Troubleshooting guide

UV Curable Coatings

Gloss, matt, slip and the coater unit: the faults that actually stop UV coating jobs on flexo, gravure, roller, duct and screen.

Uniform Industries · Greater Noida, India · ufindustries.com
Figures are typical for the class and carry their test method. The values that govern a supply are those on the Technical Data Sheet for the grade.

Before you start

How to use this guide

Work down the page. Cheap checks before expensive ones, and measure before you change anything.

Handling comes before diagnosis. Uncured UV coating is a skin irritant and a potential sensitiser, and sensitisation does not reverse: wear nitrile gloves and eye protection for the pan, the circuit, wet or suspect-cured sheets and for every cure check in this guide, so every thumb twist below is a gloved thumb twist. Never clean coating off skin with solvent, wash with soap and water. MEK and wash-up solvents are highly flammable and defatting, so use them in ventilation, away from lamps and ignition sources, with lids closed and rags in a closed metal bin.
Isolate and lock off before any hand goes into the machine. Lamp housings and quartz stay hot enough to burn long after switch-off, so let them cool and prove the UV extraction has cleared before opening a housing; handle quartz and LED windows with clean gloves only, since skin oil etches quartz and shortens lamp life; inspect and turn rollers only with the drive isolated; and change emitter to substrate distance only within the machine builder's setting, with every guard and interlock restored before the lamps are struck.
Confirm cure before you diagnose anything else, because an under-cured film also reads dull, blocks, smells and scuffs, and four faults chased separately are one fault.
Measure coat weight gravimetrically on the same sheets before and after coating rather than reading the coater setting, since anilox wear, blade wear and a warm pan all move delivered weight while the setting stays where you left it. Sheet to sheet grammage varies by 2 to 3 per cent inside one pallet, which on 300 g/m2 board is several times the coat weight, so a coated sheet weighed against a different uncoated sheet reports substrate noise, not coating.
Separate a wetting fault from a cure fault by when it happened: de-wetting, crawl and craters are decided in the seconds before the lamp, so more dose only sets them harder.
Never add anything to a UV coating at the press. It is close to 100 % solids, so every addition is reactive monomer, silicone or wax, and each one buys the symptom back at the folder-gluer, the foiling machine or the laminator.
Treat the base coat and top coat of a drip-off, and the ink and the coating over it, as one specified system; a general gloss grade over a leafing metallic, a conventional oxidative ink or a low-migration ink undoes the grade underneath it.
Adhesion, glue bond, foil hold, seal and migration belong to the finished construction, not to the coating, so qualify them on the customer's own board, laminate and adhesive after 24 h and again after 7 days.

UV Curable CoatingsHow to use this guide

The process

What happens, in order, and the one variable held at each step.

The process

What is controlled at each step

UV Curable CoatingsThe process

Control variables

What to hold, and how to measure it

Variable	Target	How it is measured	How it drifts
Coating viscosity at the pan	Roller gloss 45 +/- 5 s; high-porosity gloss 65 +/- 5 s; metallised-PET gloss 55 +/- 5 s; gravure OPV 35 +/- 5 s. These are Ford No. 4 figures at the house temperature	Ford cup No. 4 to ASTM D1200 at 25 +/- 1 C as the house condition, or ISO 2431 with a 4 mm ISO cup at 23 +/- 0.5 C, which is a different scale and gives a different number. Quote cup, standard and temperature with every reading, sample and cup both at temperature, drawn from the pan and not from the drum. Screen, tactile and crystal grades are outside the range of any flow cup: control them on a rotational viscometer with the spindle, speed and temperature stated	Falls roughly 10 to 15 per cent for every 5 C rise as the press hall and the pan warm, which is about 5 to 7 s on a 45 s cup and 7 to 10 s on a 65 s cup, then climbs again if stray UV or heat starts to gel the return.
Coating temperature	22 to 28 C in the pan or chamber, held, and below 30 C in the return line	Probe thermometer in the pan, read hourly at the same point, plus a reading on the return line	Rises with lamp radiant heat, pump work and press hall temperature. A 10 C rise is a visible gloss and weight change on its own.
Applied coat weight	Class indications, not a specification; verify gravimetrically on your own stock. Narrow-web flexo OPV 1.5 to 3.0 g/m2; gravure OPV 1.5 to 3.0 g/m2; sheet-fed chambered anilox gloss 3.0 to 5.0 g/m2; two-roll or three-roll roller gloss 4 to 6 g/m2 on coated board and 6 to 10 g/m2 on porous or uncoated board; matt grades 2.0 to 3.0 g/m2 with a ceiling near 3 g/m2 above which the matt is buried, and this matt window overrides the general window of whatever unit the matt is run on; screen flood gloss 12 to 20 g/m2; screen tactile, crystal or haptic 25 to 40 g/m2. A UV coating is close to 100 % solids, so applied and cured weight are within a few per cent	Gravimetric on the same sheets: mark and weigh five sheets to 0.01 g, feed them through the coater, weigh them again within a few minutes of cure and divide the mean gain by the sheet area. Where marked sheets cannot be fed, weigh twenty coated and twenty uncoated sheets from the same pallet and report the mean with its spread, stating that substrate variation and not the coating sets the uncertainty. On web, punch samples of known area from the same reel position coated and uncoated. State the application method with every figure	On a roller or doctor-roll coater weight falls as the coating warms and thins, because film split at the metering nip scales with viscosity. On an anilox or gravure system the dominant drift is cell plugging and blade wear; a warmer, thinner coating usually releases from the cells slightly more completely, so temperature moves weight less and can move it the other way. Say which system you are on before you use temperature as a lever.
Delivered dose at the substrate	Indicative for this class: 80 to 150 mJ/cm2 for a thin flexo or gravure OPV; 100 to 200 mJ/cm2 for a roller-coated gloss; 200 to 400 mJ/cm2 for a heavy matt or a heavy roller gloss; 400 to 1200 mJ/cm2 for a screen-applied tactile, crystal or haptic film, the figure rising with film weight. Confirm through cure at the heaviest section rather than assuming the window	Band-matched radiometer carried through the press at running speed and read on the coated area. Match the head to the emitter: UV-A (about 320 to 390 nm) for mercury and for 365 and 385 nm arrays, UV-V (about 395 to 445 nm) for 395 and 405 nm arrays, or an instrument with an LED-specific calibration. Read across the width as well as down the centre. Record the instrument, the band and the calibration date with every reading, and never compare a reading from one instrument with a window set on another	Falls in inverse proportion to line speed at constant lamp power: double the speed and you halve the delivered dose. It also falls with lamp hours, quartz devitrification, dirty reflectors and dirty LED windows.
Peak irradiance at the substrate	Mercury medium-pressure 1.5 to 4 W/cm2 in the band measured; UV LED 4 to 16 W/cm2 in the array's band. Quote the band and the instrument with the figure	Same radiometer pass, reading peak rather than integrated dose	Falls as the emitter ages and as the web or sheet path moves away from focus. What matters in service is the trend against the reading logged when the emitter was new, not the absolute number. The same dose at half the irradiance is not the same cure.
Lamp or LED output	Log hours and log the radiometer reading against them; replace on measured output, not on a calendar	Radiometer pass logged at every lamp change and weekly thereafter, same head, same band, same speed, same position	Mercury loses short-wave output first, which is exactly the band that cures the surface, so the film goes tacky on top long before the broadband reading looks bad. An LED array loses output slowly and evenly, so the trend line is the only warning you get.
Substrate and printed surface energy	Above 40 mN/m for BOPP, LDPE and CC-PET; above 38 mN/m as a floor for surface work	Dyne pens or ISO 8296 test inks, or ASTM D2578, on the reel or the pile about to run and on the printed area, never taken from the film certificate	Corona decays in storage, faster in heat and humidity; a reel treated eight weeks ago may be 4 to 6 mN/m below its certificate.

UV Curable CoatingsControl variables

Control variables

What to hold, and how to measure it

Variable	Target	How it is measured	How it drifts
Specular gloss	Standard roller gloss above 90 GU at 60 degrees with the 20 degree reading logged beside it; metallised-PET board above 95 GU at 60 degrees; satin 40 to 60 GU at 60 degrees; deep matt 10 to 20 GU at 60 degrees with the 85 degree reading logged	ISO 2813. Use 60 degrees as the reference geometry, and add 20 degrees above about 70 GU and 85 degrees below about 10 GU as ISO 2813 directs, because 60 degrees stops discriminating at both ends. Report the geometry with every figure, five points across the sheet and at head, middle and tail, mean and spread both logged	Drifts down with coat weight and coating temperature; on matt grades it drifts up as the matting agent settles out of the pan or out of the drum.
Pile or reel temperature at delivery	Below 35 C at the pile centre before wrapping or before the next operation	Probe pushed into the centre of the pile within 5 min of delivery, and again after 30 min	Climbs through a long run as lamp heat accumulates in the stack. A pile that is fine at sheet 500 blocks at sheet 5000.
Anilox or gravure cell volume	Grade specific. Log the nominal volume and the date of the last verified clean	100x loupe or a cell-volume scan on the roll, plus the gravimetric coat weight it is actually delivering	Cells plug progressively with part-cured coating and board dust. Delivered volume falls quietly, and the operator compensates with blade pressure until the roll is damaged.
Wetted parts on the coater unit	Every roller covering, blanket, plate, hose, gasket, chamber end seal and pump part in contact with the coating specified for 100 % solids UV chemistry, normally EPDM or the machine maker's UV part, with no conventional NBR or natural rubber anywhere in the circuit	Check the specification of every wetted part against the machine and coating supplier lists, particularly on a unit converted from water-based, and inspect the parts for swelling, softening, a tacky surface or a raised end seal with the machine isolated	A wrong elastomer swells and softens over weeks, so the roller goes out of round and prints its own pattern, the end seals leak, and extractables bleed back into the coating and move its viscosity and its cure.
Coating stock in the drum	Within the stated shelf life, stored between 5 and 25 C, out of direct sun and never against a lamp housing, a radiator or a hot wall; stock rotated so the oldest drum runs first	Check the batch number and date on every drum before it goes into the circuit, log what was opened when, and roll or stir a matt or slip grade to redisperse settled matting agent and wax before it reaches the pan	Matting agent and slip additive settle in a standing drum, so the first pan off a long-stored drum is low in matt and low in slip while the last pan is high in both; warm storage depletes inhibitor and shortens the life of everything in the drum.

