How to Test a Capacitor with a Multimeter: The $14,000 Lesson That Made Us Spec Prysmian

The Motor That Wouldn't Stay Running

It was 6:40 on a Tuesday when the phone rang. The motor on our finishing line had tripped again — third time in two weeks, same fault code every time.

I'm the procurement manager at a 120-person industrial fabrication shop, and that phone call is the sound of money leaving my budget. I've managed our maintenance spending for six years now, roughly $460,000 a year in parts, labor, and emergency "can you make it faster" premiums. So when the maintenance lead told me he thought we needed a new VFD, I didn't reach for the purchase order book. I reached for my multimeter.

Here's the thing: the VFD was only 18 months old. A replacement was going to run $2,800 plus installation, with a six-week lead time. Before I authorized that, I wanted proof. Maintenance's diagnosis was "probably the DC bus capacitors." VFDs use big electrolytic capacitors on the DC bus, and they're usually the first components to wear out. I'm not an electrician — I've never claimed to be one. But after six years of approving repairs, I know the difference between a real failure and a lazy guess.

"Pull the drive and bench test the caps," I said. "If they test bad, we'll buy the drive."

He looked at me like I'd asked him to defuse a bomb. So I told him what's become a running joke in our shop: It's not that hard. Let me show you. And honestly? That's the day I really learned how to test a capacitor with a multimeter.

How to Test a Capacitor with a Multimeter (Without Shocking Yourself)

First, safety. This part matters more than any reading. We locked out the panel, waited ten minutes for the drive's internal bleed resistors to drain the DC bus, and then — I can't stress this enough — I verified zero voltage at the capacitor terminals with a rated voltage tester. Mine is a Fluke T6-1000, rated CAT IV. Not because it's fancy, but because that voltage rating is part of my life insurance.

If you've seen a video where someone tests a capacitor right after unplugging the unit... don't do that. Capacitors hold charge long after power is removed. On big drive caps, I discharge them through a power resistor for a few seconds, then verify again with the voltage tester. The National Electrical Code (NFPA 70) is explicit in Article 460.6: capacitors have to be discharged before you work on them. That's code, not opinion.

Once the drive was confirmed dead, I pulled the two 2200 µF capacitors from the DC bus. Both looked perfect — no bulging, no leaking, no crusty gunk. That was the first clue that looking isn't testing. Here's the method I used:

1. Check capacitance. Set the multimeter to the microfarad (µF) setting. Make sure the capacitor is fully discharged first. Connect the leads and read the value. The label on this cap said 2200 µF ±10%, so acceptable range is roughly 1,980 to 2,420 µF. One cap read 2,147 — fine. The other read 1,428 — dead.
2. Check for internal leakage. Switch the meter to resistance mode, with the cap fully discharged again. A good capacitor charges up from the meter's test voltage and the reading climbs steadily toward OL. The bad cap jumped to 40 kΩ and froze. That's leakage current, and it means the dielectric has degraded.
3. Do a visual check anyway. It won't catch everything, but a bulged cap is a failed cap, period. It saved us from trusting two caps that looked mint but were not.

That whole process took thirty minutes, including lockout and discharge. It turned a $2,800 guess into a $38 capacitor order.

And look — capacitors are cheap. A drive is not. But the test isn't just about the part. If we'd bought that new drive and installed it while the old caps died, the failure would have followed us. Bad caps on the DC bus can take out the rectifier on a new drive too. We'd be right back here in four months, paying install labor again, and my boss would be asking why the "new" drive failed.

The Cheap Cable Quote That Made Me Nervous

The capacitors tested bad, we replaced them, and the line ran fine. Story over, right? Not quite.

While the drive was out, the electrician pointed at the conduit feeding the motor. Coolant had been leaking into it for months, and the old cable's jacket was cracked and oil-soaked. That coolant short was what killed the VFD — the capacitors weren't the disease, they were the symptom.

