Spec support for HVAC, pump and fan motors EN / DE / ES / ZH - North America - Europe - Asia Pacific

Application Note

Your Ringfeder Torque Spec Was Right. The Coupling Still Failed.

If you've ever read a Ringfeder torque specs sheet, you'd be forgiven for thinking coupling selection is a lookup exercise. The column of numbers looks like the answer. It's usually the one thing that gets you pointed in the wrong direction.

The call that solved the wrong problem

I handle rush orders at a power transmission supplier. Six years, north of 300 expedited coupling jobs — including same-day turnarounds for OEMs and maintenance teams running against shutdown clocks.

In March 2024, a client's maintenance engineer called at 2 p.m. on a Tuesday, 48 hours out from a scheduled shutdown window. The coupling on their main drive had failed, production was down, and the replacement had to arrive before Thursday morning. “Torque is the same,” he said. “3,500 Nm continuous. Matches the spec sheet.”

I pulled the Ringfeder torque specs chart. His ac induction motor's nameplate checked out too. On paper, the selection was correct. We arranged a rush shipment. The problem was never the torque spec — but nobody had asked whether the torque spec was the right question.

Three weeks later, the same client lost a second coupling on the same shaft. It failed at less than 20% load. The spec sheet had been right both times.

Why a correct spec sheet can send you in the wrong direction

Coupling torque ratings are measured under controlled conditions — bench-tested, constant speed, ideal alignment, no external axial load. That number is precise. It's also, on a factory floor, the least interesting thing about what your system is actually delivering to the coupling.

When I walked the client's timeline backward, I found three overlapping load paths no torque chart accounts for:

Misalignment. The motor's base shims had been off by 0.3mm. Nobody had flagged it, because 0.3mm isn't a problem. Except over six months, that offset pushed the bending moment on the shaft to roughly twice the continuous torque. The coupling tolerated the torque fine — it couldn't tolerate the moment it was carrying on top of it.

Axial load and the thrust bearing. This was the sneaky one. Thermal growth in the ac induction motor was pushing axial load into the bearing track that thrust bearings are supposed to absorb. The thrust bearing was near end-of-life — invisible when the machine was running. So the coupling ended up carrying axial load it was never designed for. Torque specs don't mention thrust bearings because coupling suppliers reasonably assume the bearing side is healthy. Every link in the chain assumed someone else had the thrust bearing covered.

Starting torque peaks. Direct-on-line ac induction motors push peak torque at somewhere between 1.5× and 2× the steady-state rating during acceleration. Every single start. If the coupling was selected on continuous torque — which is what the spec sheet leads with, and what most buyers compare — it was running 40–80% over its rating hundreds of times a shift. That's not a spec error. That's a spec mismatch that only shows up under load.

None of these are exotic. They're ordinary system behavior. But a torque chart can't flag them, because a torque chart starts from the assumption that everything else is healthy.

The part nobody said out loud

There was also a communication layer I missed entirely on the first call. The client said “same torque.” The supplier said “same torque.” What got compared was steady-state continuous torque. What was killing the coupling was start-stop peak torque, sustained over a three-shift schedule.

Same words, different numbers, off by a factor of 2.5.

And there was one more piece in the same machine — the linear rail on the frame. Someone had swapped the carriage bearing from an 8×15×24 unit to a longer lm8luu size (8×15×45) because the original configuration had been measured short. The machined housing had been reworked to accept it. The new bearing fit, but the stiffness profile was different, and a high-frequency vibration crept into the frame. That vibration traveled through the frame into the coupling housing. Nobody — including me — connected it to the coupling failure, because a linear bearing size and a shaft coupling are in completely different conversations.

They aren't, though. They're just distant nodes in the same system diagram. The coupling was doing its job inside a system that had stopped agreeing with itself.

What the calls actually cost

Across the first coupling, the second coupling, and the third repair visit, the combined total on rush freight, lost production hours, and extra labor came out to roughly $4,200. The coupling itself — a legitimate Ringfeder power transmission replacement — was around $1,100. The part was never the expensive part.

The second failure was the phone call that stung. The client's procurement lead asked, very politely, whether we thought they should try a different supplier. I knew we weren't the problem. We'd sold them the right part. Neither of us had been solving the right problem.

The third failure was the one that mattered. A 22-hour unplanned outage cascaded to a downstream client's line. The trust hit was worse than the cost — they later ran their own system-level vibration study and concluded the coupling would have failed again no matter how many times it was replaced.

I don't have a clean resolution to that story. That's the part I still sit with. We kept answering the right question while solving the wrong one, and the wrong one kept winning.

What to do before your next swap (it's shorter than it sounds)

If you're about to replace a coupling, the checklist is genuinely brief:

  • Pull the motor's starting torque peak — not just nameplate continuous torque.
  • Ask about the thrust bearing's condition — couplings tolerate axial load only within a narrow band. Call your motor service shop.
  • Run a laser alignment, not a feeler-gauge “good enough” check.
  • Look at vibration signatures around the coupling housing — linear bearings, rails, and mounting surfaces included.
  • Confirm everyone is comparing the same number — continuous vs. peak torque, written down and read back.

It sounds like extra work. It is significantly cheaper than a repeat replacement.

The best kind of rush order is the one that doesn't come back. For me, that's what being good at this job actually looks like — not the speed of the first shipment, but the fact that three weeks later, nobody's calling.

Spec desk note

Our application notes are written for contractors, distributors and maintenance teams comparing motors by duty, not just catalog family.