How to Evaluate a UV Printing Machine: A Production Acceptance Checklist

Why a demonstration sample does not prove production performance

A demonstration is designed to show a machine at its best. The file has been prepared and proofed, the operator knows the machine well, the item is often a clean sample, and only a few pieces are printed. None of that is dishonest, but it answers a narrower question than the one a buyer has to answer.

Production adds the variables a demonstration leaves out: material from several batches, the hardest elements of your artwork, shift-length runs with loading and cleaning, operators with different experience, and the handling and exposure your product meets after printing. Acceptance testing brings those variables into the evaluation before money and responsibility change hands.

Not every purchase needs a formal factory acceptance test. For a single machine, a thorough print trial followed by a site acceptance test is often enough. What matters is that the criteria are written down and measured, at whichever stage they are checked.

Step 1: Define the application in writing

Acceptance is only as clear as the specification it is measured against. Before any trial, write a short application specification that both buyer and supplier sign. It turns "print our bottles" into a description that a test can be judged against. In this guide, "supplier" means the machine supplier; the company that makes your bottles, panels or coatings is the "material supplier".

Be honest about what is still unknown. If the perfume formula or the bottle coating has not been finalised, the trial result applies only to the version tested. Where adhesion is uncertain, start with the untreated part and add one pre-treatment at a time, so that the route which passes is the one written into the specification.

Step 2: Prepare a representative trial

A trial is representative when it uses what production will use. The buyer supplies the materials from normal production stock rather than hand-picked samples, and the supplier prints them in the configuration that would be delivered.

• Materials: supply enough pieces for the timed run, every durability test and the approved samples, from at least two batches or deliveries where possible, including every coating colour or finish that will be printed. If only one batch exists, record that the result is limited to it. Print items at the temperature they will have in production and record it, because cold or filled items brought into a warm room can carry condensation you cannot see.
• Artwork: use real production files plus a test file with the smallest positive and reversed text, the finest lines, a smooth gradient, large solids, brand colours, any white or varnish layers you need, and a solid patch large enough for an adhesion test.
• Configuration: print with the ink set, primer or pre-treatment, print mode, curing set-up and fixtures that would be supplied, not a higher-specification demonstration build. If the trial uses pre-treatment equipment you will not have, record that as a limit of the result. Flame and plasma treatment and some primers lose effect with time, so record the interval between pre-treatment and printing and keep within the window the primer or pre-treatment maker gives.
• Fixtures: use the production fixture, or record that a temporary one was used. Check that every position holds the item at the same place and height, because height differences change the print distance and sharpness and can bring an item close enough to strike the print heads. Ask who makes each new fixture, how long it takes and whether it is in the quotation.
• Quantity: agree a timed run long enough to include reloading, at least one planned stop and restart, and samples from each material batch. A handful of pieces cannot show rejects or drift.
• People: note who operates the machine. If your own staff will run production, ask for part of the run to be loaded and inspected by someone with comparable experience.
• Criteria first: write the acceptance criteria, test methods and number of samples per test before printing starts, and agree who judges borderline results.

Round and tapered items add their own checks, such as how the wrap closes and whether colour holds across container sizes; see what to test on a cylindrical printer. A wall printer works on site, so its acceptance happens on site too: ask for a timed test print on a wall like yours.

Step 3: Measure accepted output, not rated speed

Accepted output per hour is the number of pieces that pass inspection in a shift, divided by the length of the shift in hours. It is the figure to compare between machines, because rated speed is measured under the manufacturer's conditions and makes no allowance for your loading, inspection, cleaning or rejects.

Manufacturers rate speed under stated conditions. UV flatbed printers are usually rated in square metres per hour for a print mode, which says little about pieces per hour until it is timed on your fixture, because the carriage also travels over the gaps between items. Cylindrical printers are rated in pieces per minute or hour: LSINC, for example, gives the speeds of its PeriOne, PeriQ360 and Perivallo360m as maximums for 4 inch (101.6 mm) artwork, and says speed depends on image height, resolution and print mode. Hourly figures like these assume uninterrupted printing, with no stated allowance for cleaning, stoppages or rejects. They are useful for a first comparison only.

