Grounding and Bonding Problems in Industrial Facilities

Most of the strange electrical problems I get called out for in industrial plants trace back to grounding and bonding, not the equipment everyone wants to blame. Nuisance trips, VFDs faulting for no reason, controllers locking up, motors humming and dying early. When I start pulling on those threads, the root is almost always a ground path that isn't doing what someone assumed it was doing.

Here's the plain version: grounding gives fault current and lightning somewhere to go, and bonding ties all the metal together so it sits at the same voltage. When either one is done wrong, you don't always get a dead short. You get a plant that acts haunted.

Why does bad grounding cause "ghost" problems instead of just tripping a breaker?

Because a poor ground path doesn't fail cleanly. It works most of the time and then misbehaves under load, under a fault, or when a big motor kicks on.

A loose or corroded bond, a missing connection between two steel structures, or a ground electrode that's degraded will let small voltage differences build up between things that are supposed to be at the same potential. That difference is what your sensitive electronics feel. A PLC input sees noise. A drive sees a reference that's drifting. An operator gets a bite off a machine frame that "isn't even energized."

None of that shows up on a quick meter check during normal operation. It shows up intermittently, which is exactly why these problems get chased for months before someone calls me.

What are the most common grounding and bonding mistakes I find?

After walking a lot of plants across South Carolina, the same handful of issues come up again and again:

• Equipment grounding conductors that were undersized or left out when someone added a machine and just tied into the nearest steel.
• Isolated ground receptacles wired without a real isolated ground path, so they're isolated in name only.
• Ground loops created when a piece of equipment ends up with two different ground references fighting each other.
• Neutral and ground bonded in more than one place downstream, which puts current on the grounding system that was never supposed to be there.
• Bonding jumpers missing across conduit fittings, painted steel, and expansion joints, so the metal isn't actually electrically continuous.
• Corroded ground rods and connections, which get worse fast in our humidity and in plants with any chemical exposure.

That last one matters more here than people think. South Carolina heat, moisture, and coastal salt air toward the Lowcountry all work against buried and exposed connections. A ground system that tested fine when the building was new can quietly degrade over years.

How do I actually find the problem on site?

I don't guess and I don't start replacing parts. I test the system the way it behaves under real conditions.

First I map what's actually there against what the prints say, because in an older facility those two things rarely match after decades of additions. Then I check bonding continuity across the metal that's supposed to be tied together: structural steel, raceways, machine frames, panel enclosures. A lot of "grounding" problems are really bonding gaps.

From there I measure ground reference differences between points that should read the same, and I look at what the power itself is doing. Harmonics from drives and switching supplies ride on the system and interact with a weak ground in ugly ways, so I bring in power quality analysis when the symptoms point that direction. Voltage on the ground conductor, current where there shouldn't be any, and noise timed to a specific motor starting all tell me where to dig.

I'll also run thermal imaging on connections under load. A bond that's carrying current it shouldn't, or a high-resistance joint, usually runs warm before it fails outright.

Why does this get worse as a facility ages and adds equipment?

A plant is never wired once. It gets a new line, a new compressor, another row of machines, and each addition gets tied into whatever ground was closest at the time.

Over twenty or thirty years, that turns into a patchwork. The original grounding electrode system was sized for the original load and layout. Nobody goes back and re-evaluates the whole ground path when the third VFD-driven line goes in, so the system ends up carrying noise and current it was never designed to handle.

Modern equipment makes it worse. Older motor loads were forgiving. Today's drives, servos, and control systems are sensitive to the same voltage differences that a contactor and a relay never noticed. Good industrial grounding in South Carolina facilities matters more now precisely because the equipment cares more than it used to.

What does fixing it involve?

It depends entirely on what the testing turns up, and I won't pretend otherwise. Sometimes the fix is adding proper bonding jumpers and cleaning up connections. Sometimes it's correcting a neutral-to-ground bond that's in the wrong place. Sometimes the electrode system itself needs work.

What I don't do is throw a longer ground rod at a bonding problem, or add an isolated ground to chase noise that's really coming from a ground loop. That's how facilities end up spending money and still having the same faults a month later.

The right sequence is measure first, understand the whole path, then correct the specific thing that's broken and verify it under load. If there's permitted or inspected work involved, your local authority having jurisdiction can confirm what applies to your facility before you commit to a scope.

When to Give Me a Call

If your plant has intermittent faults, drives dropping out, control systems glitching, or you're just not confident your ground system keeps up with everything you've added over the years, that's exactly the kind of work I do. I'll come out, test what's actually there, and tell you what's real instead of guessing.

Call or text (803) 565-0783 and we'll set up a time for me to look at your facility.