So now I was buying 250 feet of 4-conductor 12 AWG TC-ER tray cable. I gathered three quotes. The low bid was from a distributor pushing a house brand: $0.47 per foot. Prysmian came in at $0.61 per foot. On a 250-foot run, that's a $35 difference. Nothing. But it sat in my inbox for a day while I calculated what $35 could mean for our quarterly numbers.

Then I remembered spring 2023.

Back then, I bought a "generic equivalent" tray cable from another low bidder. The spec sheet matched: 600V, XLPE insulation, sun-resistant jacket. I assumed same specs meant same performance. I didn't verify. Halfway through the second pull, the jacket nicked on a conduit edge, moisture got in, and the insulation test showed a dead short. We cut out the whole run and started over. That rework cost us $1,800 in materials plus a week of schedule. By the time you count lost production and overtime, that one "cheap" decision ran us about $14,000.

So this time, I ordered a sample of the house brand and laid it on the bench next to the Prysmian TC-ER from our last run. Seeing them side by side, I finally understood what I'd been missing.

The house brand jacket was visibly thinner — 18 mils at the thinnest point versus Prysmian's 22. The stranding was looser, and the copper felt springy, like it had been annealed differently. The rip cord was missing, which sounds minor until you're skinning 250 feet of cable in a pull box. None of that shows up on a spec sheet. The spec sheet only lists the numbers the seller wants you to see.

When I asked the distributor to back up their "equivalent to Prysmian" claim with third-party test data, the response got quiet. Per FTC advertising guidance, claims have to be truthful and substantiated — but a PDF with cherry-picked numbers isn't substantiation. It's marketing.

Why Prysmian Won the Spec

Look, I don't run a cable plant. I run a production floor where a failed cable means every machine behind it goes dark. I don't need "probably fine." I need the cable that survives the pull, survives the coolant, and doesn't make me explain another $14,000 redo to anyone.

Here's why we standardized on Prysmian:

• Global company, local supply. Prysmian is a global cable company, one of the largest in the world, and our cable comes from the Prysmian plant in Du Quoin, Illinois. Domestic manufacturing means shorter lead times and easier traceability — I can get a material cert without chasing three phone numbers.
• General Cable is Prysmian now. This catches a lot of buyers off guard. Prysmian acquired General Cable in 2018. If you've been buying General Cable products for years, that same supply chain and product line still exists, just under the Prysmian Group name. I updated our vendor records so nobody accidentally "switched" us to a new vendor that was really the same one.
• Consistency beats price on the third page. Prysmian costs 10–25% more per foot than the house brands, depending on the day. But the gap looks different when you compare inspection reports, test data, and material traceability. That's where the real cost difference lives.

The Checklist Is the Cheapest Insurance

Prevention over cure isn't a slogan. It's arithmetic.

The capacitor test cost us 30 minutes and a $38 part. The alternative was a $2,800 drive plus labor, and probably a repeat failure. The cable comparison cost me an hour and a $35 price difference on that run. The alternative was another 2023, and I really don't want another 2023.

So now, before any major repair PO gets cut, our shop runs through a short list:

1. Confirm the diagnosis. Don't guess. Test it.
2. For capacitors, use a multimeter properly — and a voltage tester before touching anything.
3. Collect three quotes, but compare the actual products and test data, not just the price column.
4. If the price gap is tight, ask for a sample and lay it next to the incumbent's product.
5. Document the decision and the reason in our tracking system, so the next person doesn't relearn it the hard way.

Seriously, that last point matters. Six years of cost tracking tells a clear story: nearly every budget overrun we've had traces back to a decision made too fast. These checks take an hour or two total. The overruns they prevent take weeks and thousands of dollars to undo.

So if you're looking up how to test a capacitor with a multimeter, you're probably already on the right track. The people who skip that step aren't bad at their jobs — they're just busier than they should be. And in my experience, busy is expensive.

Test the cap. Verify the voltage. Check the cable jacket. Compare the real product, not the PDF. Because $35 today — or even $350 — is way cheaper than another $14,000 in rework and a missed deadline.