The $14,000 Lesson I Learned About Checking Before You Buy

So, it was a Tuesday afternoon. Not even a dramatic day—just a quiet one, sifting through purchase orders for our quarterly automation supplies. I'm a procurement manager for a mid-sized manufacturing outfit here in the Midwest. For the last six years, I've managed a budget of about $180,000 a year, negotiating with vendors large and small. I'm supposed to be good at this. I'd like to think I'm good at this.

But let me tell you about the time I almost threw over $14,000 down the drain because I didn't take five minutes to check something.

How It Started: The Normal Tingle of a Good Deal

We needed a replacement controller for a line that was starting to act flaky. Our maintenance lead, Jack—we call him Jack Gold Rush because he's always chasing the next emergency, and he actually found a bit of gold up in Alaska once, wild story—Jack flagged the Omron CJ2 we had operating one of our packaging cells. It was throwing random stop codes. Not critical yet, but annoying, and it was starting to impact our throughput.

Our go-to vendor, a local distributor I've worked with for years, quoted us a brand-new Omron CJ2-CPU32 at about $2,800. Good price, solid service. But our finance team, in a push to cut costs, asked me to shop around. So I did. I found a surplus electronics dealer who listed the exact model for $2,100. Sizable saving—about 25%. And the listing looked solid: good photos, decent description. It said it was 'pulled from a working environment, tested.' That's a red flag term actually—or rather, a yellow flag. 'Pulled from working' is better than 'untested,' but it's not a guarantee.

The numbers said go with the surplus option. Save $700. The budget looked better. My gut said stick with my distributor. Something felt off. I couldn't put my finger on it—maybe the lack of a warranty, maybe the slightly evasive email replies. But I went with the numbers. Because numbers don't lie, right?

The Middle: When Numbers Reveal the Truth

The PLC arrived on a Thursday. Brown box, decent packaging. Jack installed it Friday morning. By Friday afternoon, it worked for about 45 minutes, then started throwing the same stop codes. Not exactly the same—the original was a timing issue, this looked like a power supply issue.

I was annoyed, but I wasn't panicking yet. I figured it was a simple configuration error. I spent Saturday morning at the shop—well, Saturday late morning, after my daughter's soccer game—and double-checked all the wiring against the manual. Everything looked right. I then pulled out my multimeter to check the incoming voltage to the PLC.

And there it was. The incoming line voltage was stable at 118 VAC. Good. But the relay output on the Omron CJ2 wasn't switching properly. I set my meter to DC and checked the input signal from the sensor. The sensor was sending 24 VDC. The PLC was receiving it. But the output was sitting at like... 8 volts. Not enough to trigger the next relay.

I pulled the PLC and did a continuity test on the output terminals. The internal relay was welded shut. Flat out broken. The 'tested' surplus unit was defective. The listing said it worked. It was clearly being returned because someone else had the same issue and didn't want to deal with it—or rather, they didn't have the time to test it properly, they just shoved it back in a box.

The Hidden $7,000 Overrun

Now, the PLC itself was $2,100. That was the visible cost. But here's where the TCO kicks you. We had already installed it. We had already wired it. Labor cost for Jack? Four hours at $65 an hour. That's $260. But we also lost a production day on Friday while Jack was troubleshooting—that's about $4,500 in lost output. Then we needed a rush order from my original distributor because I refused to try another surplus unit. My distributor quoted $3,200 for an overnight express—$400 premium on the base price. Plus the shipping cost of returning the defective unit, which was on us because the seller didn't disclose the exact defect—$65. And the weekend I spent? That doesn't have a price tag, but it's non-zero.

Total overrun from trying to save $700: $4,500 lost production + $400 rush shipping + $260 labor + $65 return shipping = $5,225. Plus the $2,100 sunk cost for the defective unit. That's a seven-thousand-dollar mistake on a two-thousand-dollar purchase.

The Plot Thickens: The DuraxV Extreme Twist

Okay, so I'm already feeling like an idiot. But the story doesn't end there. While I was inspecting the unit, I noticed the part number on the sensor was a DuraxV Extreme—a heavy-duty inductive proximity sensor we use for high-vibration applications. This wasn't even supposed to be on this line. Someone, somewhere down the chain—I suspect our night shift lead—had swapped it in from a different cell as a 'temporary fix' for a different issue. And the sensor was drawing just slightly more current than the CJ2's internal power supply could deliver for its outputs. That's why the internal relay welded shut. The PLC wasn't broken. It was overloaded.

So the PLC we returned? Probably not defective. The testing didn't check for load. The test was just 'does it power on,' not 'does it function under load.' That's a huge difference.

The Reckoning: How a Multimeter Saved (and Cost) Us

I had mixed feelings about this mess. On one hand, I was furious I trusted a cheap source. On the other hand, I was relieved we figured it out before we ordered a new PLC and cooked that one too. The real lesson wasn't about the PLC brand—Omron makes solid gear, the CJ2 is a workhorse. I've seen them running for a decade on some lines. The lesson was about process.

What I Changed: The 12-Point Check

I created a simple 12-point checklist. Not a fancy system, just a spreadsheet. Now, before any new equipment is installed in a cell, we do three things:

1. Measure input voltage.
2. Measure the output load. Use a multimeter in series (or a clamp meter if you have one) to see how many milliamps the sensor actually draws.
3. Check the PLC's internal power supply spec. The Omron CJ2 datasheet says its internal 24 VDC power supply can handle 300 mA total for outputs. If you exceed that, you need an external power supply. The DuraxV Extreme draws 200 mA by itself.

That 12-point checklist I created after my third mistake has saved an estimated $8,000 in potential rework so far. Five minutes of verification—or in this case, thirty seconds with a multimeter—beats five days of correction.

The Numbers Don't Lie, But They Don't Tell the Whole Story

I've got to come clean about something else. Part of me still thinks about that $2,100 deal. Part of me wants to be smarter and find the hidden gem. Another part of me knows that the only reason we got away with that was because of the structural problems in our own plant. I'm still on the fence about prioritizing pure price over total security, but I've built a compromise. We still shop around, but we always get a quote from 3 vendors minimum. And we always, always budget for a contingency of 15% on any automation upgrade—because if it can go wrong, it will. It sounds pessimistic, but it's worked so far.

So, How to Use a Multimeter to Test Voltage?

If you're reading this and thinking, 'I should check my gear too,' here's the short version. It's not complicated, but it's everything.

1. Set the dial to V~ (AC) for standard wall power; V— (DC) for most sensor and PLC signals.
2. Plug the black lead into COM, the red lead into VΩmA.
3. Touch the leads to the terminals you want to measure. Red on positive (or hot), black on negative (or neutral).
4. Read the display. For a standard 24 VDC sensor, you should see something between 21.6 and 26.4 VDC per the manufacturer spec. For a 120 VAC line, between 108 and 132 VAC is acceptable.

Per USPS guidelines (okay, not USPS—per IEC 61131-2, the standard for PLCs): the nominal voltage for a 24 VDC system is permissible within a range of -15% to +20%. That's 20.4 VDC to 28.8 VDC.

Parting Thoughts

I've attached my 12-point checklist PDF for anyone who wants it (just email me). It's saved us a lot of headaches. And I'll leave you with this: don't be me. Don't learn this lesson the hard way. The five minutes you spend checking now could save you five days of explaining to your CFO why the production line is down.