Your Breaker Is Fine. Your Coordination Isn't.

The Trip That Started This

In March 2024, a 400A molded case breaker on our injection molding line tripped three times in one week. Maintenance reset it, the line came back up, and forty minutes later it tripped again. Reset. Trip.

The maintenance supervisor's read was simple: the breaker's going bad. Which is a reasonable thing to say. Breakers do wear out. We ordered a replacement, it landed in two days, we installed it. Three weeks of nothing. Then April came around and it started tripping again — this time taking the whole feeder with it.

Quick context: I'm the office administrator for a 140-person manufacturing company across three sites. I manage all the facilities maintenance ordering — roughly $190,000 annually spread across nine vendors, reporting up to both operations and finance. I'm not an electrical engineer, and I don't pretend to be. But I'm the one who places the order, so this became my problem.

If you're the person who buys this stuff, you've probably had the feeling already: the trip is rarely about the breaker.

The First Mistake I Made

In my first year handling maintenance purchasing, I made the classic beginner error: I upsized the problem. The 400A breaker kept tripping, so we ordered a 500A. The supplier flagged it — vaguely — with a line about "confirm compatibility with upstream equipment." We skimmed it.

It tripped more. Of course it did. The 630A main upstream now acted first on a fault, because the branch breaker was no longer downstream of it in any meaningful sense.

That's when the first real lesson landed: breakers don't coordinate themselves. They're just installed next to each other. Selectivity coordination — making sure the breaker closest to a fault trips, and nothing upstream of it does — is a design decision, and somebody has to actually make it before anyone signs a purchase order.

Here's where the communication breaks down, by the way. I said "compatible with what's there." The supplier heard "same amperage, same mounting, same pole count." Both of us were using correct words. We meant completely different things. Our electrical contractor — the only person who asked what was upstream — caught it from the drawings.

What Selectivity Coordination Actually Requires

From the outside, buying breakers looks like buying parts. The reality is that you're buying a relationship between devices, and that relationship has to be verified on paper before it exists in copper.

Selectivity means comparing trip curves across every overcurrent range — long-time, short-time, and instantaneous — and confirming there's no overlap. During an overload, the upstream device has to wait long enough for the downstream one to clear. During a short circuit, same thing. There's almost always a band, usually at high fault current, where two devices race each other. In that band you don't have coordination. You have luck.

Three tells that nobody did this work on your behalf:

• The vendor sends a catalog page instead of a selectivity table
• Nobody asks what your upstream device is, or what rating it carries
• When you ask for time-current curves, you get steered somewhere else

That last one is a red flag I ignored for two months. Our first supplier kept saying "standard coordination" and never once produced a curve.

Per IEC 60947-2, manufacturers publish coordination data for their own breaker families. The catch is that this data is verified for tested combinations, usually of the same make. Mix a downstream device from one manufacturer with an upstream device from another and what you have is a calculated guess, not a tested result. That distinction cost us most of a quarter.

Electronic Trip Unit vs Thermal Magnetic: Installation Isn't the Hard Part

There's no shortage of pages titled some version of "how to install electronic trip unit vs thermal magnetic." Most of them cover wiring, mounting, torque specs. All correct, all deservedly generic. But installation is the easy half of the job and, honestly, the wrong half to worry about.

A thermal-magnetic breaker has a fixed curve. The settings are baked into the frame size at the factory. You install it and it behaves. You can also install it badly, but you can't make it behave differently than it was built to.

An electronic trip unit is the opposite. Adjustable long-time, short-time, instantaneous settings; ground fault on some models; communications. It can do almost anything you want — provided somebody calculated the settings, wrote them in, and tested them.

I learned that one in the fall of 2023, when we swapped a feeder onto an electronic-trip molded case breaker and then... left it alone. Nobody programmed anything. The factory defaults had no relationship whatsoever with what was downstream. For months it sat there uncoordinated, and after the next trip our contractor plugged in a laptop and read back the values, all of them untouched.

Fine hardware. Completely uncalibrated.

So if you're the one drafting the RFQ and the line item says "matching model," add a second line: submit the setting schedule and confirm field verification. Price it separately if you can. Otherwise what arrives is a device, not a coordination design.

"IEC Circuit Breaker" Is Not a Specification

This one still irritates me.