UV Curable CoatingsControl variables

The mechanism

Why gloss is levelling, and why matt inverts the rule

The same lever moves gloss and matt in opposite directions. Gloss needs enough film to bridge the roughness of the ink and board underneath; matt needs its particles to stand proud of the film, so above the grade ceiling extra weight buries the effect. This is why more coating rescues a dull gloss and destroys a matt, and why the two faults must never be treated with the same adjustment. Read the weights against the application method: a roller coat and a narrow-web flexo OPV are a different scale of film.

UV Curable CoatingsThe mechanism

02 | 20 faults

Faults

Symptom, cause, check, change, confirm. The trap is the move to avoid.

Fault 1 of 20

Gloss down, or gloss patchy across the sheet

WHAT YOU SEE

The sheet reads flat against the approved sample, or the heavy-ink areas and the fold panels go dull while the margins stay bright. It is normally seen at the stacker under the inspection light, not at the coater.

WHAT IT LOOKS LIKE

CAUSES, MOST LIKELY FIRST

Coat weight below the window for the application methodGloss is specular reflection and needs a film smooth on the scale of the wavelength of light. Below about 4 g/m2 on a roller coat, or about 1.5 g/m2 on a narrow-web flexo OPV, the film follows the micro-roughness of the ink and board instead of filling it.
Coating too warm at the pan, so a roller or doctor-roll coater meters lighter than the setting saysViscosity falls roughly 10 to 15 per cent for every 5 C rise and the metering nip splits less material at the same gap. On an anilox or gravure system this lever is much weaker and can run the other way, so look at cell condition first.
Absorbent, uncoated or low-print-area stock drinking the coatingThe vehicle wicks into the fibre before cure and leaves a resin-starved surface that scatters rather than reflects.
Ink not set, or too much anti-set-off powderA mobile ink surface and powder grains both give the coating a rough foundation that no amount of levelling can bridge.
Wrong grade for the jobA standard gloss laid on a porous, low-print-area board is the exact case the high-performance grade exists for, and no press setting substitutes for it.

UV Curable CoatingsGloss down, or gloss patchy across the sheet

Fault 1 of 20 · continued

Gloss down, or gloss patchy across the sheet

CHECK, IN THIS ORDER

Confirm cure first, gloved: MEK double rubs after ASTM D5402 at the house conditions, looking for 50 or more with no breakthrough, since an under-cured film also reads dull
Weigh coat weight gravimetrically on the same marked sheets before and after coating, and compare against the window for the application method in use
Read gloss to ISO 2813 at five points across the sheet and at head, middle and tail, adding the 20 degree geometry on high gloss above about 70 GU, and log the spread and not just the mean
Measure pan temperature with a probe and pan viscosity with a Ford No. 4 cup to ASTM D1200 at 25 C, or an ISO 2431 4 mm cup at 23 C, on coating drawn from the pan, and state which
Inspect the metering with the machine isolated: anilox or gravure cells under a 100x loupe, or doctor roll setting and nip parallelism across the width

THEN CHANGE

Bring coat weight back inside the window for the unit you are running before changing anything else
Cool the coating supply to 22 to 28 C and hold it there; jacket or chill the pan if the press hall runs hot
On porous or low-print-area board move to the high-performance gloss grade, the UniCure-11RC class, and run it at 6 to 10 g/m2 on a roller coater rather than laying more of a standard grade on
Cut spray powder to the minimum the pile needs and give conventional ink adequate set time, or prime the board first
Where the stock is the limit, prime and then coat rather than coating twice

CONFIRM IT IS FIXED

Gloss to ISO 2813 back inside the grade figure, above 90 GU at 60 degrees for a standard roller gloss with the 20 degree reading logged beside it, the spread across the sheet inside 5 GU, and gravimetric coat weight inside the window on three sheets taken an hour apart.

THE TRAP

Adding a reducer to make the coating flow better. A UV coating is close to 100 % solids, so the only thing available to thin it is reactive monomer: it lowers viscosity, lowers delivered weight at the same setting on a roller coater, lowers gloss further and shifts the cure demand at the same time.

UV Curable CoatingsGloss down, or gloss patchy across the sheet (continued)

Fault 2 of 20

Surface tacky, body of the film hard

WHAT YOU SEE

A gloved thumb twist marks the coating while a fingernail will not scratch through it. Sheets mark in the pile, and the tack is often worst at the edges of the width or on the areas furthest from the lamp centre.

WHAT IT LOOKS LIKE

CAUSES, MOST LIKELY FIRST

Dirty reflectors, dirty LED windows or a fouled quartz plateCoating overspray and board dust on the optics cut delivered irradiance by tens of per cent with no change in lamp setting, and this is the thing that has usually changed since the job last ran well.
Lost output from an aged lamp or a devitrified quartz sleeveSurface cure depends on the shorter wavelengths absorbed in the first microns, and those go first as a mercury lamp ages while a broadband reading still looks acceptable.
Coat weight crept above the windowA heavier film needs more energy to reach the same conversion, and weight has usually drifted up before the tack appears.
Speed raised or lamp power dropped, so delivered dose fellDose is inversely proportional to speed at constant power: double the speed and you halve it.
Substrate or web away from the lamp focusIrradiance falls steeply with distance, and the same dose at half the peak irradiance does not give the same surface cure.

UV Curable CoatingsSurface tacky, body of the film hard

Fault 2 of 20 · continued

Surface tacky, body of the film hard

CHECK, IN THIS ORDER

Gloved thumb twist on the coated area plus MEK double rubs after ASTM D5402, so surface cure and through cure are judged separately
Take a radiometer pass at running speed on the coated area, reading both dose and peak irradiance, with the head band matched to the emitter: UV-A for mercury and 365 or 385 nm arrays, UV-V for 395 or 405 nm arrays
Read at both edges of the width as well as down the centre, since a tacky edge and a sound centre is an emitter length or reflector width fault
Read lamp hours against the last logged radiometer figure, and inspect the quartz sleeve for a milky, devitrified look
With the machine isolated and the housing cool, wipe and inspect reflectors, LED windows and the quartz plate, then repeat the radiometer pass to see what the clean recovered

THEN CHANGE

Clean the optics first, with the machine isolated and locked off and the housing cool; this is the cheapest and most common recovery. Lint-free cloth and the maker's cleaner, and clean gloves on quartz because skin oil etches it
Replace the lamp or the sleeve on measured output rather than on hours
Raise delivered dose by lamp power or by adding a lamp, not by slowing the press
Bring coat weight back to the window
Where the coated width exceeds the effective cured width, move the image inboard or fit the correct emitter length; do not try to cover it with power

CONFIRM IT IS FIXED

No mark from a firm gloved thumb twist immediately at the delivery, 50 or more MEK double rubs after ASTM D5402 with no breakthrough at the house conditions, and a logged band-matched radiometer reading inside the grade window at full running speed, taken at both edges and the centre.

THE TRAP

Slowing the press. On presses whose lamp power tracks web speed the dose does not rise at all, so nothing changes; and where it does rise, slowing has only masked a lamp, a reflector or a window that has lost output and will lose the next job as well.

UV Curable CoatingsSurface tacky, body of the film hard (continued)

Fault 3 of 20

Coat weight drifts down through the run

WHAT YOU SEE

Gloss and rub were right at make-ready and are marginal three hours in, with no setting changed. The operator has usually already opened the blade a little to compensate.

WHAT IT LOOKS LIKE

CAUSES, MOST LIKELY FIRST

Anilox or gravure cells plugging with part-cured coating and board dustDelivered volume falls progressively while the nominal roll specification stays on the job ticket. This is the dominant drift on any engraved-roll system.
Doctor blade wear or uneven blade loadingA worn blade meters a wedge rather than a line, so weight falls and varies across the width at the same time.
Coating warming in the pan and thinning, on a roller or doctor-roll coaterFilm split at the metering nip scales with viscosity, so a thinner coating carries less out of the nip at the same gap. On an anilox or gravure unit the same warming usually moves weight slightly the other way, so do not cool the pan to chase a falling weight on an engraved roll.
Coating level falling in the chamber, or the pump drawing airAn air-laden feed gives an intermittently starved cell fill and a lighter, streaky film.
Viscosity rising from stray UV or heat, starting to gel the returnGel particles both raise apparent viscosity and plug the cells, so weight falls even as the cup reading rises.

UV Curable CoatingsCoat weight drifts down through the run

Fault 3 of 20 · continued

Coat weight drifts down through the run

CHECK, IN THIS ORDER

Gravimetric coat weight every 30 min through the run on marked sheets weighed before and after, logged, not eyeballed against the sample sheet
Anilox or gravure cell inspection under a 100x loupe with the machine isolated, looking for cells filled flush rather than open
Doctor blade contact line checked across the full width, and blade edge inspected for a wear flat
Pan viscosity by flow cup against the drum viscosity, same cup, same standard, same temperature, plus pan and return line temperature
Chamber level and pump suction checked for foam and air entrainment

THEN CHANGE

Clean the anilox or cylinder by the method specified for the roll, and do it on a schedule rather than when the weight has already gone
Change the blade and reset loading to the minimum that gives a clean wipe
Cool and shield the pan and the return line to hold 22 to 28 C, which recovers weight on a roller coater and stops gelling on any unit
Keep the chamber full and raise the pan level so the pump is never drawing air
Take gelled or heat-aged coating out of circulation instead of running it down

CONFIRM IT IS FIXED

Three gravimetric coat weight checks an hour apart inside the window for the unit, with gloss to ISO 2813 stable within 5 GU head to tail at the stated geometry.

THE TRAP

Compensating a plugging anilox by opening the blade or raising blade pressure. It gives the weight back for an hour, wears the blade and the cell walls, and leaves a roll that will no longer clean, so the next job starts from a worse baseline.

UV Curable CoatingsCoat weight drifts down through the run (continued)

Fault 4 of 20

Blocking and set-off in the pile or the reel

WHAT YOU SEE

Sheets stick face to back when the pile is fanned, or the reel shows gloss patches and ink pick on the reverse when it is unwound. It usually appears in the middle of the pile, not at the top.

WHAT IT LOOKS LIKE

CAUSES, MOST LIKELY FIRST

Marginal surface cureA tacky top few microns is enough to bond two faces under the weight of a pile, even when the bulk of the film is hard.
Heat carried into the stackLamp radiant heat accumulates in the pile; conversion goes on and the softened film flows into contact under load.
Coat weight above the windowA heavier film cures less completely at the same dose and presents more mobile material at the interface.
Insufficient slip in the grade for the pile height being runSurface slip is what lets two coated faces separate; a resist or stampable grade is deliberately low in slip and will block where a high-slip grade would not.
Rewind tension too high on web workContact pressure in the reel core is far above what a flat pile applies, so a film that passes in the pile fails in the roll.

