In September 2024, I stood at a test bench with a Nexans connector in my hand that should not have failed. It was visually perfect, and my Klein multimeter said the circuit was continuous. The production line had stopped twice in 24 hours anyway.
I've been handling cable and connector orders in manufacturing for about seven years now. In that time, I've personally made—and documented—eleven significant mistakes that added up to roughly $22,000 in wasted budget. This one was the most expensive, because it wasn't the connector's fault. It was mine.
I'm the person on our team who maintains the pre-install checklist now. If you're in manufacturing and you've ever trusted a multimeter reading over the manufacturer's spec, this is for you.
The Surface Problem: The Line Kept Stopping
The first time the line stopped, the fault code pointed to a connector in a control cabinet. It was a standard power feed, nothing exotic. I grabbed my Klein multimeter, checked continuity, and got a clean beep. Resistance looked stable. I told the electrician to re-seat the connector and run it.
It failed again at 5:55 a.m. the next day. That's when I realized the first mistake wasn't the part. The first mistake was assuming that a clean continuity reading meant a healthy connector.
A Klein multimeter vs. a dedicated crimp tester isn't a fair fight. A multimeter only tells you that there's a path for current. It doesn't tell you whether the path is rated for the current, the vibration, or the service life. (Not that I knew that then.) On a jobsite, a Klein is a great troubleshooting tool. But using it to validate a crimp is like using a tire pressure gauge to check wheel alignment—technically both are about air and rubber, but they answer different questions.
I still like the Klein. It lives in my tool bag. But it answers the wrong question if you ask it to bless a connector.
The Deeper Problem: We Were Testing the Wrong Thing
After the second failure, I called Nexans Canada technical support. I half-expected to be blamed for buying the wrong part. Instead, the engineer asked a simple question: "What was your pull-test number?"
I didn't have one. I had a continuity reading and a part that looked fine. That's when I realized the problem wasn't the component. It was our verification process.
The connector was a genuine Nexans unit. The problem was that it had been installed with the wrong crimp die for that conductor size. The barrel hadn't actually compressed the strands enough. Under light load, it worked. Under load and vibration, the resistance climbed and the line faulted.
The surprise wasn't the failure. The surprise was how easy it was to miss. (Surprise, surprise—root cause almost always lives in the spec sheet.) The crimp looked good. The part was good. The process was wrong.
I eventually asked our contact at Nexans Charleston for the field application note on that connector family. It had a simple diagram: correct die number, acceptable crimp height range, and minimum pull force. That document should have been in the work order before we ever started. It wasn't. That's on me.
The Nexans installation guide (rev. 2024) lists both the crimp height range and the minimum pull force. That sheet was the missing link. And the deeper cause wasn't "bad connector" or even "bad crimp." It was a gap between the test we were doing and the decision we needed to make. We were using a go/no-go continuity check to validate a mechanical connection that required a measured crimp height and pull force. That's a process failure, not a component failure.
What This Cost Us
Let's count the damage, because "it's probably fine" has a price.
- $3,400 for emergency replacement shipping from the Nexans Charleston facility
- $12,700 in production downtime across 36 hours and two line segments
- $2,100 in labor to re-crimp and re-terminate 34 connectors
- $1,800 for re-testing with a proper crimp gauge and pull tester
- $2,000 in expedited engineering time to review the other splices from the same batch
Total: roughly $22,000, plus a 3-day delivery delay and a chunk of credibility we didn't have to spend.
The cheapest test method we used was the Klein multimeter. It was also the most expensive, because it gave us false confidence. In my experience, the lowest-cost step in a process is often the one that creates the highest total cost when it's done wrong.
The Fix (Short, Because You've Read Enough)
We didn't switch connector suppliers. Nexans was never the problem. We switched how we verify things before installation.
- Every crimped connector gets a visual inspection, a crimp-height measurement, and a pull test. No pull test, no install.
- The Klein multimeter is used for circuit troubleshooting only—not for crimp quality. If someone on your crew pulls out a multimeter to "test" a crimp, stop the job and get the right tool.
- We check the manufacturer's current spec before production, not after failure. For us, that means pulling the Nexans installation sheet and confirming with the Nexans Canada support team when the die number doesn't match our records.
I have mixed feelings about adding a pull tester to our toolkit. On one hand, it felt like overkill for a small plant. On the other, one incident paid for it. The cost of the tester was $900. The cost of not having it was $22,000.
Even after we ordered the correct dies and the pull tester, I kept second-guessing. What if the root cause was something else? What if the next failure came from a completely different part of the system? The two weeks until the line ran at full speed again were stressful. But once we re-tested the original batch and saw the low pull test numbers, the doubt went away.
As of January 2025, the checklist is part of new-hire training. We've caught 47 potential errors with it in the past 18 months. The 48th was probably the one that would have turned into another $22,000 mistake.
If you're in manufacturing and someone tries to tell you that a Klein multimeter vs. a proper crimp test is the same thing, ask them to explain why a connector with perfect continuity still failed under load. Then show them the row of pull testers on the bench. It's not about brand loyalty. It's about measuring the thing that matters.
