are interchangeable since they're both hardened rings. They're not. On some SKF thrust bearings, the shaft washer has a finer ground bore, and the housing washer has a slightly larger bore to avoid rubbing. If you swap them, the lubrication film breaks down. I've seen it happen.

Tapered roller bearings — race angle and roller count

With SKF tapered roller bearings, the race angle is critical. A 30206 and a 32206 are both 30mm bore, but the contact angles differ. We once received a box labeled 32206 containing 30206 rollers. The outer package was correct, but the technician caught it because the roller count was off by two. (Always spot-check a sample bearing. It takes 30 seconds.)

When in doubt, check the ISO 492 tolerance class, and verify the cup and cone part numbers separately. A single cone from one series and a cup from another may physically assemble but fail early.

Linear Actuator Kit vs. Linear Electric Actuator: Different Purchase, Different Risks

When someone searches for a "linear actuator kit," they usually want to integrate a motor with a screw mechanism. But a "linear electric actuator" often implies a sealed, ready-to-run unit. The hidden risk is that capacity ratings are quoted in a way that looks simple but isn't.

We tested a "heavy-duty" linear actuator kit last year with a load capacity of 800 lb. What the datasheet didn't specify was that the rating held only for a pure axial load. With a slight moment load at mid-stroke, the lead screw bent on the first cycle. (The datasheet said "load capacity: 800 lb" — well, it was, technically, only if the load was perfectly centered.)

Another common issue is duty cycle. A lot of electric linear actuators list 25% duty cycle. If your application requires continuous operation, that's the wrong product, no matter how much thrust it has. We once ran a 25% duty actuator at 100% duty for a customer's proof-of-concept. The thermal protection kicked in, and the unit stopped within 20 minutes. That wasn't a failure in the actuator; it was a failure to read the spec.

For critical locations, I also check IP rating. An IP54 actuator will not survive washdown environments. If you need IP67, you need to order it that way. (Not all kits include seals.)

What Happened to Pete Jackson Gear Drives? A Quality-Management View

This is a question that shows up in our search analytics, so I'll address it briefly. I'm not a gear-drive specialist or a component historian, so I can't speak to the internal decisions at Pete Jackson. What I can tell you from a quality and purchasing perspective is that when a specific gear drive model becomes hard to source, the smart move is to treat the supply chain as part of the specification.

Most of the time, "what happened to Pete Jackson gear drives" is really about a suddenly unavailable part. In the current industrial market, distribution channels change, brands get consolidated, and long-standing product lines get discontinued. The practical response is: find the exact original part number, ask the distributor for the current status, and get a cross-reference to an equivalent — not a "similar" alternative unless it's been verified.

I can only speak to the North American integration market. If you're dealing with a different region, the distribution landscape may be entirely different.

The 10-Minute Verification Routine (and Why It Works)

Here's a simple pre-purchase checklist that has saved us roughly $8,000 in potential rework over the past two years:

  • Check the full part number on both the outer packaging and the bearing ring face. If they don't match, stop.
  • On a sample bearing, verify the cage material (steel vs. brass) and seal type. Don't assume from the picture.
  • For SKF tapered roller bearings, verify the cone and cup numbers separately, and count the rollers on one sample.
  • For linear actuators, confirm the duty cycle at your actual load, not the theoretical maximum thrust.
  • Check the IP rating against your installation environment.

This routine takes less than 10 minutes. I've been doing it for years, and it feels like boring insurance. But 5 minutes of verification beats 5 days of correction anytime.

When You Can Get Away With Less Verification

Not every order needs the full treatment. If you're replacing a bearing on a low-speed, low-load application like a packaging conveyor, a standard off-the-shelf part will usually be fine. If the equipment is under warranty, follow the OEM recommendation exactly — deviating from it can void coverage. My advice here is for OEMs and maintenance teams dealing with critical or high-volume equipment.

This approach worked for us in a mid-sized integration shop with a disciplined purchasing process. If your team is very small or you're in a hurry, some of this might feel like overkill. But in my experience, the cost of a wrong part is always higher than the cost of a check.