UV Curable CoatingsBlocking and set-off in the pile or the reel

Fault 4 of 20 · continued

Blocking and set-off in the pile or the reel

CHECK, IN THIS ORDER

MEK double rubs after ASTM D5402, plus a gloved thumb twist for surface cure specifically
Probe the pile centre temperature within 5 min of delivery and again after 30 min
Block test: two coated faces, or a coated face against the uncoated reverse, in contact under a stated load, typically 1 kg/cm2 as an accelerated condition, held at 40 C for 24 h, then peeled and examined for transfer and gloss change
Coefficient of friction, coated face to coated face and coated face to uncoated reverse, at 23 C and 50 % RH, by the method agreed with the converter
Gravimetric coat weight, and rewind tension on the web

THEN CHANGE

Bring delivered dose to the top of the grade window, cleaning the optics first with the machine isolated
Cool the delivery: chill roll, cooling fans, lower pile height, and fan the stacks before wrapping
Reduce coat weight to the window
Where the pack allows it, move to a higher-slip grade such as the UniCure-16RC class, or the high-slip gravure OPV class where a coefficient of friction below 0.2 is wanted
Slacken rewind tension and rewind smaller diameters on marginal jobs

CONFIRM IT IS FIXED

No transfer and no measurable gloss change after the block test at 40 C for 24 h under the stated load, with the pile centre below 35 C before wrapping. Note that 1 kg/cm2 is about ten times the pressure a 1 m pile of carton board applies, so it is a severe screening condition and not a simulation of the pile; where a marginal result has to be judged, repeat it at the pressure and temperature the actual pile sees.

THE TRAP

Dusting the pile with spray powder to stop the sticking. Powder on a freshly coated surface destroys the gloss the job was coated for, and it is the first thing that shows up as a crater or a bond failure in a downstream lamination, foiling or gluing operation.

UV Curable CoatingsBlocking and set-off in the pile or the reel (continued)

Fault 5 of 20

Fisheyes, craters, pinholes and specks in the coating

WHAT YOU SEE

Small round holes where the coating has pulled back to a clean rim, or specks sitting under the film, seen best by holding the sheet at a glancing angle to a point light. Often in one area of the sheet or one lane of the web rather than everywhere.

WHAT IT LOOKS LIKE

CAUSES, MOST LIKELY FIRST

Silicone contaminationSilicone lowers surface tension locally, the surrounding coating pulls away from the low-tension spot and leaves a crater with a clean centre.
Anti-set-off spray powder on the printed surfaceEach grain is a non-wetting particle the coating cannot cover, giving a crater with a solid nucleus at 30x.
Wax or slip additive migrating out of the ink filmThe additive blooms to the ink surface during set and presents the coating with a low-energy patch, worst over heavy solids.
Air entrained in the supply, or the chamber foaming and spitting at speedA pump drawing air, a low pan level or a chamber whipped by the roll at speed carries bubbles into the film that burst at cure and leave pinholes; a failed or lifted chamber end seal adds drips and splash marks at the edges that are the same fault at a different scale.
Board dust and fibre pulled onto a charged sheet before the coaterA sheet leaving the press carries static, and dust attracted to it sits under the film as a speck with a meniscus around it rather than as a clean crater.

UV Curable CoatingsFisheyes, craters, pinholes and specks in the coating

Fault 5 of 20 · continued

Fisheyes, craters, pinholes and specks in the coating

CHECK, IN THIS ORDER

Inspect ten consecutive sheets at a glancing angle to a point light and count defects per 100 cm2
Look at the centre of a defect at 30x: a solid nucleus means powder, dust or dried coating, a clear centre means silicone or oil, a clean round hole with no rim means entrained air
Check the pan and the return for foam, check pan level against the pump suction, and check the chamber end seals for leakage and drips at the edges of the coated width
Check the anti-static bars are live and the hall air filtration is working, since the same dust returns as soon as the sheet is charged again
On duct application, check fount conductivity, pH and feed, and look for emulsification in the coating on the roller train

THEN CHANGE

Find and remove the silicone source. Ban silicone aerosols and silicone-containing anti-set-off compounds from the coating hall, not just from the coater
Strip and clean the whole coating circuit including pump, hoses, filter and chamber, since a contaminated circuit reinfects fresh coating
Cut spray powder to the minimum and change to a coarser, harder powder where it cannot be removed
Raise the pan level, fit or clean the return filter so the pump is never drawing air, and renew chamber end seals that are leaking or lifted
Prove the anti-static bars and blow the sheet off before the coater

CONFIRM IT IS FIXED

Ten consecutive sheets free of craters, pinholes and specks at 100 cm2 inspection at full running speed, with gloss to ISO 2813 inside the grade window.

THE TRAP

Adding a flow additive or a silicone slip aid to cover the craters. Using a silicone to cure a silicone defect drops the surface tension of the entire film, which does stop the craters, and also stops anything adhering to the coating afterwards. Glue, foil, laminate and overprint all fail downstream, by which time the job is converted.

UV Curable CoatingsFisheyes, craters, pinholes and specks in the coating (continued)

Fault 6 of 20

Orange peel, ribbing or roller pattern

WHAT YOU SEE

The coated surface has a regular texture that shows as banding or a dimpled look at a glancing angle. Gloss readings are respectable but the sheet does not look right next to the sample.

WHAT IT LOOKS LIKE

CAUSES, MOST LIKELY FIRST

Coating too cold and too viscous to levelLevelling is driven by surface tension against viscosity; a coating below its grade temperature simply does not flow out before it reaches the lamp.
Too little open time between the metering nip and the lampThe film is frozen by cure before it has finished levelling, so it preserves the pattern the nip put into it.
Worn, out of round or wrong-durometer application rollerA roller with a flat or an uneven crown prints its own profile into the film and no amount of coating covers it.
A roller covering, blanket, hose or seal that is not UV-compatible, swelling in the coatingConventional NBR and natural rubber parts swell and soften in a 100 % solids acrylate, so the roller goes out of round and prints its own pattern within weeks, and the swollen part bleeds extractables back into the coating and moves its viscosity as well. This is the usual answer on a unit converted from water-based, and only EPDM or the maker's UV-specified part belongs in the circuit.
Anilox too coarse for the weight being laidCoarse cells lay discrete beads that must coalesce; if the coating is viscous or the run is fast they do not fully merge and the cell pattern survives.

UV Curable CoatingsOrange peel, ribbing or roller pattern

Fault 6 of 20 · continued

Orange peel, ribbing or roller pattern

CHECK, IN THIS ORDER

Measure pan temperature and viscosity by flow cup, stating cup, standard and temperature
Note the pattern's direction and its pitch. Bars running across the web and repeating at the circumference of the application or metering roller point at runout, a flat or nip bounce, and you can confirm it by measuring the repeat and matching it to a roller diameter. Continuous stripes in the running direction point at the metering: a nicked or worn blade edge, a plugged lane of cells, or film-split ribbing that gets worse with speed and with viscosity
With the drive isolated, check roller durometer with a Shore A gauge, turn the roller slowly by hand looking for flats and runout, and measure the diameter at three points for swelling against the specification
Check the specification of every wetted elastomer on the unit, particularly on a coater converted from water-based
Measure the nip or doctor roll gap at three points across the width

THEN CHANGE

Bring the coating to the grade temperature, warming rather than thinning it
Increase the distance or the time between the coater and the lamp where the press allows it, or switch off an interdeck lamp that is freezing the film early
Regrind or replace the application roller and reset the nip parallel across the width
Replace any conventional NBR or natural rubber covering, blanket, hose, gasket or end seal in the coating circuit with EPDM or the machine maker's UV-specified part, and check every wetted part on a converted unit rather than the obvious one
Move to a finer anilox appropriate to the coat weight

CONFIRM IT IS FIXED

No pattern visible at a glancing angle to a point light on ten consecutive sheets, with gloss to ISO 2813 varying by no more than 5 GU across the width at the stated geometry.

THE TRAP

Increasing coat weight to flood the pattern out. A heavier film of the same over-viscous coating carries the same ribbing, and it now also brings cracking at the crease, blocking in the pile and a bigger cure demand, so one cosmetic fault becomes three mechanical ones.

UV Curable CoatingsOrange peel, ribbing or roller pattern (continued)

Fault 7 of 20

Coating crawls off silver, gold or matt ink

WHAT YOU SEE

The coating pulls back from the metallic or matt areas into islands and beads, leaving the print looking dull and patchy exactly where it should look richest. Sharp edges appear where the coating has retreated to the boundary of the ink.

WHAT IT LOOKS LIKE

CAUSES, MOST LIKELY FIRST

Leafing metallic pigment presenting a low-energy surfaceLeafing aluminium and brass flake float to the top of the ink film and carry the fatty acid coating that makes them leaf, which the UV coating cannot wet.
Wax or slip additive in the ink underneathThe additive blooms to the ink surface as it sets, dropping the surface energy below what the coating needs.
Matt ink loaded with silicaThe rough, particle-filled surface holds air, so the coating bridges rather than wets and then pulls back as it shrinks on cure.
General purpose coating used over a metallic instead of the matched gradeThe matched grades are formulated with the surface tension and film build to wet a leafed metallic; a standard gloss is not.
Drip-off base coat carried outside its imageBase coat over-transfer onto adjoining areas makes the gloss coating de-wet where it was never intended to.

UV Curable CoatingsCoating crawls off silver, gold or matt ink

Fault 7 of 20 · continued

Coating crawls off silver, gold or matt ink

CHECK, IN THIS ORDER

Dyne pens or ISO 8296 test inks on the printed metallic solid and on an adjacent process solid, on the sheet itself, not on the bare substrate
Look at whether the coating has pulled back to a clean rim or broken into islands, which separates a wetting fault from a cure fault
Check the ink data sheet for wax or slip additive, and check what was added at the ink duct during the run
Print a test strip with the coating at two weights to see whether more film overcomes the de-wetting
Confirm the base coat image on the drip-off plate or cylinder is not over-transferring

THEN CHANGE

Use the coating matched to the metallic. For UV offset printed metallised-PET board that is the UniCure-14RC class; the leafing silver and brass-flake gold metallic inks are specified with a matched UniCure coating over them and should not be run under a general gloss
Raise coat weight within the window so the film has the volume to stay wet over the low-energy area
Remove wax and slip additions from the ink at the duct
Where a metallic and a coating genuinely fight, lay the coating through a second unit at higher weight, or move the effect to screen application where the film weight is 12 to 20 g/m2 rather than 4 to 6
Correct base coat over-transfer on drip-off work

CONFIRM IT IS FIXED

No de-wetting anywhere on the metallic solid across a full pile, and gloss to ISO 2813 on the metallic area within 10 GU of the adjacent process area at the same geometry.

