By Jon Robinson
Part II of The pulse of print heads focuses on the advances of manufacturing piezo and thermal systems for use in inkjet presses taking greater aim at commercial printing and packaging
With the growing range of investment options, PrintAction is producing a series of articles, called The pulse of print heads, to better understand one of the most-critical components of any production inkjet press. In Part 1, last month, we took a look at the relatively simple discussion of drop size, primarily because print head R&D and inkjet messaging for more than a decade focused on printing ever smaller drops of ink with the goal of improving overall inkjet quality, even as some commercial settings may require larger drops for higher volume work.
This month, Part II of The pulse of print heads focuses on the manufacture of print heads and how it relates to the adoption of inkjet presses for a wider range of commercial-printing applications. When a production inkjet system requires dozens of print heads each costing a few thousands dollars, for example, the manufacture of print heads also relates to the initial purchase price of inkjet systems and subsequent print head replacement costs.
Crystals, diaphragms and heat
The past few years have seen the rise of two important technical terms in relation to the key piece of hardware – print heads – of production inkjet presses: Nanotechnology and MEMS. Print head makers and their press-building OEM partners – if not one and the same – have put both nanotechnology and MEMS into play for decades now. Short for Microelectromechanical Systems, MEMS basically describes any type of microscopic device, particularly devices with moving parts.
MEMS manufacturing, therefore, relates more directly to piezo print heads that eject ink with moving mechanical elements, walls or diaphragms. Thermal print head manufacturing is experiencing similarly important advances, albeit with different process definitions, as developers of both print head types absorb massive upfront factory costs to propel the printing industry’s adoption of inkjet.
“When we talk about MEMS, Xaar talks very holistically about our whole product portfolio – older [print heads] and new stuff. The difference being that we now use silicon MEMS, as well,” says Jason Remnant, Product Line Manager with Xaar, which has built inkjet print heads since 1990. He explains silicon is more or less used to form the base of the print head, providing it with fluidic chambers before a film is applied with PZT (piezoelectric pumping components).
Xaar’s older generation print heads were built with what the company refers to as Bulk PZT that would be cut down to make the actuator ejection device, with control signals and a source of energy. The advances in silicon PZT manufacturing provides print-head makers with scalability and accuracy, resulting in an ability to fit more nozzles onto the given size of a print-head plate, with corresponding drivers, at less cost – even if the head may not be as durable as a Bulk PZT build.
In 2007, Xaar started working toward silicon-based MEMS production and in May 2016 introduced its next-generation 5601 print, which is also built with what manufacturers describe as Thin Film technology for holding PZT components. “It has to be biggest thing to come along from Xaar in a decade,” says Remnant. Over the past decade, print head developments ensured the mass adoption of wide-format inkjet for commercial work, as well as ceramics printing and print products with lower quality requirements like the inner pages of books, statements and forms. The commercial printing industry – with its many applications and quality demands – requires a print-head evolution that is well under way.
“The 5601 is a new platform of print heads that will absolutely drive the opportunity to digitize more print in the world,” says Remnant. In addition to reaching higher manufacturing levels at smaller micro-scales (nanotechnology), the new generation of print heads for commercial work, packaging and laminates, need to jet fluids other than solvent and UV. Remnant explains the 5601 can jet low-viscosity fluids, including aqueous and latex-type inks, which also opens up inkjet to the world of textiles.
To deploy the 5601, Xaar is working closely with Ricoh, which holds significant press interests in commercial and high-speed printing markets. “Past print heads have included silicon MEMS techniques and now new designs are being developed. MEMS and thin-film technology are not changing Ricoh’s print head position, but rather, these two technologies are enhancing and expanding Ricoh’s inkjet print head capabilities,” says Joseph Ryan, Director Business Development, Ricoh Printing Systems America.
The most-advanced print head manufacturing models today integrate components to create more of a print chip than a print head. “MEMS is a bit of a misnomer for HP thermal inkjet technology,” says Ross Allen, Senior Technical Specialist, HP Inkjet Technology Platform, who first joined the company as an engineer in 1981. “There are no moving mechanical elements in an HP print head. The ink is the only moving part. So, HP thermal inkjet is a MicroElectroFluidic System, and that term is not in common use.”
HP builds its newest generation of print heads with silicon and photolithographic polymer technologies. Allen explains this allows the entire print head, including on-board electronics, to be built with technologies that were originally developed for manufacturing integrated circuits like computer chips. HP’s MicroElectroFluidic advances resulted in the launch of its Scalable Printing Technology (SPT) around a decade ago. Allen explains SPT enables fine structures, both electronic and fluidic, to be defined, precision-aligned and built on a silicon substrate.
