EN
Title: I've Cut Open 500+ Failed Crimps – Here's What Low-Quality Tools Do to Your Connections
By our failure analysis & quality team
I've spent the last eight years cutting open failed crimps. Not in a lab with white coats – on a shop floor, with a hacksaw and a microscope, trying to figure out why a connection that looked fine on Monday failed on Wednesday.
Here's the uncomfortable truth: most of those failures were preventable. And almost all of them started with the tool.
The ratchet is not optional – we learned that the hard way.
We had an operator who preferred the old plier-style crimper because it was "faster." No ratchet, just squeeze and release. His crimps looked fine from the outside. Then we started having intermittent faults on a control panel line – about one in every 50 units would fail power-up testing.
We cut open 20 of his crimps. Fifteen of them had visible gaps between the wire strands and the terminal barrel. No cold weld. Just a friction fit that had worked long enough to leave the bench but not long enough to survive shipping. That operator's "faster" tool cost us a full rework of 47 panels. We replaced every plier-style crimper on the floor with ratcheting units the next day.
The $740 lesson that changed our inspection protocol.
Ponemon Institute's $740 average roadside failure cost stuck with me because I lived it. Not on a highway – on a production line. A customer's equipment shut down at a printing facility. The root cause? A single under-crimped wire on a solenoid valve. The terminal had passed visual inspection but pulled out with less than 50 lbf of force – half the spec requirement.
The customer's downtime cost: $4,200. The rework cost: $1,100. The crimping tool that caused it? A $35 non-ratcheting unit that had been used for three years without a single calibration check. That $35 tool caused $5,300 in direct losses. We now specify ratcheting mechanisms as mandatory on every tool we buy.
Die wear is invisible until it isn't.
We had a die set that had run about 8,000 crimps. Still looked fine to the naked eye. We sent it for routine measurement, and the die cavity was 0.06mm wider than spec. That's less than the thickness of a human hair.
We ran a test: 50 crimps with the worn die, 50 with a new one. Pull test results: the worn die averaged 1,520N. The new die averaged 1,850N. That 0.06mm cost us 18% of pull-out strength – but you couldn't see it without a gauge.
Now we measure die cavity width weekly. Not monthly. Weekly. And we replace dies at 5,000 cycles – not 10,000 – because by 8,000 they're already degrading.
Under-crimping doesn't fail immediately – it fails on the customer's site.
That's the insidious part. A slightly under-crimped terminal might pass a continuity test. It might pass vibration testing on the bench. It'll work fine during factory acceptance.
Then it gets installed in the field. Ambient temperature cycles from 10°C to 40°C. The wire expands and contracts. Vibration from nearby equipment loosens it another fraction. And then one day – usually at the worst possible moment – it opens. No warning, no gradual decline. Just a system that stops working.
We tracked 47 field failures over 18 months. Forty-one of them were under-crimps that had passed visual inspection but failed pull testing. We no longer rely on visual inspection alone. Every high-reliability crimp gets a pull test or a crimp-height measurement before it leaves the shop.
When the wire pulls out, the tool is almost always the root cause.
I've personally pulled wire out of failed terminals on three separate customer sites. Every time, the terminal looked deformed – but not enough. The wire strands were still shiny. No cold weld had formed. The metal hadn't fused; it had just been pushed together.
We traced every single one back to either a worn die, a mis-set tool, or a non-ratcheting crimper. Not one was an operator error in the sense of technique – it was always a tool that couldn't deliver the right force or geometry.
That changed how we buy. We now treat crimping tools as precision instruments, not consumables. They get serial numbers, calibration logs, and replacement schedules – just like torque wrenches.
The training gap that almost shut us down.
We had a new operator who was producing crimps that looked perfect. He was using the right tool, the right die, the right wire. But we caught a batch where every crimp was under the pull spec. Why? He was gripping the handles too far inboard, reducing the mechanical leverage. The tool was fine. The training wasn't.
We now do a practical crimp test as part of operator certification – not just a PowerPoint. Every new operator has to produce five consecutive crimps that pass pull testing before they're allowed to work unsupervised. That simple requirement cut our operator-related defects by 80%.
What we now do differently, every single day:
-
Ratcheting mechanism as a non-negotiable spec. No exceptions.
-
Die cavity measurement every week – we have a gauge on every bench.
-
Die replacement at 5,000 cycles, not 10,000.
-
Pull test on every high-reliability crimp – not random sampling.
-
Operator practical certification, not just classroom training.
-
Tool serial numbers with calibration logs – we track every tool's cycle count.
The bottom line, from someone who's seen too many failures:
A cheap tool isn't cheap. It'll cost you in rework, downtime, and field failures. We've paid that bill more times than I'd like to admit. We don't pay it anymore.