Count each minute once. Some activities stop the machine and add to every cycle, some come off the available printing time, and some happen while the machine prints.

• In every cycle (the machine waits): loading and unloading on a single fixture, and swapping fixtures. Every pass the artwork needs, including white, varnish and texture, belongs in the timed print cycle itself.
• Off the available time: start-up, nozzle checks, cleaning and artwork changeovers. Put each activity either in the cycle or in this list, never both.
• In parallel (the machine keeps printing): loading a second fixture while the first prints, inspecting and packing finished pieces and preparing the next batch, provided one person can do the work in the time available.
• Check the overlap: if the operator who loads also inspects, add both tasks together and compare them with the print cycle. If they take longer, the operator sets the pace, not the machine.

Hypothetical example: flat-sided bottles on a UV flatbed printer

In this hypothetical example, the same machine, artwork and single operator give about 75 or about 112 accepted bottles an hour, against a print-only rate of 160. The only difference is a second fixture that lets the operator load while the machine prints, and the gain depends on the swap time: with a 3 minute swap, the two-fixture result falls to about 92. The example assumes the 9 minute print includes every layer, that the first fixture is loaded during start-up and the last unloaded during cleaning. It leaves out things a real trial should time: artwork changeovers, pre-treatment, packing, ink top-ups and longer stops.

A trial run is usually shorter than a shift. Time the cycles, log every stop and count rejects during the trial, calculate the shift figure with the agreed allowances, and confirm it at site acceptance with a full-shift output run. Treat a reject rate from a short run as provisional: 4 rejects in 100 pieces could come from a real rate anywhere between about 1 and 10 percent.

Rejects also cost material and machine time. At 4 percent, an order for 1,000 good bottles needs about 1,042 printed, and reprinting a few rejects on a part-filled fixture can take almost a full cycle. Record ink use per piece as well: weigh a full fixture of pieces before and after printing, and measure refills over the run, because refills also include the ink used for purges and cleaning. Feed the measured figures into a cost model rather than a rated-speed calculator; our bottlers' guide sets out a cost-of-ownership framework.

Step 4: Judge print quality against an approved sample

An approved sample is a printed piece of your own product, signed by buyer and supplier, that sets the standard for colour, finish and defects. Agree in the test plan how it will be made, for example from the first pieces printed in the trial, and sign it before the timed run starts. Print quality is easiest to accept when every criterion refers to a measurement or to that sample. Judge pieces under controlled lighting, because colour and defects look different under warehouse lights, daylight and a viewing booth. ISO 3664 sets out viewing conditions for comparing prints.

Colour needs particular care, because we have found no ISO standard that sets a colour tolerance for UV inkjet printing on finished objects. ISO 13655 defines how print colour is measured, and ISO/CIE 11664-6 defines the CIEDE2000 colour-difference formula, often written ΔE00 or Delta E 2000. ISO/TS 15311-2, which covers print quality for commercial digital printing on flat sheets, states that it does not set aims or tolerances, because they differ widely between applications. The figure you write is therefore a commercial agreement between buyer and supplier. Where your customer or brand owner already sets a colour tolerance for your current decoration, start from that, and confirm at the trial that the machine holds it on your item.

On a curved item, use an instrument with an aperture small enough to sit flat on the patch, and measure at the same marked position each time. Where that is not possible, measure patches printed in the same mode on a flatter piece of the same material and coating, record that they are a substitute, and still judge the item itself against the approved sample. Record the measured values of the approved sample when it is signed, because a physical sample can change with time. If the colour reference was made by another process, such as offset or screen printing, compare under more than one light source, because two prints can match under one light and differ under another.

For tactile and Braille signs, measure dot height and spacing rather than judging by touch, and check them against the code that applies to the project. Our Braille and tactile signage guide explains which rules apply in Dubai.