Just as Xaar faced limitations producing Bulk PZT, HP also previously faced manufacturing challenges with its original thermal heads because they employed separately fabricated nozzle plates that had to be mechanically aligned and adhered to a silicon substrate with fluidic channels and chambers. Allen explains more complexity came from the use of different material properties, such as thermal expansion between an electroformed nickel nozzle plate and the silicon (polymer) component.
“By building fluidic – ink – chambers, passages, and nozzle plates out of the same photo-imageable polymer in layers up from the surface of a silicon wafer – with its electronic circuits – larger and more complex print heads may be produced,” says Allen. “HP thermal inkjet print heads are essentially integrated circuits that eject ink.”
Like Xaar’s 5601, Epson’s PrecisionCore and Fujifilm Dimatix’ Samba technology, HP SPT is print head platform, meaning it continues to receive R&D dollars to include what Allen describes as smaller fluidic structures: Smaller drop generator chambers, ink passages, nozzles and built-in filters that catch particles in the ink.
“This means that current generations of an HP print head chip – typically about an inch long – can have thousands of identical nozzles and deliver two or four different colours of ink. These chips are placed end-to-end, staggered – and with a small overlap – to build print heads that are 4.25- and 8.5-inches wide.”
Compact nozzles and zones
The ability to design nozzle-dense print heads – and manufacture them on a grand scale – is critical for inkjet-press adoption in commercial printing for a number of reasons from quality to cost. Technically, nozzle-dense heads allow press makers to build larger format presses with smaller print zones. Xaar’s 5601 is built in a Z-pattern to interlace the print heads and reduce the printing area of – ideally – a single-pass inkjet press built by one of its partners.
A smaller print zone reduces potential printing complications with fast moving paper. “Being able to assemble a number of print heads into large arrays allows large systems to be assembled,” says Ryan. “Aligning print heads, especially in high-resolution printing applications, has always been a challenge to system designers. Almost all print heads have alignment techniques using precision locating pins, flat control surfaces, and incorporating physical configurations, such as Z forms and trapezoidal configurations for interlocking and alignment.”
Employing traditional print heads in a single-pass production inkjet press, explains Xaar’s Jason Remnant, typically required staggering the print heads on a print bar to address issues like number of applicable colours and redundancy, particularly as press format sizes increased. Staggering heads can equate to deeper print bars, which in turn increases the print zone. “A small print zone is really critical because it has a [reduced] cost on the build of your machine and it also has a big influence on the print quality of your output,” explains Remnant. “If you are making a huge single-pass printer and it turns out that your print zone is two-metres wide, you have to control your substrates [to] get them from the first colour all the way to the nozzles of the last colour – and [the paper must] be where you expect it to be, so the drops end up where you want them.”
Challenges of running a larger print zone are exacerbated, explains Remnant, because it allows for more swelling when paper is hit with fluids, particularly if absorbing water. “Part of the design of this  head was to allow the OEM to make a very compact print zone and, in fact, the concept for a four-colour system with our print speed would actually mean you are printing quicker than the swelling of the paper.”
The application of staggered print head bars, of course, becomes efficient when building integrated print chips with super-packed nozzles. For the first generation of print heads used in the HP PageWide Web Presses, Allen explains nozzles were spaced in two offset columns of 600 nozzles per inch to print at 1,200 dpi across the web. “The newest generation of HP print heads, called High Definition Nozzle Architecture, places small drop weight nozzles between the original high drop weight nozzles for dual drop weight printing. Across each ink feed slot – a slot through the silicon chip that supplies ink to the fluidics layer – these print heads feature 2,400 nozzles per linear inch,” says Allen. “A low drop weight nozzle prints in the same dot row as a high drop weight nozzle across the ink feed slot, so the printing resolution is still 1,200 dpi across the web.”
HP’s print head build with integrated circuit technologies means many hundreds of its print head chips can be made on one silicon wafer. “This leads to large economies of scale in manufacturing,” says Allen, “where many different print head series can be built in the same HP factory.” Economies of scale provided by today’s print-head manufacturing results in lower-cost products that will ultimately affect the price of production inkjet presses and introduce a wider range of lower-cost, smaller-format systems for commercial printing. With growing use of total-cost-of-ownership investment models, printers should also consider the cost of replacing silicon-based print heads.
“I don’t see any breakthroughs coming in any inkjet technology that could be considered a dramatic reduction in replacement cost. HP SPT already delivers manufacturing economies of scale that are reflected in print head price,” says Allen. “What could happen to reduce effective print head cost-to-print is longer print head life, which drives down cost per square metre. Of course, HP and others are always working to develop longer life, more reliable print heads, but lower prices will be evolutionary and not a dramatic breakthrough.”