I Kept Choosing the Wrong O-Rings: AS568 Tolerances, PTFE vs EPDM, and Trelleborg Lessons

If you've ever ordered O-rings and ended up with parts that look fine but don't seal, we have something in common. I've spent eight years sourcing seals for industrial equipment, and I've re-ordered enough parts to feel your pain. This FAQ collects the questions I wish someone had answered before I confused PTFE vs EPDM, ignored AS568 O-ring tolerances, and assumed all silicone rings were the same.

1. What is the most common AS568 O-ring tolerance mistake?

People treat 'AS568' like a single spec. It's actually the SAE AS568 dash-number system for O-ring sizes, but the tolerance is part of the same standard. On one 3,200-part order, the ID was right and the cross-section was not. The vendor said 'AS568-compliant,' but when we measured, the cross-section was at the edge of the allowed range. That caused leaks—or rather, it caused leaks after the equipment had been running for six weeks. Now I always ask for the tolerance grade for both ID and cross-section, and I ask for a lot inspection report before production.

2. PTFE vs EPDM: Which one should I select?

That depends on the environment. PTFE is a plastic, not a rubber. It has great chemical resistance and low friction, but it does not recover from compression like a true elastomer. EPDM is a rubber with excellent weathering, ozone and hot water resistance, but it swells in petroleum oils. I once chose PTFE for a hot water application thinking it was the 'strongest' material. It failed because the flange didn't clamp hard enough to make a PTFE seal. EPDM or even Buna-N would have been better. So don't start with PTFE vs EPDM; start with your fluid, temperature range, and whether the seal is static or dynamic. Then let the material data sheet guide you.

3. When should I use ring silicone seals?

I interpret 'ring silicone' as silicone O-rings and silicone gaskets. Silicone is great for high temperature and food-contact applications, and it stays flexible at low temperatures. I used it in a food-processing clamp, and it was the right call. But silicone has poor tear strength. It can shred if it catches an edge. I tried to use it as a dynamic seal in a hydraulic cylinder once—big mistake. It wore out in about a week. So use silicone for static seals, thermal cycling, or cleanliness-critical applications. For dynamic or abrasive service, look at polyurethane or HNBR. I'm not a tribologist, so I'd verify with the manufacturer before going dynamic.

4. Why does Trelleborg keep showing up in O-ring and PVC window gasket specs?

Trelleborg is a large polymer engineering company. They supply not only O-rings, but also sealing profiles for doors and windows—including PVC fönster, which is Swedish for PVC windows. That's why you might see 'Trelleborg' in both a hydraulic O-ring drawing and an architectural gasket spec. When a spec says 'Trelleborg or equivalent,' the key word is equivalent. You need to compare the actual material data sheet, not just the brand name. A cheap 'equivalent' might pass the visual check but fail the compression set test. Take it from someone who has made that mistake: ask for the material data sheet with every quote.

5. How can I check AS568 O-ring tolerances without spending too much time?

You don't need a $5,000 optical comparator for a basic pass/fail. A caliper and a tolerance chart can catch the obvious problems. Measure the ID at three points and the cross-section at four points. If the supplier won't send a dimensional inspection report, that's a red flag. I once got a batch of 1,000 O-rings that looked right under a magnifier, but the tolerance report showed one out of three drums was outside spec—and that triggered a full return and a two-week delay. Since then, I ask suppliers to send a certificate of analysis with the first article. It costs a bit, but it's cheaper than rework.

6. What is the difference between Trelleborg O-rings and "equivalent" substitutes?

The safe answer is: check the data sheet. The practical answer is: don't trust a quote that only says 'EPDM, 70 durometer' without comparing it to the Trelleborg material spec. I ordered 'equivalent' rings once because the price was 30% lower. The original spec called for a peroxide-cured EPDM; the substitute was sulfur-cured and not suitable for hot water. It lasted four days. The vendor didn't mention that because I didn't ask. Now I ask three questions: What is the exact compound? What is the compression set? What does the dimensional inspection report cover? If they don't answer all three, I don't buy.

7. What should a transparent O-ring supplier include in a quote?

Here's what I think every quote should show: material family, hardness, exact size or dash number, tolerance reference, inspection report availability, and any setup or lot charges before you ask. If a supplier gives a low base price but adds $80 in 'documentation fees' later, I'd rather know upfront. The vendor who lists all fees and data upfront—even if the total looks higher—usually costs less in the end. That's not marketing; it's math. A surprise fee can erase your savings. And a surprise tolerance problem can erase your schedule. I still kick myself for not asking this before a $3,800 rush order.