THE TRAP

Turning up the lamps because the metallic area looks dull and reads as under-cured. De-wetting is decided in the seconds between the nip and the lamp; more dose only sets the defect permanently and takes the sheet past the point where the coating could still have been washed off and the job reprinted.

UV Curable CoatingsCoating crawls off silver, gold or matt ink (continued)

Fault 8 of 20

Coating cracks on the crease and the fold

WHAT YOU SEE

A white line of broken coating along the crease, worst across the board grain, often not visible until the cartons come off the folder-gluer or until the next day.

WHAT IT LOOKS LIKE

CAUSES, MOST LIKELY FIRST

Coat weight too highA thick cross-linked film has more stored shrinkage stress and less ability to accommodate the strain at the outside of a crease.
Over-cureBeyond the cure window the network keeps tightening; a harder, more highly cross-linked film is a more brittle one.
Rigid grade used where a flexible one was specifiedA high-gloss, high-hardness roller coating is built for scuff resistance, not for a 180 degree fold on a heavy cover stock. The flexible grades are formulated with a much lower glass transition so that they still have useful elongation at press hall temperature.
Creasing rule and matrix wrong for the board calliperA narrow matrix forces a tight radius, and the coating is asked to stretch further than any UV film of that weight can.
Cold, dry stock at the folderTwo things change together below about 15 C. Cold, dry board fibre creases harshly and the crease radius the coating has to follow gets tighter, and a flexible grade loses part of the elongation it was chosen for as it moves towards its glass transition. A hard gloss grade is glassy at any normal press hall temperature, so on that grade the board condition is the whole story.

UV Curable CoatingsCoating cracks on the crease and the fold

Fault 8 of 20 · continued

Coating cracks on the crease and the fold

CHECK, IN THIS ORDER

Fold coated sheets across and along the grain and examine the crease at 10x, at 1 h and again at 24 h
Measure coat weight gravimetrically
Run a radiometer pass and compare the delivered dose with the grade window, looking for over-dose as readily as under-dose
Measure the creasing matrix width against the board calliper and check the rule height
Take the stock temperature at the folder-gluer infeed and check the board moisture and how the pallet was stored

THEN CHANGE

Reduce coat weight to the lower half of the window first, since it is the fastest and most reliable lever
Bring dose down to the lower end of the window, confirming the MEK rub count still passes
Move to a flexible grade: the UniCure-17RC class for book covers and high-gsm paper, or the UniCure-15RC class where gloss and foldability are both needed
Correct the creasing matrix and rule for the calliper, and crease along the grain where the design allows
Condition the stock and the folder-gluer hall above 18 C and keep the board at its normal moisture before folding

CONFIRM IT IS FIXED

No cracking visible at 10x on a 180 degree fold across the grain on five sheets taken from the pile, assessed 24 h after coating on stock conditioned to the folding hall.

THE TRAP

Turning the lamps up because the crack looks like under-cure. Cracking is usually the opposite fault: more dose tightens the network further, makes the film harder and more brittle, and the next fold is worse than the last.

UV Curable CoatingsCoating cracks on the crease and the fold (continued)

Fault 9 of 20

Rub and scuff failure, worst on matt

WHAT YOU SEE

The coated surface burnishes to a shine or transfers onto the facing sheet when rubbed, and matt areas go glossy in the handled zones. Gloss jobs usually pass the same test comfortably.

WHAT IT LOOKS LIKE

CAUSES, MOST LIKELY FIRST

Matt film by definitionThe effect is made by matting particles protruding through the surface, and those particles are the first thing a rub block touches, so a matt is inherently lower in rub than a gloss of the same chemistry.
Under-cureAn incompletely cross-linked binder does not hold the matting particles and does not resist abrasion, so cure must be excluded before the grade is blamed.
Coat weight below the matt windowA thin film leaves too little binder around each particle and too little total material to absorb the abrasion.
Standard matt specified where a high-rub matt was neededHigh-rub matt grades use harder, better-bound particles and a tougher binder; a decorative matt on a handled pack is the wrong product, not a bad batch.
Matting agent settled in the drum or the panSettled matting agent gives a first pan that is short of particles and a last pan that is loaded with them, so rub and gloss both wander through the drum.

UV Curable CoatingsRub and scuff failure, worst on matt

Fault 9 of 20 · continued

Rub and scuff failure, worst on matt

CHECK, IN THIS ORDER

MEK double rubs after ASTM D5402 first, to take cure out of the argument
Rub resistance on a Sutherland-type rub tester after the practice of ASTM D5264, with the house conditions stated because the practice does not fix them: 4 lb block, 100 cycles, print to print, at 23 C and 50 % RH, assessing both transfer and gloss change on the rubbed area
Gravimetric coat weight against the matt window of 2.0 to 3.0 g/m2
Gloss to ISO 2813 to confirm which matt level is actually on the sheet, adding the 85 degree geometry on a deep matt below about 10 GU
Repeat the rub after 24 h and after 7 days, since a UV film continues to harden

THEN CHANGE

Confirm and correct cure before anything else
Move to a high-rub matt grade: the UniCure NX Matt Coating MT-6030 class for narrow-web flexo, or the UniCure-19RC class high-matt for roller coating
Raise coat weight towards 3 g/m2, the top of the matt window, and no further, because above that ceiling the binder buries the matting particles and the finish is lost
Roll or stir the drum to redisperse settled matting agent before it reaches the pan, and keep the pan moving
Where the pack is handled hard, propose a satin at 40 to 60 GU instead of a deep matt, which buys rub resistance with a small loss of contrast

CONFIRM IT IS FIXED

The number of cycles agreed with the customer on a Sutherland-type rub tester, typically 100 cycles at 4 lb print to print at 23 C and 50 % RH, with no transfer to the facing sheet and no gloss rise beyond the agreed figure, tested 24 h after coating and repeated at 7 days. The conditions are quoted with the result, since the practice does not fix them.

THE TRAP

Raising matt coat weight to make it tougher. Above roughly 3 g/m2 the binder buries the matting particles, so the rub improves a little, the matt disappears and the job is now both the wrong finish and still short of the rub specification.

UV Curable CoatingsRub and scuff failure, worst on matt (continued)

Fault 10 of 20

Coating lifts on met-PET, BOPP or laminated board

WHAT YOU SEE

The coating comes away with the tape, or peels from the die-cut edge, on metallised or laminated stock while the same coating is solid on uncoated board. Often fine at the delivery and failing at the die-cutter the next day.

WHAT IT LOOKS LIKE

CAUSES, MOST LIKELY FIRST

Surface energy below what the coating needsBelow about 40 mN/m a UV coating cannot wet a polyolefin or polyester surface well enough to form the interfacial contact adhesion depends on.
Slip additive or silicone at the laminate surfaceBOPP and PE films carry slip agents that bloom to the surface over time, and a laminating adhesive can carry silicone; both defeat wetting regardless of corona.
Cure shrinkage stress on a non-absorbent surfaceThe film shrinks as it cross-links, and on a surface it cannot key into, that shrinkage pulls the film off rather than into the substrate.
Wrong grade for the constructionMetallised-PET board and matt-BOPP laminated board each have a grade built for them, and a general gloss substituted for either will fail the tape.
Aged corona on a stored reel or an aged laminateCorona treatment decays in storage, faster in heat and humidity, so a certificate from eight weeks ago says nothing about today's reel.

UV Curable CoatingsCoating lifts on met-PET, BOPP or laminated board

Fault 10 of 20 · continued

Coating lifts on met-PET, BOPP or laminated board

CHECK, IN THIS ORDER

Dyne pens or ISO 8296 test inks, or ASTM D2578, on the actual reel or pile about to run, target above 40 mN/m
Tape pull with a defined tape across a coated solid, removed at a stated angle and rate, at 24 h and again at 7 days, reported as percentage of coating retained
Cross-cut to ISO 2409 only where coating and substrate are thick enough that the blade does not cut through the substrate; on thin film use the straight tape pull instead
Where a laminate is delaminating rather than the coating lifting, measure the bond properly: ASTM F904 or a T-peel to ASTM D1876 on 15 or 25 mm strips at 300 mm/min, reported in N/15 mm with the mode of failure stated
Gravimetric coat weight, since shrinkage stress scales with film thickness

THEN CHANGE

Where the line has in-line corona, treat immediately before coating to above 40 mN/m rather than relying on the supplier treatment. On sheet-fed and roller-coating work, where in-line corona is rarely fitted, the levers you actually have are the four below
Use the matched grade: the UniCure-14RC class for UV offset printed metallised-PET board, the UniCure-62SR class for gloss over matt-BOPP laminated board through screen
Prime where the construction allows it
Reduce coat weight to the bottom of the window to cut shrinkage stress
Measure the dyne value on the actual pile before the job starts and reject reels and laminates whose treatment has decayed, rather than trying to coat through the problem

CONFIRM IT IS FIXED

Full retention on the tape pull across a coated solid 24 h after coating on three sheets from the pile, and no loss when the same test is repeated at 7 days on the customer's own laminate. Adhesion is a property of the construction and is qualified on the customer's material.

THE TRAP

Testing adhesion at the delivery and calling the job good. A UV film goes on developing for hours and the interface stress rises with it, so a pass at the delivery and a failure at the die-cutter the following morning is the normal sequence, and by then the whole pile has been coated.

UV Curable CoatingsCoating lifts on met-PET, BOPP or laminated board (continued)

Fault 11 of 20

Coating lifts or cures poorly over conventional oxidative ink

WHAT YOU SEE

The coating is sound over the margins and over UV ink and strips away in sheets over the heavy conventional solids, usually at the die-cutter or the folder-gluer the next day rather than at the delivery. The film over those solids may also mark and smell when the rest of the sheet does not.