Step 5: Test each durability property separately

Durability is several properties, not one. A print can pass a cross-cut adhesion test and still rub off in transit, soften under perfume or fail in a dishwasher. Each exposure in your specification needs its own test, its own samples and its own pass criterion.

Three rules apply to every test. First, do not test straight off the machine, because adhesion and resistance keep developing after printing. PRO TECHnology's practice is to leave printed parts 24 hours before testing; if the ink's technical data sheet gives a longer time, use that. Waiting does not make up for too little UV light at the lamp, and if production fills, packs or stacks parts sooner, test at that interval as well. Second, record the cure-to-test time, the room conditions and the exact test conditions with the result. Third, use your own product and handling wherever the specification names them, not a generic substitute.

For the cross-cut, test the solid patch in the test file, printed with the same layers as the artwork, because the cut spacing depends on the total film thickness and primer, white and varnish all add to it. Record whether the test was on the item or on a flat panel, and where any failure occurred: at the substrate, at the primer or between ink layers. The result is a classification, not a measured bond strength.

Pre-treatment can improve adhesion, but it does not by itself give chemical resistance. Resistance to perfume or alcohol also depends on the ink chemistry, the cure, the film thickness, any protective varnish and the product formula, so run a rub test with your own fragrance rather than relying on alcohol alone.

If a print fails, find the cause before changing the machine or its settings. Common causes are too little cure; too much cure, which can make the film brittle; an ink or varnish that does not resist the test solvent even when cured; a layer too thick to cure through; an ink that does not suit the material; a contaminated, untreated or wrongly treated surface; a material supplier's coating that the ink does not bond to; and testing too soon. How a print fails is a clue. If it softens, smears or leaves colour on the cloth, look first at the cure and the ink. If it lifts away as a film or flakes at the edges, look first at the surface, the pre-treatment and the coating. Our guide to UV printing problems by material helps separate the causes.

Wording matters in the record. A workshop check based on ISO 2409 is useful evidence, but it is not a test to ISO 2409 unless the full procedure, equipment and conditions in the standard were followed. If your customers or regulators need formal results, commission an accredited laboratory and attach its report to the acceptance record.

Ink data sheets are not acceptance criteria. Marabu, for example, says the information in its UltraJet DUV-CP data sheet "should not be used for specification purposes". The test on your own item is what decides. Step-by-step procedures for the cross-cut, alcohol rub and soak tests are in our UV printing adhesion guide.

PRO TECHnology does not rate or promise UV print as dishwasher safe or microwave safe. Where washing matters, run real wash cycles on a sample of your own item, and record the outcome as a test result for that item, not as a rating.

Step 6: Prove repeatability across runs, restarts and batches

Repeatability means the machine gives the same accepted result across a long run, after a stop, on a new material batch and with a different operator. Test it by taking samples at fixed points and comparing each one with the approved sample.

• Within a run: keep the first, a middle and the last pieces of the timed run, from each fixture position or spindle, and check colour, white opacity, registration and defects on each. Print the machine's nozzle check pattern at the start and end of the run and keep it with the samples.
• After a stop: stop the machine as production would, for a break, a cleaning cycle or overnight, then restart with only the routine start-up procedure in the manual and check the first pieces again without further adjustment. Check white in particular, because white pigment settles when ink stands.
• Between material batches: print every batch and coating colour supplied, because a change in the material supplier's coating can change adhesion.
• Between operators and shifts: have a second person run the job from the written recipe (the settings, materials and fixture that passed), without help from the person who set it up.

Curing settings alone do not show how much UV light reaches the print. Lamp output falls with age and with ink mist on the lamp window, and less light reaches surfaces further from the lamp, such as the sides of a curved item or a recess. Where the machine allows it, measure UV intensity and dose at the print position with a radiometer suited to the lamp type, because an instrument made for mercury lamps can misread LED output. Record the readings at the trial and at site acceptance as the baseline for later checks; where a radiometer cannot be placed, agree another cure check, such as a solvent rub on a test patch.

