The $4,950 Lesson: How a Komori Press, a 3D Printer, and an Epson Workforce Taught Me to Check Everything

The air in the shop that morning was a mix of solvent, ink, and the faint sweet stench of a 3D printer running way too hot. I remember thinking: This is what production looks like. Then the phone rang. It was one of our best clients, asking where their 5,000 brochures were.

They were on the floor. Half of them. The other half were already in the recycling bin.

I've been handling commercial print orders for 13 years. I've personally made—and documented—seven significant mistakes, totaling roughly $24,000 in wasted budget. This was mistake number seven. The one that finally forced me to write the pre-production checklist we now run on every job. The one that taught me the difference between looking at a machine and understanding it.

The Komori Was the Easy Part

In early 2021, we upgraded from a digital printer to a used Komori Lithrone. It came with the KHS Hyper System—Komori's automated make-ready setup. Before we bought it, I spent a week reading every spec sheet I could find. I searched for "khs komori hyper system offset printing" and learned about automatic plate changing, remote ink control, the works. It sounded bulletproof.

From the outside, a Komori offset printing machine with KHS looks like a fast path to productivity. The reality is, it's only as fast as your pre-press process. All the automation does is make mistakes faster. I didn't understand that until the day we ran 5,000 copies of the wrong color.

We also didn't send anyone to a formal training class. I figured the KHS's touchscreen would walk us through it. It has a lot of screens. What it doesn't have is a voice saying: "Your proof is from an inkjet, genius."

Here's what we had on the floor that morning:

• A Komori offset printing machine—the Lithrone, with the KHS Hyper System, freshly installed.
• A cheap open-frame 3D printer we used for making custom brackets and rollers.
• An Epson Workforce all-in-one in the office, for invoices and, it turned out, color proofs.

Three tools. All three "working" at the same time. All three contributing to the mess.

The Proof Is in the Proofing

The job was a 5,000-piece brochure for a medical device company. The client's marketing lead had approved a color proof—printed on the Epson Workforce. Our customer service rep, who was new, actually asked, "Is the Epson Workforce an inkjet printer?" I said yes. It is. It's an inkjet. I even added, "It's a good one."

But that was about as useful as saying the Komori's feeder elevator goes up and down. We didn't ask the next question: Is an office inkjet a contract proof for offset lithography? It isn't. Not even close.

The brochure had a medical device logo with a specific teal—Pantone 3385. On the Epson proof, it looked saturated and crisp. The client approved. We loaded the plates on the Komori, entered the KHS setup parameters, and let the press run.

The first 300 sheets were perfect. Then the teal shifted. Not a lot—maybe 5% Delta E. But on a medical logo, "not a lot" is a fail.

We had the press operator make an ink adjust. Overcorrected. Then overcorrected again. We burned three hours chasing a target that was moving because we were comparing to an uncalibrated inkjet proof. The surprise wasn't the color deviation. It was that an $80 inkjet cartridge could derail a $100,000 press.

The KHS system is genuinely good at what it does—shortening changeovers, stabilizing ink flow. But it's a tool, not a supervisor. We treated it like a supervisor. That's not a fair complaint about the press; it's a confession about us. And the Komori itself? It never wavered. It held registration to 0.05mm. We were the variance.

The Warped Part

While we fought with color, the 3D printer was in the corner trying to print a new transport bracket for our folding machine. It had been running for two hours with no 3D printer enclosure with ventilation—just a bare frame on the same shelf as the water cooler. The part came out covered in stringing and with a layer height so bad it was practically an abstract sculpture.

I'd seen the guides. I'd read about ABS fumes and particulates. But the 3D printer was a one-time hobbyist thing until the day it became a production tool. That was the difference. We later found the printer's 3D printer OS was three releases behind. A known bug in that old version caused the first-layer height to vary. The system had been prompting us to update for weeks. I saw the notification, told myself we'd do it after the order. After turned into never. The bug was in the slicing engine, not the printer itself, but the OS controlled the slicer. Same result.

The Numbers

Let's be specific about what that day cost:

• $2,300 in paper and ink—wasted test runs on the Komori.
• $1,800 in overtime—reloading, washing up, reprinting the final run.
• $120 for the overnight replacement bracket.
• $730 in lost labor for the 3D printer, the setup, and the failed parts.

Total: $4,950. Plus a 4-day delay and a client who didn't renew. It could have been worse. If we hadn't caught the color shift, the client would have mailed 5,000 miscolored brochures to doctors. Under FTC advertising guidelines (ftc.gov), those visual claims would have been misleading. We don't want to test that legal boundary.

And if we'd sent them as direct mail, USPS size and thickness rules would have been another trap. A 6.125-by-11.5-inch piece is fine if it's under 0.25 inch thick. Ours wasn't in danger, but we didn't check that either. We were the definition of "ready to fail."

What Finally Fixed It

After the third rejection in Q1 2022—yes, there was a third, a different job—I created a 12-point pre-production checklist. It's the best $4,950 I ever spent. Since we put it into place, we've caught 17 potential errors in 14 months. Maybe 16. I'd have to check the log. The number doesn't matter. The habit does.

Here's the part that matters:

1. Confirm final artwork against the original spec sheet.
2. Run a controlled proof on a calibrated device. Not the office inkjet.
3. Verify the Komori's ink density with a spectrophotometer before the first production sheet.
4. Check the 3D printer's enclosure and outdoor vent before any long part print.
5. Update the 3D printer OS and slicing software. Set a reminder.
6. Ask, "Is this proof actually what the press will produce?" If the answer is "probably," stop.

Prevention Over Cure

I used to think that checking twice slowed things down. It doesn't. A five-minute verification can save you five days of correction. That's not a slogan; it's a spreadsheet formula.

The KHS Hyper System on our Komori offset printing machine is a serious piece of engineering. It sets up the press, but it doesn't tell you that your proofer is a glorified document printer. It doesn't tell you that your 3D printer's enclosure is missing, or that the OS hasn't been updated in months.

(Should mention: we now have a proper 3D printer enclosure with ventilation. The air quality is better. The noise isn't great. Better than inhaling ABS fumes.)

We also answer the "Epson Workforce" question internally every time someone asks. Yes, it's an inkjet printer. No, it's not used for proofs anymore. It's for invoices and labels.

The final check: Before you commit to any deadline, ask what you haven't verified. The answer will tell you more than any spec sheet. If it's "nothing," you haven't thought hard enough.

"The press never fails on its own. It fails because someone skipped a check." — The note I keep next to the coffee machine.

That's the long way of saying: check first. I have the wavy bracket, the $4,950, and the checklist to prove it.