WHAT IT LOOKS LIKE

CAUSES, MOST LIKELY FIRST

Conventional ink not dry through, only surface setOxidative ink dries from the top down by taking up oxygen over hours to days. A UV coating laid over a film that is still wet underneath seals it, stops the oxygen it needs, and leaves a soft layer between the board and a hard coating that the die-cutter tears apart.
Driers and antioxidants at the interfaceCobalt and manganese driers and the antioxidants that keep the ink open in the duct are radical scavengers, so the first few microns of coating over a heavy conventional solid reach a lower conversion than the same film over bare board at the same dose.
Wax and slip in the conventional inkConventional inks carry more wax than UV inks and it blooms to the surface as the ink sets, so the coating sits on a low-energy layer rather than keying into the ink.
Anti-set-off powder in the pilePowder is the price of running conventional ink and it is also a non-wetting particle under the coating, so the same job carries both problems.
No primer, and an ink series not specified for UV overprintingThe designed answers are an in-line primer or an overprintable, low-wax conventional ink series. A general UV coating laid straight onto an unmodified oxidative ink set is not a specified system, and the interface is where it fails.

UV Curable CoatingsCoating lifts or cures poorly over conventional oxidative ink

Fault 11 of 20 · continued

Coating lifts or cures poorly over conventional oxidative ink

CHECK, IN THIS ORDER

Tape pull across a coated conventional solid and across a coated unprinted margin on the same sheet at 24 h and again at 7 days: a pass on the margin and a failure on the solid puts the fault at the ink, not the coating
MEK double rubs after ASTM D5402 over the solid and over the margin, and compare the two counts
Record the time between printing and coating, and rub an uncoated printed sheet from the same pile to see whether the ink is dry through or only set
Dyne pens or ISO 8296 test inks on the printed solid and on the margin
Check the ink series data sheet for overprintability with UV coating, and check whether wax, drier or anti-set-off compound went in at the duct

THEN CHANGE

Let conventional ink dry through before coating: coat off-line on a dry pile rather than in-line over wet ink where the construction allows it
Move to the overprintable, low-wax conventional ink series specified for UV coating, or print the job in UV ink
Prime with the specified primer or sealer and coat over the primer rather than over the ink
Cut spray powder to the minimum the pile needs and use a coarser, harder powder
Reduce coat weight to the bottom of the window, since shrinkage stress on a weak interface scales with film thickness

CONFIRM IT IS FIXED

Full retention on a tape pull over the heaviest conventional solid at 24 h and again at 7 days on three sheets from the pile, with the MEK rub count over the solid within 10 rubs of the count on the unprinted margin. This is an ink, board and coating property together and is qualified on the customer's own printed stock.

THE TRAP

Raising the dose because the coating over the conventional solid reads soft. The soft layer is the ink underneath, not the coating. More dose hardens and shrinks the film sitting on top of a wet ink, so the sheet strips more cleanly at the die-cutter, not less.

UV Curable CoatingsCoating lifts or cures poorly over conventional oxidative ink (continued)

Fault 12 of 20

Slip out of specification, cartons jam or slide

WHAT YOU SEE

Cartons will not feed through the folder-gluer or the cartoner, or the opposite: a stack of coated sheets slides apart on the pallet. The coating looks perfect in both cases.

WHAT IT LOOKS LIKE

CAUSES, MOST LIKELY FIRST

Wrong grade for the required coefficient of frictionSlip is formulated in. High-slip, resist and stampable grades sit at very different friction levels by design, and no press setting moves a grade from one to the other.
Under-cureA partly cured surface is tacky, so static friction rises sharply and feeding stops.
Coat weight away from the windowSlip additives are surface-active; too thin a film does not carry enough to the surface, too thick a film dilutes the surface concentration.
Slip additive settled in a long-stored drumThe first pan off a standing drum is short of the additive the grade is specified on, so the friction figure moves without anything on the press changing.
Wax or slip additive from the ink reaching the coated surfaceMaterial blooming through a thin coating changes the friction the converter measures without changing anything the press sees.

UV Curable CoatingsSlip out of specification, cartons jam or slide

Fault 12 of 20 · continued

Slip out of specification, cartons jam or slide

CHECK, IN THIS ORDER

Coefficient of friction, static and kinetic, coated face to coated face and coated face to uncoated reverse, at 23 C and 50 % RH, 24 h after coating, by the method agreed with the converter: ASTM D1894 horizontal sled where the board can be mounted flat, or the TAPPI T 815 inclined plane for packaging board, which does not give the same number
MEK double rubs after ASTM D5402 to exclude under-cure
Gravimetric coat weight
Repeat the friction test at the temperature the converting line actually runs at
Confirm which grade is in the pan, how long the drum stood before it was opened, and whether anything was added during the run

THEN CHANGE

Specify by coefficient of friction and by method, not by description: the high-slip roller grade class where cartons must feed, and the high-slip gravure OPV class where a figure below 0.2 is wanted
Where the pack must not slide on a pallet, move the other way to a resist or stampable grade and accept the lower slip
Correct cure and coat weight before changing grade, and roll the drum to redisperse settled additive
Remove any press-added slip and requalify
Agree the acceptance figure, the method and both face combinations with the converter before the run, not after the jam

CONFIRM IT IS FIXED

Coefficient of friction inside the band agreed with the converter, on both coated to coated and coated to reverse, by the agreed method, measured 24 h after coating and repeated at the converting line temperature.

THE TRAP

Adding a slip additive at the press to get the line running. It fixes the feed that shift and leaves a low-energy surface that the glue, the foil and the laminate will not hold, so the fault moves from the cartoner to the customer's filled pack.

UV Curable CoatingsSlip out of specification, cartons jam or slide (continued)

Fault 13 of 20

Curl on low-grammage stock and thin web

WHAT YOU SEE

Coated sheets lift at the corners in the pile, or the web will not lie flat at the die station. Worst on light grammage and on one-sided full coverage, and the direction can change overnight.

WHAT IT LOOKS LIKE

CAUSES, MOST LIKELY FIRST

Cure shrinkage of a one-sided filmA UV film contracts as it cross-links; on a thin substrate with nothing on the reverse to balance it, that contraction bends the sheet towards the coated face.
Coat weight too high for the grammageShrinkage force scales with film thickness, so the same grade at the top of its window curls stock that it holds flat at the bottom.
Dose above what the grade needsExtra conversion means extra shrinkage and a tighter, less compliant film.
Stock moisture loss under the lampsLamp and infra-red heat drives moisture out of the sheet, and because the coating seals the coated face the loss is usually greatest from the uncoated reverse, which shrinks that face and curls the sheet the other way. The moisture curl may add to or oppose the cure curl and it relaxes as the pile re-equilibrates, which is why curl is measured on a cut square at 1 h and again at 24 h rather than judged at the delivery.
Grade not formulated for low-curl workThe narrow-web coating range is built specifically for very low curl on low-grammage printed substrates; a sheet-fed roller grade is not.

UV Curable CoatingsCurl on low-grammage stock and thin web

Fault 13 of 20 · continued

Curl on low-grammage stock and thin web

CHECK, IN THIS ORDER

Cut a 100 mm square from the coated sheet, lay it coated side up on a flat plate and measure corner lift in mm at 1 h and again at 24 h, noting the direction each time
Gravimetric coat weight
Radiometer pass for delivered dose against the grade window
Measure stock moisture before and after the coater where a meter is available
Confirm the grade in the pan is the one specified for the grammage

THEN CHANGE

Reduce coat weight to the bottom of the window first
Move to a low-curl narrow-web grade run at 1.5 to 2.5 g/m2 rather than trying to fix a sheet-fed grade on thin stock
Bring dose to the lower end of the window while confirming the MEK rub count still passes
Cool and rehumidify the delivery, and let the pile stand and re-equilibrate before converting
Balance the construction with a reverse coat or reverse print where the design allows

CONFIRM IT IS FIXED

Corner lift on 100 mm squares cut from the head, middle and tail of the run inside the figure agreed with the converter, measured at 24 h, with cure still passing the agreed MEK rub count.

THE TRAP

Cutting dose on its own to reduce shrinkage. It does reduce the curl, and it leaves an under-cured film that then blocks in the pile, smells and scuffs, so the job fails at the next station instead of this one.

UV Curable CoatingsCurl on low-grammage stock and thin web (continued)

Fault 14 of 20

Coating gelling in the pan, tray or chamber

WHAT YOU SEE

Viscosity climbs through the shift, seeds and lumps appear in the pan, and comet-shaped streaks start running through the coated film. The return line is usually thicker than the drum.

WHAT IT LOOKS LIKE

CAUSES, MOST LIKELY FIRST

Stray UV reaching the coatingLight leaking from the lamp housing, a reflector bouncing back along the web, or direct daylight, starts polymerisation in the pan itself.
Heat in the coating circuitAbove about 35 C thermal initiation and inhibitor depletion both accelerate, and the coating begins to build viscosity on its own.
Contamination with amine, peroxide or a foreign chemistryTraces from another product, a badly cleaned circuit, or extractables bleeding out of a swollen non-UV hose, seal or roller covering can initiate or accelerate polymerisation.
Heat-aged coating returned to stock and re-usedCoating that has been round a hot press has lost inhibitor, so it gels far sooner the second time.
Stagnant coating left in the circuit overnightStill coating in a warm chamber next to a lamp housing is the worst case for both heat and stray light.

UV Curable CoatingsCoating gelling in the pan, tray or chamber

Fault 14 of 20 · continued

Coating gelling in the pan, tray or chamber

CHECK, IN THIS ORDER

Find the leak without exposing anyone. With the lamps off and the machine isolated, inspect housing seals, shutter edges, end plates and the web path for gaps, and check that pan and chamber covers are in place
To confirm a live leak, place UV-sensitive indicator strips or a radiometer head at the pan and along the return line and run the lamps with all guards closed, then read them afterwards. If anyone must be in the hall with lamps running they wear UV-rated eye protection and cover exposed skin, extraction is proven running, and nobody looks towards the source
Probe the pan and the return line temperature and log both against the drum temperature
Compare pan viscosity with drum viscosity by flow cup, same cup, same standard, same temperature, at the start and after 4 h
Check whether returned coating has been mixed back into fresh, check the batch date on the drum, and check the last circuit clean

THEN CHANGE

Shield the pan, the chamber and the return line from the lamp housing and from daylight, and fit proper covers
Cool the circuit to hold below 30 C, jacketing the pan where necessary
Strip and clean the full circuit when a foreign chemistry is suspected, and replace any non-UV hose, seal or covering that is swollen or softened
Do not return heat-aged coating to stock. Take it out of the circuit into a clearly labelled container, leave it vented rather than sealed, keep it cool and away from lamps and direct sun, do not consolidate it with fresh material, and dispose of it as reactive chemical waste through the route on the safety data sheet. Gelling is exothermic and accelerates as it warms, so if a container is warming or firming, isolate it outside and tell the supervisor
Empty and clean the chamber and lines at the end of the shift rather than leaving coating standing

CONFIRM IT IS FIXED

Pan viscosity within 3 s of the drum viscosity on the same cup at the same stated temperature after 4 h of running, with a clean filter and no comet streaks in the coated film.

