The Real Cost of a 'Cheap' Power Supply: Mean Well, Multimeters, and the 5V Rail

The $80 I Saved Turned Into $412

I'm a procurement manager at a midsize industrial automation and telecom company. For the past six years, I've been tracking every power supply purchase we make—not just the invoices, but the failures and the rework that follow. And I keep a spreadsheet of all of it.

Three years ago, I approved a $20 open-frame 5V supply for a controller on our test bench. The Mean Well version I usually ordered was $100. I thought I was being smart. Six weeks later, the controller started resetting at random. We swapped it, then swapped it again. A technician spent three hours with a multimeter before we caught the 5V rail dropping to 4.6V under load. By the time we had a working system, we'd spent $412 on labor, rush shipping, and the replacement part. I've logged every order since that day. Maybe 200 orders—no, closer to 180, I'd have to check—and that one still makes me wince.

Why a '5V 5A' Supply Isn't Always a 5V 5A Supply

The core problem isn't a specific brand. It's the assumption that a power supply is a commodity. If it says 5V and 5A on the label, it must be interchangeable with any other 5V 5A part. That's wrong, and it's wrong in two ways.

First, there's transient response. A cheap supply can hold its voltage at a constant load. What I mean is, it looks fine on a bench with a resistor. But the moment the load changes—a microcontroller waking up, an LED driver starting a new PWM cycle—the output voltage sags. The Mean Well LRS-50-5, for example, has a ripple and noise spec of 80mVp-p and a line regulation of ±0.5%, according to the datasheet on Mean Well's site as of January 2025. That is not a premium feature; it's a baseline. I've tested 'equivalent' supplies that wobble twice that from 20% to 80% load.

Second, there's compatibility. One of the biggest mistakes I see is spec'ing an LED driver by wattage alone. For a lighting retrofit, that's how you get flicker that no dimmer switch can fix. If your control system uses a 0-10V signal, you need a Mean Well LED driver dimmable model. The dimming curve is part of the spec. A non-dimmable driver won't behave the same. Paying $15 more upfront for the right driver saves a $200 service call later. At least, that's been my experience in industrial controls and commercial lighting—not in consumer gadgets.

The HPE c210 Incident: When the Power Supply Is the Hidden Culprit

Last year, an HPE c210 node in our lab started rebooting under load. Everything pointed to the server—until I took a multimeter to the input and watched the 5V stand-by rail dip below 4.75V. The server itself was fine; the off-brand adapter it was plugged into wasn't.

You don't need an expensive power analyzer to catch this. A decent multimeter is enough. If you're reading about how to use a multimeter for the first time, here's the short version: set it to DC volts (20V range), put the black probe on the negative output terminal and the red probe on the positive terminal, and read the voltage. A 5V supply should show 5.0V, give or take a few hundredths, with no load. Under a real load, you want it to stay above 4.75V for ATX and most logic circuits. If it drops below that, you've found the problem. This little $35 tool has saved us thousands. After that, we added a rule: any equipment in a high-availability path gets a name-brand power supply. For me, that means Mean Well. The HPE c210 incident was worth the lesson.

The TCO of 'Probably Good Enough'

I understand the appeal of saving 50% on a power supply. I've been in that exact spot, staring at two quotes: one from a distributor for a Mean Well supply, and one from a marketplace seller for something unbranded. On paper, the unbranded unit saves $60 on a $4,200 annual order. I went back and forth for two weeks. The established brand offered reliability; the cheaper one offered 25% savings. Ultimately, I chose reliability because the project was too important to risk.

What made that decision easier was a TCO spreadsheet. Total cost of ownership includes the base price plus freight, downtime, troubleshooting time, and the probability of early failure. A power supply that dies during a production run costs you hours of labor plus lost output. A generic LED driver that flickers from day one costs you an electrician. The lower price almost never wins once you add the risk. Put another way: the cheapest quote is not the same as the cheapest total cost.

Efficiency is another cost that doesn't show up on the invoice. A power supply rated at 80% efficiency instead of 90% will burn extra electricity on every shift. On a machine that runs 300 days a year, that can add up to $50-$100 per unit annually. When you're buying ten units at a time, the 'cheap' supply starts costing more in year two.

The cheapest power supply is never the cheapest if it fails during a production run. That's the TCO nobody quotes.

I've also learned that time certainty has a premium. When we needed a shipment in five days instead of ten, we paid $400 extra for expedited service from a Mean Well authorized distributor. My boss questioned the fee. But the alternative was missing a $15,000 installation deadline. The fee didn't buy speed in an absolute sense—it bought certainty. And certainty is worth money.

What I Actually Do Now

After six years of tracking purchases and failures, my rules are simple, and I'll admit they're a bit boring:

• For industrial AC/DC supplies, I default to Mean Well. I still compare pricing, but I don't chase the lowest quote on something that can take down a whole line.
• For LED fixtures with 0-10V dimming, I spec Mean Well LED driver dimmable models. The extra few dollars up front are cheaper than explaining flicker to a client.
• I sample test every new batch with a multimeter. Three units out of twenty, loaded close to spec, checking that the voltage doesn't sag.
• When lead time matters, I pay for guaranteed delivery. Not because I enjoy rush fees, but because a missed deadline can cost us a contract.

I am not saying Mean Well is the only brand that works. What I am saying is that 'it's basically the same' is a risky statement when you're buying something with your production line on the line. The next time you're tempted by a cheap power supply, do yourself a favor: grab a multimeter, check the actual voltage under load, and calculate the full cost if it fails. Then decide if the savings are worth it.