What to inspect at a machine demo, and on a used machine

A demo is usually run by the seller, on their file, on their material, in their favourite print mode. None of that is dishonest — it is just not a test. Here is what to do instead, whether you are watching a new machine or inspecting a used one.

Take this list on your phone. Every item is something you can do in someone else’s showroom in under an hour.

Print these five things, in this order

1. A nozzle check — before and after a long print

Ask for one cold, at the start. Then run a substantial print, and ask for another immediately afterwards.

A machine that passes cold and fails hot has a thermal or ink supply problem, and it is the single most revealing test you can run in a showroom. Nobody demonstrates this voluntarily.

2. A solid light grey, and a solid black

Not a colourful photograph — those hide everything. Banding lives in flat midtones, and light grey is the most unforgiving thing you can print. A large flat black shows gloss banding, mottling and cure differences.

3. The speed mode they quoted you

If the quotation says 100 m²/h, print at 100 m²/h — not at the slow, beautiful mode.

And watch the cure at that speed. One manufacturer publishes this on its own specification page: “Inks would not be cured when printing speed is faster than that of draft mode.” Machines can move faster than they can cure. See what pass count really means.

4. Unidirectional, then bidirectional

Same file, both ways. If the unidirectional print is clean and the bidirectional is not, alignment or head height is off. That is adjustable — but you want to know before you buy, not after, and it tells you something about how the machine has been kept.

5. Your own material, and then leave it alone

Bring your substrate and your artwork. Print it. Look at the cure immediately — but do not run a tape test on the spot.

Two ink manufacturers independently specify that final adhesion is reached after 24 hours. An immediate tape test will fail a job that would have passed the next morning. Take the sample home, test it tomorrow, and judge it then. Full method in will the ink stick?

Measure these four things

• Media feed accuracy. Print a one-metre line and measure it with a steel rule. It should be a metre.
• Vacuum hold-down across the whole bed, not the middle. Put a light sheet at each corner and see whether it holds. Corner vacuum loss is common and it is what lifts a sheet into the carriage.
• Head gap against the published figure. Real published ranges: Mimaki JFX200 is 1.2–3.0 mm with a 1.5 mm default; Mimaki UJF MkII is 1.5–5.0 mm; Roland LEF-20 auto-sets 1.0 mm. A machine running much higher than its default is telling you something — usually about flatness or past head strikes, and a raised gap is itself a cause of head damage.
• Media flatness. Mimaki specifies surface unevenness of 0.5 mm or less. Check the board they are demonstrating on, and check whether your own stock meets it.

Ask these questions about the head

• Does it recirculate? This changes your daily routine and matters enormously if you run white. The Ricoh Gen5 and Epson i3200 do not; several Konica, Toshiba and Kyocera heads do. See why white UV ink settles and the printhead guide.
• Show me the nozzle map — how many nozzles are already compensated? On a used machine this is the question. Software compensation hides dead nozzles beautifully until it runs out of room.
• Is there an hour or pass counter, and where do I read it? On a used machine, “lightly used” is a claim; a counter is a fact.
• What is the head warranty, in writing? Be direct. We could not obtain a published printhead warranty term from any machine manufacturer, and the norm in this market is around twelve months on the machine with the printhead either excluded or covered for a shorter period. Get the actual answer on your actual quotation.

The training question most buyers forget

Trade guidance from Nazdar puts it well: buyers “must ask up front what is included in the training. For instance, will it cover maintenance? RIP navigation? Profiling? Repair?”

Their expectation is eight to sixteen hours of training on wide-format equipment, and a useful warning: “The training phase will take longer than the installation.”

Ask who is trained, for how long, on whose material, and what happens when that person leaves.

The ten reasons machines break

A ready-made list from the same source, and it is worth reading as a buyer because most of it is about you rather than the machine: not reading the manual; improper maintenance; poor electrical connections or bad power; overrunning the machine’s capability; not replacing worn parts; untrained staff; maintenance skipped for scheduling; not understanding the equipment; operators who do not care; and no investment in calibration tools or training.

The third and fourth deserve special attention in India. Power quality is covered in setting up the room. Overrunning capability is what happens when a shop buys the cheapest machine that can technically do the job and then runs it at capacity every day.

Commercial checks

• Serial number and manufacture date — verify, do not accept verbally
• Firmware version, and whether it can still be updated
• RIP licence: is the dongle legitimate, and does it transfer to you? A non-transferable licence is a cost you discover later.
• Ink policy and warranty. Ask directly what third-party ink does to the warranty. Xaar notes that ink past its expiry “may affect performance and… cause damage to the printhead.”
• Spare parts: what is held in India, and what has to be ordered
• Who services it, from which city, and how — our own answer is on the service page

The reserve nobody budgets

Trade press quoting Nazdar suggests budgeting 10 to 15 per cent of machine cost per year for consumables, printheads included, with heads lasting one to three years.

Whatever the exact figure for your machine, the point stands: a machine bought with nothing left over for consumables is a machine that will stop. Put that reserve into the payback calculation from the start.

On buying used

Everything above applies, plus three things.

Ask why it is being sold. A shop upgrading is a different story from a shop that could not make it pay, and the second usually means the machine was run hard and maintained badly.

Look at the maintenance evidence, not the paint. Capping station, wiper, waste tank and the area around the head tell you more than the panels. A spotless exterior with a filthy capping station is a machine that was cleaned for viewing.

Price the head as a near-term cost. If nobody can tell you the head hours, assume you are buying a head replacement in the near future and negotiate accordingly.

One honest caveat

This checklist is engineering practice, not a published standard. No manufacturer issues a buyer’s inspection protocol, and we are not going to imply one exists. It is what experienced operators do, and every item traces to a documented failure mode covered elsewhere on this site.

If you want to run it on our machines, that is exactly what a demo is for — bring your material and your file. Compare specifications side by side first on the compare page.

Sources

• Mimaki specification pages — cure limitation above draft speed, head gap ranges, media flatness requirement
• Roland DG documentation — head gap auto-set, bidirectional correction
• Epson documentation — unidirectional print as a diagnostic
• Marabu and Nazdar technical data — 24-hour final adhesion
• Wide-Format Impressions / Nazdar — training scope and duration, the ten reasons machines break, consumables budget (trade press)
• Xaar — ink expiry and printhead damage