THE TRAP

Topping the pan up with fresh coating to bring the viscosity back. The gel particles stay in the circuit, plug the anilox or gravure cells, and print as comets for the rest of the run; and the fresh material is now seeded with partly polymerised coating so it gels faster than the batch it replaced.

UV Curable CoatingsCoating gelling in the pan, tray or chamber (continued)

Fault 15 of 20

Drip-off contrast weak or edges blurred

WHAT YOU SEE

The matt and gloss areas look closer together than the approved sample, or the boundary between them is soft rather than a clean line. The gloss usually looks acceptable on its own.

WHAT IT LOOKS LIKE

CAUSES, MOST LIKELY FIRST

Base coat weight too lowThe effect works by surface-tension repulsion, so the base coat must cover the image completely; a partly covered image gives a partly repelled gloss.
Gloss top coat weight too lowThe gloss coating needs enough volume to pull back from the matt image and pile up in the open areas; too thin a film cannot move.
Base coat partly cured before the gloss goes onAn interdeck lamp left on freezes the base coat surface and destroys the surface-tension difference the effect depends on.
Base coat and top coat are not the matched pairContrast comes from a designed difference in surface tension between two specified products; two gloss and matt coatings picked at random will not repel one another.
Edge build and over-transfer at the base coat imageCoating piles at the edge of a plate or a cut cylinder and prints a heavy rim, a halo at the trailing edge and base coat outside the image, all of which blur the boundary. A lifted or worn chamber end seal does the same at the edges of the width and adds drips.

UV Curable CoatingsDrip-off contrast weak or edges blurred

Fault 15 of 20 · continued

Drip-off contrast weak or edges blurred

CHECK, IN THIS ORDER

Run each unit alone and measure the base coat and the gloss coat weights separately, gravimetrically on marked sheets
Confirm every interdeck lamp between the base coat unit and the gloss unit is off and shielded
Check the two products in use against the specified matched pair: the UniCure-31DT, -50DT or -75DT class base coats with the UniCure-12RC or -20RC class texture top coats
Read gloss to ISO 2813 in the matt area and in the gloss area and record the difference, not the absolute values, adding the 20 degree geometry on the gloss area and the 85 degree on the matt
Examine the boundary at 10x for edge build, a trailing-edge halo and base coat spread outside its image, and check the chamber end seals

THEN CHANGE

Set both coat weights to specification, base coat first, and only then judge contrast
Switch off and shield the interdeck lamp
Substitute the specified matched pair rather than mixing products from different series
Raise the gloss top coat weight to give the film the volume it needs to pull back
Relieve or remake the base coat plate or cylinder edge where it is building, and renew leaking chamber end seals

CONFIRM IT IS FIXED

A 60 degree gloss difference to ISO 2813 of at least 50 GU between the matt and gloss areas, with a clean boundary at 10x on ten consecutive sheets.

THE TRAP

Increasing the matt base coat weight to get more contrast. The contrast comes from a surface-tension difference, not from thickness, so a heavier base coat barely moves the gloss difference while it spreads outside the image and blurs the edge, which is the more visible fault of the two.

UV Curable CoatingsDrip-off contrast weak or edges blurred (continued)

Fault 16 of 20

Yellowing on white board and light colours

WHAT YOU SEE

The coated area reads warm or dirty against the uncoated margin or against the approved sample, worst on white board and pastel tints and most obvious in the customer's viewing booth. It is often acceptable at the delivery and clearly out the next morning.

WHAT IT LOOKS LIKE

CAUSES, MOST LIKELY FIRST

Photoinitiator and amine synergist residuesCleavage products and unreacted amine absorb in the blue, so the film reads yellow. The floor is set by the raw material set, and it is highest on the grades that carry the most initiator: heavy films, matts and low-migration grades.
Post-cure yellowing over the first 24 to 48 hConversion and oxidation go on after the sheet leaves the press, so the colour at the delivery is not the colour the customer will see.
Dose above what the grade needsPast the cure window the extra energy goes into initiator fragments and thermal load rather than into conversion, and colour rises with it.
Aged mercury lamp with shifted outputAn ageing lamp gives proportionally less short-wave and more long-wave and infra-red, the operator raises power to recover cure, and heat and initiator fragmentation both climb.
Coat weight above the windowAbsorbance scales with film thickness, so the same grade reads yellower simply for being laid heavier.

UV Curable CoatingsYellowing on white board and light colours

Fault 16 of 20 · continued

Yellowing on white board and light colours

CHECK, IN THIS ORDER

Measure CIELAB b* on the coated and the uncoated area of the same sheet with a spectrophotometer, D50 illuminant and 2 degree observer, at 1 h, 24 h and 7 days, and report delta b* rather than a visual opinion
Radiometer pass for delivered dose, looking for over-dose as readily as under-dose, and read lamp hours against the last logged figure
Gravimetric coat weight
Hold one coated sheet in the dark and one in daylight for 7 days and remeasure, since part of the colour bleaches back
Confirm the grade in the pan and whether the job called for a low-yellowing grade on white board

THEN CHANGE

Bring dose back to the specified window, confirming the MEK rub count still passes
Replace an aged lamp on measured output and clean or replace dichroic reflectors that are passing heat into the sheet
Reduce coat weight to the bottom of the window
Move to the low-yellowing grade for white board and pastel work, and to a TPO-free or amine-free formulation where the specification asks for one. These are formulation statements; conformity of the finished pack is assessed on the customer's own construction
Cool the pile below 35 C and leave it unwrapped overnight before it is judged or despatched

CONFIRM IT IS FIXED

Delta b* between the coated and uncoated area inside the figure agreed with the customer, measured at 24 h and again at 7 days on sheets from head, middle and tail, with the film still passing the agreed MEK rub count.

THE TRAP

Judging the colour at the delivery and signing the job off. A UV coating yellows for a day or two after it cures, so a pile that matched at the press can be outside tolerance when the customer opens it, and the whole order is coated by then.

UV Curable CoatingsYellowing on white board and light colours (continued)

Fault 17 of 20

Odour or taint in the coated stock

WHAT YOU SEE

The pallet smells when the wrap comes off, or the customer reports taint in a food or tobacco pack. The coating passes a rub test and looks fully cured.

WHAT IT LOOKS LIKE

CAUSES, MOST LIKELY FIRST

Under-cureOdour in a UV film is almost always residual unreacted monomer and photoinitiator cleavage products, and residual monomer means incomplete conversion.
Standard grade used where a low-odour or low-migration grade was specifiedThe raw material set is what sets the odour floor; a standard coating has no route to a low-odour result at any dose.
Coat weight above the windowMore film at the same dose means lower conversion through the depth, and more residual material to give off odour.
A hot pile wrapped before it cooledWrapping seals residual monomer against the stock and drives it into the uncoated reverse, which on a carton is usually the food-contact face.
A standard coating over a low-migration inkA low-migration ink underneath is cancelled by a standard overprint coating on top, which reintroduces everything the ink was chosen to exclude.

UV Curable CoatingsOdour or taint in the coated stock

Fault 17 of 20 · continued

Odour or taint in the coated stock

CHECK, IN THIS ORDER

Radiometer pass for delivered dose at running speed, plus MEK double rubs after ASTM D5402 at the heaviest film on the sheet
Sensory test to EN 1230-1 for odour and EN 1230-2 for off-flavour, the Robinson-type transfer test, at a defined sample area to headspace ratio in clean glass, typically 24 h at 40 C, scored 0 to 4 by a trained panel against an uncoated control
Gravimetric coat weight
Probe the pile temperature at the point the pallet was wrapped, and examine the uncoated reverse of sheets from the middle of the pile for transfer
Confirm the grade in use, and confirm that every layer on the pack, ink, coating and any adhesive, is the low-migration variant where one is specified

THEN CHANGE

Raise delivered dose to the top of the grade window, cleaning optics and replacing an aged lamp first
Move to the specified grade: the UniCure-2010RC class low-migration low-odour coating, the UniCure-24GR class low-odour gravure coating, or the UniCure Gloss LOC 82RC class for narrow web
Reduce coat weight to the window
Cool the pile below 35 C before wrapping, and leave the pallet unwrapped and ventilated overnight
Specify the low-migration coating wherever a low-migration ink is used, since the two are one system

CONFIRM IT IS FIXED

A sensory result to EN 1230-1 and EN 1230-2 scored by a trained panel against an uncoated control and meeting the score agreed with the customer, typically 1 or better on the 0 to 4 scale, with the MEK rub count passing and delivered dose logged inside the grade window. Raw materials are selected to the EuPIA Charter for the low-migration grades; migration on the finished pack is assessed separately, on request, on the customer's own construction.

THE TRAP

Wrapping the pallet tightly to keep the smell in until despatch. Wrapping a warm pile seals residual monomer against the stock, and it transfers into the uncoated reverse of the sheet, which on most cartons is the face that ends up against the product.

UV Curable CoatingsOdour or taint in the coated stock (continued)

Fault 18 of 20

Glue, foil or laminate will not hold on the coating

WHAT YOU SEE

Glue flaps pop open at the folder-gluer or a week later in the box, hot foil does not take, or a laminate delaminates from the coated face. The coating itself is faultless.

WHAT IT LOOKS LIKE

CAUSES, MOST LIKELY FIRST

Full-coverage coating over the glue lapA UV coating is a closed, low-energy, non-absorbent surface by design; a water-based glue meant to key into board fibre cannot wet it and cannot reach the fibre.
Slip or silicone added at the pressAny surface-active addition lowers the surface tension of the whole film, which is exactly what an adhesive, a foil size and a laminating adhesive all need to overcome.
Wrong grade for the downstream operationA release coating, a high-slip coating and a stampable coating are three different products, and only one of them is meant to be glued or foiled.
Over-cureA harder, fully converted film is slightly less polar and offers a little less for an adhesive to key into, but this is a second-order lever. Carton glue is overwhelmingly water-based PVA or dextrin and it is not reacting with the coating at all: the bond comes from wetting, water absorption and penetration into the board fibre, which is why fibre tear is the pass criterion.
Contamination on the coated surfacePowder, dust or oil at the coated face defeats a bond that the coating alone would have supported.