If your RFQ says "IEC circuit breaker" and nothing else, you've written a sentence without content. IEC 60898-1 covers household and similar overcurrent protection, generally up to 63A, with defined breaking capacity categories. IEC 60947-2 covers industrial circuit breakers with entirely different test criteria and a different way of expressing what the device can actually interrupt. Writing the family name without the standard number means one supplier quotes you 60898 and another quotes you 60947-2, both legitimately "IEC," and you've just installed two different object types and called them the same thing.

Nobody flags this. It doesn't show up on a price sheet. It doesn't appear as a returned invoice six months later. It shows up as a device that interrupts less than the available fault current, on the one day it matters.

My RFQ now requires: standard number and part, frame rating, poles, Icu, and Ics. That last one is the one to chase. Icu is what the device can interrupt once. Ics is what it can interrupt repeatedly. Plenty of quotes list only the first, because the first is the larger number.

Residual current breakers have their own version of this problem. We had a panel where a 30 mA RCD sat downstream of a 300 mA RCD, and neither was given a delay setting. A ground fault would take both of them out together, which defeated the entire point of having an upstream device. Two manufacturers, two compliant products, no coordination. When we re-quoted those circuits we specifically asked for delay class on the data sheet, and VoltShield was one of the few that printed it on the spec table instead of burying it in the manual.

Arc Flash Mitigation Is a Design Outcome, Not a Product Feature

This is the term I'd most like to see retired from product pages.

We went looking at arc flash mitigation circuit breaker options in 2024, mostly because our insurer asked a question we couldn't answer. What I found is that nothing you can buy off a shelf reduces incident energy by itself. What reduces incident energy is a combination: a maintenance-mode setting on the trip unit that forces faster instantaneous action under a defined current, short-time delay chosen so that the upstream device clears in one or two cycles, and — critically — an arc flash study that actually calculates the numbers for your system.

The accepted method is the one in IEEE 1584. Calculations based on it tell you what incident energy you have at each working distance, with each breaker configuration. Buy a breaker with a maintenance-mode switch and never use it, and you've paid a premium for a feature that exists only on the label.

That's roughly what happened to us. We paid extra for two devices specifically because they carried that wording, then discovered our maintenance team had never been trained to enable the mode and our electrical contractor had never been asked to document the setting. The equipment was correct. The process around it was empty.

What It Cost

Not catastrophic. Expensive enough to be embarrassing.

Roughly $4,300 in unnecessary hardware — the wrong-sized breaker, the second replacement we didn't need, and one device with a feature we never used. Add maybe fifteen hours of contractor time across three visits, most of it spent re-diagnosing. Add production losses from four trips, one of which took down a whole building for almost an hour.

And then the part that doesn't show up on any invoice. In October 2024, a customer's procurement team was on-site for a supplier audit. At 7:10 in the morning, the feeder on Building 3 tripped. Nobody was hurt, the audit continued, the auditor was gracious about it. But she wrote it down, and she mentioned it twice more that day.

That's the piece I keep coming back to. Your plant is the first thing a customer sees about your company. When a nuisance trip puts half your floor in the dark while they're standing in it, that's not an electrical problem anymore. That's the impression they carry home.

What I Changed

Short list. The analysis took months; the fixes took an afternoon of paperwork.

• Selectivity table required at quote. Not offered — required. No table, no order.
• Trip unit type and setting range written into the RFQ, with a separate line for the setting schedule and field verification.
• Standard number and part in the specification, plus Icu and Ics as separate fields. If Ics is missing, I ask once, then move on.
• Arc flash work goes through a study first. IEEE 1584 calculation, documented settings, and a training session for the maintenance team before any device with a maintenance mode gets installed.
• One vendor conversation instead of nine. We consolidated the breaker and switchgear spend with a supplier willing to sit in a room with our contractor, which VoltShield did — they shipped a setting schedule with each electronic trip unit and put the adjustment ranges on the data sheet where the rest of us can read them.

The Short Version

If your line keeps tripping and the answer you're getting is "replace the breaker," ask one more question before you sign: who confirmed the coordination? If nobody can answer that, the replacement will work for about three weeks, like ours did.

This worked for us at 140 people, three buildings, and a fairly predictable load profile. If you're running a plant with seasonal spikes or a lot of motor starting, your curves will look different and my numbers won't transfer. Get the study done either way.

The bottom line: the money you save on uncoordinated equipment gets paid back in downtime, in contractor callbacks, and in whatever a customer thinks when the lights go out during their walkthrough.