Then freeze the recipe that passed: artwork version, colour profile, print mode, ink and primer, pre-treatment settings, curing settings and fixture. Keep a signed set of approved samples, labelled with the date, material batch and settings and stored away from light. Run a quick adhesion check on each new batch, and keep the tested piece with the job record, as our guide to UV printing on anodised aluminium describes.

Agree in advance which changes trigger a new test: a new material supplier, substrate, coating or finish, a new ink or primer, a new type of artwork, changed print or curing settings, a firmware or RIP (raster image processor) software update, a new colour profile, and major service work such as a print head or UV lamp replacement. After head work, check head alignment and colour before production restarts.

Step 7: Document what the machine needs to run

Acceptance should confirm what the machine needs to keep printing, not only what it prints. Most of these points come from documents the supplier can provide before purchase, so ask for them early and check them against your site.

LSINC's site requirements for its PeriOne and Perivallo360m printers, summarised in our cylindrical UV printing guide, are a useful example of what a site sheet covers. They list power, compressed air, temperature and humidity ranges, clearances, venting of fumes to the outside and an internet connection for remote service. Figures can change between documents: LSINC's 2024 PeriOne brochure gives the supply as 230 VAC, while its 2026 documents give 200 to 240 VAC. Always work from the current document for your exact model.

Step 8: Sign the acceptance record and check the handover file

Before signing, repeat the key checks at site acceptance, even if the machine passed at the supplier's facility. Transport, local power, temperature, humidity and new operators can all change the result, the delivered unit may not be the machine the trial ran on, and the site run is the first one carried out entirely under your own conditions.

The acceptance record is the document both parties sign at the end. It lists the configuration, each criterion and method, the results and any open items, so that someone who was not present can see what was tested, on what, and with what result. If the purchase contract links a payment or the start of the warranty to acceptance, make sure the contract and the acceptance record use the same criteria.

What the acceptance record should contain

• Configuration: machine model and serial number, firmware and RIP software versions, ink series and batch numbers, primer and pre-treatment, print modes, print height above the item, curing settings with the lamp type, lamp hours and measured UV dose, fixtures and colour profiles.
• Trial conditions: date, location, operators, room temperature and humidity, item temperature, material batches and artwork versions.
• Criteria and results: each criterion as agreed, the method used, the number of samples, the measured or observed result, and pass or fail.
• Approved samples: signed samples held by both parties, with their measured values and storage conditions.
• Responsibilities: who supplied materials, who operated, who judged results, and how disputes and re-tests are handled.
• Open items: anything that failed or was not tested, with an agreed action and date.

What the handover file should contain

• Operator and maintenance manuals for the delivered configuration.
• Safety data sheets, in current revisions, for every ink, primer, flush and cleaning product.
• The maintenance schedule, the consumables list and recommended spare parts with part numbers.
• Warranty terms with the start date and exclusions, and the service contacts and response arrangements in writing.
• Software licences and backups of colour profiles, print modes and job recipes.
• Training records showing who was trained, on what, and the competence check each person passed.
• The signed site requirements, the acceptance record and the approved samples.

The production acceptance checklist

Use this checklist to plan the trial and record the results. The values in each criterion are agreed between buyer and supplier for your product, and the last column lists the evidence to record.

1. Application

2. Trial

3. Output

4. Print quality

5. Durability

6. Repeatability

7. Operating requirements

8. Acceptance and handover

When direct UV printing is not the right process

A careful trial sometimes shows that direct UV printing is the wrong choice for a product. That is a useful result, not a failure of the trial. Another decoration process may suit better when:

• The item must survive repeated industrial washing, such as the hot washing of returnable bottles or the glasswashers used in hotels. Some screen-printing inks for glass are rated by their makers against named domestic and industrial dishwasher tests; the UV inkjet ink data sheets we have checked give no such cycle count. Whatever the process, check what a rating covers and wash-test it on the actual item.
• Volumes of one design are very high and stable. Screen printing, labels or sleeves can cost less per piece once their set-up is spread over a long run. Direct digital printing is strongest with many designs, short runs and frequent changes.
• A mirror metallic finish is required. Hot foil and metallisation are the usual routes to a reflective metal effect, so compare a printed sample of the metallic effect on your item with a foil or metallised sample before you decide.
• The print would sit where food or drink touches, or on plastic food packaging. PRO TECHnology's position is that the UV printed layer is not certified for direct food contact, so keep decoration away from drinking rims and food-contact surfaces. On plastic packaging, ink components can migrate through the wall, or transfer to the inside of other pieces when printed parts are stacked or nested. Ask the ink supplier for its food-contact documents and low-migration options before printing, and have the finished pack tested for migration where your market requires it.
• The surface is out of reach of the print heads or the curing lamp, such as deep recesses, tight internal corners or areas behind handles. Drops land less accurately as the distance from the heads grows, and a recess the lamp cannot reach may stay under-cured. Pad printing or a design change may be needed.
• The product faces long, harsh outdoor exposure without protection. Light and weathering tests may show that a different ink, a clear protective layer or another process is needed.

Who wrote this guide, and where trials run

This guide was written by the PRO TECHnology Digital Printing team. PRO TECHnology has worked in digital printing since 2011, and its Digital Printing division supplies industrial printing equipment in the UAE and the wider Middle East, including UV flatbed printers from Direct Color Systems, cylindrical UV printers from LSINC and wallPen vertical printers. PRO TECHnology engineers install the machines it supplies, alongside the manufacturer's engineers where the manufacturer attends, and support them locally with training, consumables and spare parts.

Print trials and demonstrations run at Sun Studio in Al Quoz, Dubai, the group's UV printing and demonstration studio and a sister company that has been printing since 2011. Trials use your own items and artwork on the models that run in production on Sun Studio's floor; other models are shown by arrangement. A trial returns the printed samples, a net speed figure timed on your item, an ink consumption figure and the adhesion results on your material. Other checks in this guide, such as a rub test with your own fragrance, are agreed in the test plan before printing.

Frequently asked questions

Sources

Standards are cited by designation, edition and shortened title as shown in the standards bodies' online catalogue listings, checked through search on 21 October 2026; EN 12875-1 is cited through the Swedish national listing of the same standard. We did not read the full texts, which the standards bodies sell, so this guide describes methods "based on" a standard rather than claiming conformance. LSINC details come from LSINC's PeriOne and Perivallo360m site requirements form (8 July 2026), its 2024 PeriOne, PeriQ360 and Perivallo360m brochures and its 2026 PeriOne brochure. The worked output example is hypothetical.

• ISO 2409:2020, Paints and varnishes: Cross-cut test
• ASTM D3359-23, Standard Test Methods for Rating Adhesion by Tape Test
• ASTM D5264-98(2019), Standard Practice for Abrasion Resistance of Printed Materials by the Sutherland Rub Tester
• ASTM D5402-19(2024), Standard Practice for Assessing the Solvent Resistance of Organic Coatings Using Solvent Rubs
• ISO 2836:2021, Graphic technology: Prints and printing inks, assessment of resistance of prints to various agents
• SS-EN 12875-1:2005 (Swedish adoption of EN 12875-1:2005), Mechanical dishwashing resistance of utensils, Part 1: Reference test method for domestic articles
• ISO 16474-2:2013 (with Amd 1:2022), Paints and varnishes: Methods of exposure to laboratory light sources, Part 2: Xenon-arc lamps
• ISO 3664:2025, Graphic technology and photography: Viewing conditions
• ISO 13655:2017, Graphic technology: Spectral measurement and colorimetric computation for graphic arts images
• ISO/CIE 11664-6:2022, Colorimetry, Part 6: CIEDE2000 colour-difference formula
• ISO/TS 15311-2:2018, Graphic technology: Print quality requirements for printed matter, Part 2: Commercial print applications utilizing digital printing technologies
• Marabu, UltraJet DUV-CP technical data sheet (version 1, 09/2024)