Mark Andy Flexo UV Drying: 7 Checks I Learned After $6,000 of Rework

I've been running flexographic presses for eight years—the last four on a Mark Andy press with the Mercury UV curing system. In my first two years, I made UV ink drying mistakes that added up to roughly $6,000 in wasted material and rework. Not a catastrophic number in the scope of a print shop, but embarrassing. And most of those failures were preventable. I do not mean that as a theoretical statement—I mean I've lost real money to these exact mistakes.

This checklist is for anyone running UV inks on a flexo press. If you're shopping for a Mark Andy ProSeries press, the price tag is significant, but the real costs show up in consumables and maintenance. This article covers the UV drying side, which is where I lost the most. Here are seven checks I run on every UV job now. No theory, just what I've learned from messing up.

Check 1: Measure UV intensity with a radiometer

In 2017, I ran a 200-piece label order with a UV lamp that was roughly 40% below spec. The lamp looked fine. It was glowing. The hour counter said we had 300 hours of life left. But the ink wasn't fully curing. The customer called two days later saying ink was rubbing off on their fingers. That was $890 in redo, plus a one-week delay. I still kick myself for not buying a radiometer sooner.

A radiometer measures UV output in mW/cm². Your ink supplier will give you the minimum cure threshold for their chemistry. I'm not an ink chemist, so I can't speak to exact formulation interactions. What I can tell you from a production perspective is that the radiometer reading is ground truth. The hour counter is a suggestion at best.

Here's something vendors won't tell you: the lamp hour rating assumes continuous running. If your press runs in short bursts, the lamp ages faster than the counter suggests. If you run long shifts, it lasts longer. Measure, don't guess.

Check 2: Confirm your ink matches your lamp type

This sounds obvious, but it's the second-most common failure I've seen. Not all UV inks cure the same way under all lamps. A mercury vapor lamp has a different spectral output than an iron-doped lamp or a UV LED system.

I once ordered a batch of ink formulated for LED curing while our Mark Andy flexographic UV ink drying system uses a mercury lamp. The ink sat there, looking and smelling like the right thing, but it wouldn't harden properly no matter how long I ran it under the lamp. That was $450 down the drain—plus the time spent troubleshooting everything else before I checked the data sheet.

The check is simple: read the ink's technical data sheet. It specifies the recommended UV source. If it doesn't, call the ink manufacturer before you commit.

Check 3: Cure-test on the actual substrate

Substrates absorb and reflect UV differently. A test on white BOPP tells you nothing about how the same ink will cure on metallized film.

In September 2022, I was running a job on metallized film. The test sample cured perfectly. The production run didn't. The difference? The sample was cut from the outside of the roll; the production run came from the middle, which had a slightly different coating thickness. Every piece of that 500-piece order had cure issues. $650 in material, straight to scrap.

Now I cure-test on the exact material, from the actual roll, with the actual ink and lamp settings. And I test three separate areas of the web. This is different from DTF printing or inkjet. A ColDesi DTF printer, for example, has its own curing variables. And photo paper for inkjet printers relies on the coating to absorb ink. Flexo is the opposite—the ink sits on the surface and needs to crosslink. The substrate is part of the cure equation.

Check 4: Set line speed to match dwell time

The relationship is simple: total cure energy = intensity × time. If you double the line speed, you halve the dwell time under the lamp. To get the same cure, you need double the intensity.

Here's where I messed up. I increased the press speed by 15% to hit a deadline and didn't adjust the lamp power. The job looked fine coming off the press. Three days later, the ink hadn't fully hardened and the labels smeared. That cost $780 in redo plus a 3-day production delay.

When you change line speed, re-check the radiometer reading under the lamp. Most modern UV systems—including the Mark Andy Mercury system—let you set power levels that correspond to line speed. Use that feature. It exists for a reason.

Check 5: Give mercury lamps time to warm up

Mercury vapor lamps don't reach full power instantly. They need time to vaporize the mercury inside the arc tube. If you hit start and immediately run the press, you're curing at maybe 60-70% of expected output.

What most people don't realize is that the lamp's ready indicator doesn't mean full power. It means ignited. In my first year, I ran 50 sheets before the lamp reached intensity. All 50 went in the trash.

Wait at least 5-10 minutes after ignition before starting production. If your curing unit has a mounted radiometer, wait until the reading stabilizes.

Check 6: Inspect reflectors for dirt and oxidation

This is the one everyone skips. Reflectors in UV curing systems degrade over time. They collect dust, and the aluminum surface oxidizes. The lamp can be perfectly new, but if the reflector is dull, you're losing 30-50% of the UV energy that should reach the substrate.

And here's the trap: the lamp looks bright to your eye. The machine sounds normal. The radiometer, if you're checking the direct beam, might read acceptably. But the reflected energy—a major portion of what actually reaches the substrate—is gone.

I discovered this after a $3,200 order came back with inconsistent cure on one edge. We spent two days troubleshooting the lamp, the ink, the substrate. Finally, a service tech asked when we'd last cleaned the reflectors. We hadn't. Ever. The press had been running for 14 months.

Inspect reflectors weekly. Clean them with the manufacturer-recommended solution. Replace them if you see pitting. It's a 15-minute check that saves a lot of pain.

Check 7: Keep a job log

The most expensive lesson: memory isn't a reliable record. You might think you'll remember the lamp power, line speed, and ink batch you used for a job. You won't, a month later.

I keep a simple log now. Date, press, job number, ink batch, substrate, lamp hours, radiometer reading, line speed, lamp power, cure test result. Thirty seconds to fill out after each setup.

This is the same principle as looking up the WPS PIN on an HP printer. You won't need it most days, but when the printer won't connect, you're glad the manual is where you expected it to be. My job log is that manual for my press. We've caught 47 potential errors using this checklist in the past 18 months. The log is what made the diagnosis fast.

Before you call a technician

If you've run through all seven checks and still see curing problems, the issue could be in the ink formulation itself. That conversation is for your ink supplier, not a press technician.

This also gets into electrical engineering territory, which isn't my expertise. If you suspect the UV lamp driver or power supply is failing, don't open it up unless you're trained. High-voltage components hold charges after the machine is off. Call a qualified technician.

Why this matters for your brand

When labels come back with ink that rubs off, the customer doesn't think "they had a UV lamp issue." They think "this shop prints sloppy work." I've seen it happen. A $50 maintenance item turned into a lost account worth thousands.

There's also a color angle. If you're matching brand colors, the industry standard is a Delta E of 2 or less. Inconsistent curing means inconsistent color, because the ink's final appearance depends on how thoroughly it crosslinked.

The checks above are cheap. A radiometer is a few hundred dollars (as of early 2025, at least, when I last priced one). A job log is a notebook. Cleaning reflectors takes 15 minutes. Compared to the cost of rework and damaged trust, it's nothing.