UV Curable CoatingsGlue, foil or laminate will not hold on the coating

Fault 18 of 20 · continued

Glue, foil or laminate will not hold on the coating

CHECK, IN THIS ORDER

Check the coating plate, cylinder or blanket to confirm whether the glue lap and the foiling area are knocked out of the coating image
Identify the exact grade in the pan against what the job specified, and check whether anything was added during the run
Glue-bond test on a made-up flap assessed for fibre tear, with the converter's own adhesive and board
Where a laminate is involved, measure the bond to ASTM F904 or a T-peel to ASTM D1876 on 15 or 25 mm strips at 300 mm/min, reported in N/15 mm with the mode of failure stated. Where a tape peel is what is meant, say so and use ASTM D3330 on the tape, which is a different measurement
Dyne pens or ISO 8296 test inks on the coated surface where the glue or foil has to hold

THEN CHANGE

Knock the glue lap and the foiling area out of the coating image; this is the fix that works and the one most often skipped
Remove press-added slip and requalify from a clean circuit
Move to a stampable or glueable grade, the UniCure Gloss CSC 84RC class for stampable narrow-web work, where full coverage is unavoidable
Where foil is to be applied over the coating, specify the coating and the foil together and trial them before the run
Where full coverage is mandatory, qualify the customer's own adhesive on the coated surface before the job, not after it

CONFIRM IT IS FIXED

Fibre tear on the glue flap across ten made-up cartons at 24 h and again at 7 days, or a laminate bond to ASTM F904 or ASTM D1876 agreed with the converter on their own adhesive and their own board. This is a construction property and is qualified on the customer's material, not claimed for the coating.

THE TRAP

Asking for a stronger or hotter glue first. A faster-setting adhesive gives a bond that passes at the folder-gluer and opens in the packed box a week later, because the problem is the surface it is sticking to, not the strength of the adhesive; and by then the whole order is glued.

UV Curable CoatingsGlue, foil or laminate will not hold on the coating (continued)

Fault 19 of 20

Screen coating shows bubbles, pinholes or poor lay

WHAT YOU SEE

Small burst bubbles and pinholes through the coating, or a surface that looks textured rather than glass-smooth. It is most visible over dark and metallic grounds, which is where screen coatings usually run.

WHAT IT LOOKS LIKE

CAUSES, MOST LIKELY FIRST

Air worked into the coating before printingScreen coatings are viscous and hold air for a long time; stirring, pumping or pouring from a height puts bubbles in that will not come out during the print stroke.
Squeegee speed too high, or wrong durometerA fast stroke or a hard blade shears air into the film and gives it no time to flow out of the mesh marks before it is cured.
Coating too cold and too viscousBelow the grade temperature the film will not level out of the mesh pattern inside the available open time.
Mesh and stencil wrong for the gradeToo coarse a mesh lays discrete threads of coating that must coalesce; too fine a mesh starves the film and leaves pinholes.
Off-contact too greatA high snap-off stretches and snaps the film off the mesh, which entrains air along the whole stroke.

UV Curable CoatingsScreen coating shows bubbles, pinholes or poor lay

Fault 19 of 20 · continued

Screen coating shows bubbles, pinholes or poor lay

CHECK, IN THIS ORDER

Stand a sample in a clear beaker for 10 min and count the bubbles rising, to see how much air is already in the coating
Measure coating temperature at the screen and compare with the grade figure, and check viscosity on a rotational viscometer with spindle, speed and temperature stated, since a screen grade is outside the range of any flow cup
Check mesh count and squeegee durometer with a Shore A gauge against what the grade specifies
Examine a pinhole at 30x: a mesh mark around it points at the stencil and the stroke, a clean round hole at entrained air
Measure off-contact across the frame

THEN CHANGE

Let the coating stand to de-aerate at 25 C before printing, and pour rather than stir
Slow the squeegee and fit the specified durometer blade
Bring the coating to 25 C, warming it rather than thinning it
Change to the mesh the grade specifies, and use the grade formulated for high clarity, good lay and a bubble-free surface rather than pushing a general coating through a screen
Reduce off-contact to the minimum that still gives clean snap-off

CONFIRM IT IS FIXED

Bubble-free surface at 30x on ten consecutive prints, with gloss to ISO 2813 inside the grade window and through cure confirmed by the MEK rub count at the heaviest film, which on a screen flood gloss is 12 to 20 g/m2 and needs a dose to match.

THE TRAP

Stirring the coating harder to get the bubbles out. Stirring is how the air got in; a screen coating needs to be left to stand, and a batch that has just been beaten with a paddle will print worse than the one that was left alone.

UV Curable CoatingsScreen coating shows bubbles, pinholes or poor lay (continued)

Fault 20 of 20

Raised or tactile effect loses height and edge

WHAT YOU SEE

The crystal, haptic or tactile effect feels flat to the finger and the edges of the image have rounded off. Gloss looks fine, the relief simply is not there, and the body of a thick film may also be soft under a hard skin.

WHAT IT LOOKS LIKE

CAUSES, MOST LIKELY FIRST

Film weight short at the screenThe whole effect is film thickness; stencil build, mesh and stroke set how much material is laid, and 25 to 40 g/m2 is the working range for a tactile, crystal or haptic film. There is no other source of height.
Coating too warm, so the relief flows before it is pinnedA low-viscosity film levels, and levelling is the enemy of relief; the shape must be frozen before surface tension flattens it.
Too long between the screen and the first emitterEvery extra second of open time is time the raised edge spends slumping.
Peak irradiance too low to skin a heavy film quicklyIrradiance sets the shape, because it decides how fast the surface gels. Dose delivered slowly cures the film eventually but not fast enough to hold an edge.
Total dose short for the film weight, which is a separate failure from the shapeA 25 to 40 g/m2 film needs 400 to 1200 mJ/cm2 to cure through. Pinning the surface hard and leaving the total dose short holds the shape over a soft body, which then fails at the fingernail and on odour.

UV Curable CoatingsRaised or tactile effect loses height and edge

Fault 20 of 20 · continued

Raised or tactile effect loses height and edge

CHECK, IN THIS ORDER

Measure relief height with a dial gauge or a calibrated microscope on a cross-section at five points across the image
Measure coating temperature at the screen and check viscosity on a rotational viscometer with spindle, speed and temperature stated
Check stencil build thickness and mesh against the grade specification
Radiometer pass for peak irradiance and for total dose separately, since the first sets the shape and the second sets the through cure
MEK double rubs after ASTM D5402 at the thickest section, and scrape the edge of the relief with a fingernail through a glove to find a soft body under a hard skin

THEN CHANGE

Build the stencil to the thickness the effect needs and run the mesh the grade specifies
Cool the coating towards the lower end of its working range so it holds its shape
Freeze the shape first: high peak irradiance immediately after the screen, at a pinning station or the leading emitter, so surface tension has no time to level the relief
Then deliver the full through-cure dose, 400 to 1200 mJ/cm2 depending on film weight, at the following stations. Raising irradiance while leaving total dose short is the common half-fix and it produces a good-looking effect that fails later
Use the grade built for the effect: the crystal, haptic, bubble and tactile screen grades in the UniCure-66SR to -70SR block each produce a different texture and are not interchangeable

CONFIRM IT IS FIXED

Relief height within the figure agreed against the approved sample at five points across the image, a sharp edge at 30x, and through cure at the thickest section by the agreed MEK rub count with a band-matched dose reading logged.

THE TRAP

Putting a second pass over the first to build the height back. A second film over an already cured first film gives a visible step and an interlayer that separates under the fingernail, instead of the single thick film the effect is designed around.

UV Curable CoatingsRaised or tactile effect loses height and edge (continued)

At the press

One page to work from, and the methods behind every number.

Quick reference

One page at the press

Symptom	Check this first	Change this first
Gloss down or patchy	Gravimetric coat weight on marked sheets, and pan temperature	Weight back into the window for that unit, never by thinning
Surface tacky, body hard	Band-matched radiometer at running speed, edges as well as centre, and the state of the optics	Clean reflectors and windows with the machine isolated, then raise lamp power
Coat weight drifting down	Anilox or gravure cells at 100x and the blade edge	Clean the roll; do not add blade pressure
Blocking or set-off in the pile or reel	Pile centre temperature within 5 min of delivery	Cool the delivery and cut the pile height
Fisheyes, craters and specks	Defect centre at 30x: particle, clear or empty	Remove the silicone, powder or dust source, clean the circuit
Orange peel or ribbing	Pattern direction and pitch against roller circumference, then pan viscosity and temperature	Warm the coating to grade temperature; check for a swollen non-UV roller
Coating crawls off metallic ink	Dyne value on the printed metallic, not the substrate	Change to the coating matched to that metallic
Cracks on the crease	Coat weight, then dose against the upper limit	Reduce coat weight before touching the lamps

UV Curable CoatingsQuick reference

Quick reference

One page at the press

Symptom	Check this first	Change this first
Matt rubs and burnishes	MEK rubs first, then gloss to identify which matt level is actually on the sheet	Move to a high-rub matt grade
Coating lifts on met-PET or BOPP	Dyne pens on the reel about to run, target above 40 mN/m	Switch to the matched grade, and corona in line where the line has it
Coating strips over conventional ink	Tape pull over the solid against the unprinted margin at 24 h	Let the ink dry through, or prime, before coating
Cartons jam, slip wrong	Coefficient of friction at 24 h by the method agreed with the converter	Change grade to the specified friction, add nothing
Curl on low-grammage stock	Corner lift on a 100 mm square at 1 h and 24 h, noting direction	Reduce coat weight to the bottom of the window
Coating gelling in the pan	Light leak onto the pan, checked with lamps off and strips left in place, and pan temperature	Shield and cool; take the seeded batch out vented and labelled
Drip-off contrast weak	Is any interdeck lamp on between the two units	Switch it off and set both weights separately
Yellowing on white board	Delta b* coated against uncoated at 24 h, and dose against the upper limit	Bring dose down to the window and check lamp age

UV Curable CoatingsQuick reference

Quick reference

One page at the press

Symptom	Check this first	Change this first
Odour or taint	Dose at running speed and MEK rubs at the heaviest film	Raise dose and cool the pile before wrapping
Glue, foil or laminate fails	Is the lap knocked out of the coating image	Knock the lap out; do not change the adhesive first
Screen bubbles and poor lay	Bubbles rising in a beaker sample, and coating temperature at the screen	Let it stand to de-aerate; stop stirring it
Tactile relief flat or soft	Peak irradiance and total dose read separately	Pin the shape with irradiance, then give it the full dose

UV Curable CoatingsQuick reference

Test methods

The measurements this guide relies on

Measurement	Method	What it tells you, and the usual acceptance
Flow cup viscosity	ISO 2431 with an ISO cup, or ASTM D1200 for a Ford cup	Tells you what the coater will actually meter. ISO 2431 uses 3, 4, 5 and 6 mm ISO cups at 23 +/- 0.5 C and supersedes DIN 53211; a Ford cup is ASTM D1200 and is normally run at the house temperature, here 25 +/- 1 C. The three scales are not interchangeable, so quote cup, standard and temperature with every number. Sample and cup both at temperature, drawn from the pan and not from the drum. Typical acceptance on a Ford No. 4 at 25 C: roller gloss 45 +/- 5 s, high-porosity gloss 65 +/- 5 s, metallised-PET gloss 55 +/- 5 s, gravure OPV 35 +/- 5 s.
Rotational viscosity for screen and tactile grades	Rotational viscometer, spindle, speed and temperature stated	Screen, crystal, haptic and tactile grades are far too viscous for any flow cup, so they are controlled on a rotational instrument. A reading means nothing without the spindle, the speed and the temperature beside it, and these coatings are shear thinning, so compare like with like.
Coat weight, gravimetric	No standard number; weigh and divide	The only honest measure of what is on the sheet. Mark and weigh five sheets to 0.01 g, feed them through the coater, weigh them again within a few minutes of cure, and divide the mean gain by the sheet area. Where marked sheets cannot be fed, weigh twenty coated and twenty uncoated sheets from the same pallet and report the mean with its spread, stating that substrate variation of 2 to 3 per cent, not the coating, sets the uncertainty. On web, punch samples of known area from the same reel position coated and uncoated. Class indications: narrow-web flexo and gravure OPV 1.5 to 3.0 g/m2, sheet-fed chambered anilox 3.0 to 5.0 g/m2, roller gloss 4 to 6 g/m2 on coated board and 6 to 10 g/m2 on porous board, matt 2.0 to 3.0 g/m2, screen flood gloss 12 to 20 g/m2, screen tactile 25 to 40 g/m2.
Delivered UV dose and peak irradiance	Band-matched radiometer; no single test standard cited	The energy the film actually receives, and how fast it receives it. Radiometer carried through the press at running speed and read on the coated area, at both edges of the width as well as the centre. Match the band to the emitter: UV-A, about 320 to 390 nm, for mercury and for 365 and 385 nm arrays; UV-V, about 395 to 445 nm, for 395 and 405 nm arrays; or an instrument with an LED-specific calibration. Typical acceptance for this class: 80 to 150 mJ/cm2 thin flexo or gravure OPV, 100 to 200 mJ/cm2 roller gloss, 200 to 400 mJ/cm2 heavy matt or heavy roller gloss, 400 to 1200 mJ/cm2 screen tactile, crystal or haptic. Peak irradiance at the substrate: mercury 1.5 to 4 W/cm2, LED 4 to 16 W/cm2, in the band measured. Record instrument, band and calibration date, and never judge one instrument against a window set on another.
Specular gloss	ISO 2813	The number the customer is actually buying. 60 degrees is the reference geometry; add 20 degrees above about 70 GU and 85 degrees below about 10 GU, as ISO 2813 directs, because 60 degrees stops discriminating at both ends. Read at five points across the sheet and at head, middle and tail, logging spread as well as mean, and report the geometry with every figure. Typical acceptance: above 90 GU at 60 degrees for a standard roller gloss, above 95 GU on metallised-PET board, 40 to 60 GU satin, 10 to 20 GU deep matt.
Solvent rub cure check	After the practice of ASTM D5402, with the house conditions stated	Whether the film is cured through. ASTM D5402 defines the practice but not the load and not a pass level, so state the house conditions: MEK on a cotton pad under a nominal 1 kg load, counting double rubs to first breakthrough, gloved. Typical acceptance for a cured UV coating is 50 double rubs or more with no breakthrough; 30 or fewer is a marginal cure to be investigated, not a pass. Substrate, coat weight and the age of the sample all move the count, so compare like with like. It reports through cure only, so pair it with a gloved thumb twist for surface cure.
Surface energy	ASTM D2578, or ISO 8296 test inks, or calibrated dyne pens	Whether the coating can wet what it is going onto. Measured on the reel or pile about to run and on the printed area, never taken from the film certificate. Typical acceptance: above 40 mN/m on BOPP, LDPE and CC-PET, above 38 mN/m as a floor.

UV Curable CoatingsTest methods

Test methods

The measurements this guide relies on

Measurement	Method	What it tells you, and the usual acceptance
Coefficient of friction	ASTM D1894 horizontal sled, or TAPPI T 815 inclined plane for board	Whether cartons feed and whether pallets stay together. ASTM D1894 is a horizontal sled method written for plastic film and sheeting and is used on coated board where the board can be mounted flat; a converter's laboratory may be running an inclined plane method instead, which does not give the same figure. For packaging board the inclined plane method to name is TAPPI T 815; ASTM D4918 is written for uncoated writing and printing paper and does not cover a coated board, so do not quote it against a coated carton. Agree the method, the mating faces, the temperature and humidity and the acceptance band before the run. Static and kinetic, coated face to coated face and coated face to uncoated reverse, at 23 C and 50 % RH, 24 h after coating. A high-slip gravure OPV sits below 0.2.
Adhesion, tape pull and cross-cut	ISO 2409 cross-cut, with a straight tape pull as the alternative	Whether the coating is holding to the printed surface. Cross-cut only where coating and substrate are thick enough that the blade does not cut through; on thin film use a straight tape pull across a solid, removed at a stated angle and rate, and report percentage retained. Pull over a printed solid and over an unprinted margin on the same sheet, because the difference tells you whether the fault is the ink or the coating. Test at 24 h and again at 7 days.
Laminate and ply bond	ASTM F904, or T-peel to ASTM D1876	Whether a laminate is holding to the coated face. Not the same measurement as a tape peel: ASTM D3330 is the peel adhesion of pressure-sensitive tape and says nothing about a ply bond. Pull 15 or 25 mm strips at 300 mm/min, report in N/15 mm, and state the mode of failure, because where the construction fails matters more than the number.
Rub resistance	Sutherland-type rub tester, after the practice of ASTM D5264, with the house conditions stated	How the finish survives handling. The practice covers the instrument and the procedure, not the block weight and not the cycle count, so the conditions travel with every result: 4 lb block, 100 cycles, print to print, at 23 C and 50 % RH, assessed for transfer to the facing sheet and for gloss change on the rubbed area. A cycle count means nothing without them, and two laboratories must be on the same conditions and the same mating faces before a rub result can be argued. Test 24 h after coating and repeat at 7 days, since a UV film goes on hardening.
Block resistance	Described rather than cited	Whether the pile will stick. Two coated faces, or a coated face against the uncoated reverse, in contact under a stated load, typically 1 kg/cm2 as an accelerated condition, held at 40 C for 24 h, then peeled and examined for transfer and gloss change. A 1 m pile of carton board applies only about 0.1 kg/cm2, so this is a severe screening test and not a simulation of the pile; where a marginal result has to be judged, repeat it at the pressure and temperature the actual pile sees.
Odour and taint	EN 1230-1 odour, EN 1230-2 off-flavour, the Robinson-type transfer test	Whether the coated stock will be noticed in a food or tobacco pack. Coated sample and an uncoated control at a defined sample area to headspace ratio, sealed in clean glass, conditioned typically 24 h at 40 C, scored 0 to 4 by a trained panel against the control and, for taint, against a food simulant. Acceptance is the score agreed with the customer, commonly 1 or better. A coated board is rarely identical to bare board, so identity with the control is not an acceptance criterion.
Yellowing	Spectrophotometer, described rather than cited	Whether the coating is adding colour to white board. Measure CIELAB b* on the coated and the uncoated area of the same sheet, D50 illuminant and 2 degree observer, and report delta b* at 1 h, 24 h and 7 days. The 24 h figure is the one to judge on, because a UV film yellows further after it leaves the press and part of that colour then bleaches back in daylight.

UV Curable CoatingsTest methods

When the press-floor fixes have not worked

What to send us

A coated sample and an uncoated control from the same pile or reel, at least ten sheets or two metres, taken from the head, middle and tail of the run and marked as such, plus the approved sample the job is being judged against.
The radiometer log: delivered dose and peak irradiance at full running speed, the instrument, the head band and its calibration date, readings at both edges and the centre of the width, lamp or LED hours since the last change, and the reading taken when they were new.
The coating batch numbers, the drum dates and how the drums were stored, the ink batch numbers underneath and whether the ink is UV, conventional oxidative or low-migration, the grade actually in the pan, and anything added at the press with the quantity and the time it went in.
The gravimetric coat weight log for the run, taken on marked sheets before and after coating, with pan viscosity and the cup, standard and temperature it was measured on, pan and return temperature, and press hall temperature at the same times.
The substrate: an unprinted, unopened piece of the same reel or pallet with the supplier, the reel number and the treatment date, plus the dyne reading you measured on it yourself.
The coater unit itself where a pattern, a viscosity drift or an unexplained contamination is the complaint: the specification of the roller coverings, blanket, hoses, gaskets and chamber end seals, and whether the unit was converted from water-based.
The full downstream route the job takes, with the machines, speeds and temperatures it sees at folding, gluing, foiling, laminating and filling, and the adhesive, foil or film it meets there, since bond, seal and fold performance belong to the construction and cannot be assessed from the coating alone.
For an odour, taint or migration question, the finished pack construction layer by layer, the product it contains, the fill temperature and the intended shelf life, so a sensory and migration assessment can be set up on the finished pack rather than argued from the coating specification.

Technical servicesales@ufindustries.com | +91 8750051884 | E105, GNEPIP, Industrial Area Site-5, Greater Noida, Uttar Pradesh 201306, India

UV Curable CoatingsTechnical service