A light from the shadows

Last week I wrote about Xaar’s recent history with thin film printheads, but for this second instalment Xaar’s CEO, John Mills, explains his plans to concentrate on the company’s core bulk piezo technology. 

Xaar has continued to develop its bulk piezo technology.

Mills has tried to draw a line under the old Xaar, unveiling new corporate branding, refocussing the company purely as a printhead vendor and characterising Xaar’s current technology as its second generation, or ImagineX. He comments: “The first generation technology underpinned 30 years of product development. This involved a shared wall printing every third nozzle, which limited the speed to 35kHz. It was not able to print aqueous ink and used a plastic nozzle plate.”

ImagineX deals with these shortcomings but it’s not a radically new approach and should be seen as more of an evolution of Xaar’s original bulk piezo technology. And that’s a good thing because Xaar would not have survived for over 30 years if there was anything fundamentally flawed in its technology – the management over the years is a whole different issue – but the technology has worked and ImagineX represents a series of sensible improvements. 

The term bulk piezo means that the starting point is a largish block of PZT or lead zirconate titanate. This PZT material contains piezo electric crystals that can change shape when an electrical charge is applied, creating the movement that forces the ink out of the nozzle. By contrast, thin film heads use less of the expensive PZT material and when combined with a silicon MEMs manufacturing process can allow for the nozzle arrays to be packed closely together. As Mills noted in the first part of this story, thin film is very good for jetting aqueous inks at high resolution and high speed. But bulk piezo is generally better at jetting fluids with higher viscosity and larger particle sizes. Or to borrow an analogy from athletics, thin film is more of a 100m sprinter, whereas bulk piezo is like a long distance runner.

Xaar is not the only printhead vendor using bulk piezo technology. But Mills points out that Xaar has a different approach to other vendors such as Kyocera and Ricoh thanks to its shared wall architecture. Mills accepts there are some limitations: “You can only fire one nozzle in three and you get cross talk between the channels because the end of the channel is connected to the channel next to it.”

He says that other vendors have different architectures precisely because of those limitations, but adds: “The general consensus is that everybody else is migrating towards the same configuration but the challenge they have is that the ink goes into the firing chamber and there’s no way out. If you get a blockage in the firing chamber then there’s nothing you can do about it whereas with Xaar you don’t have that.” He says that the other vendors could not simply adapt their bulk piezo designs to catch up with Xaar: “They would have to completely redesign their architecture to effectively copy what we have done.”

ImagineX

ImagineX is best described as a collection of features, some of which already exist, others have seen improvements, and some are completely new. Many of these are still in development and some could take up to five years to fully come to fruition.

The obvious starting point is Xaar’s TF or through flow technology, which is not new but does underpin the entire ImagineX concept. This describes the overall architecture whereby the fluid recirculates around the head, rather than just going direct to the nozzle, passing right across the back of the nozzle so that there’s little chance of air bubbles or particles blocking the nozzle. This also allows for a longer than average nozzle open time. 

Another existing feature is High Laydown, which is a way of putting down more ink in each pass using a different print mode. It is particularly useful for certain applications, such as laying down a varnish layer, or for some 3D printers. Mills explains: “So effectively, instead of just being able to change the drop size from maybe five to seven picolitres, you can change the volume of ink by over a factor of five. So you can change mode and use the same heads to put down say a white layer.”

Xaar has also been working to ensure that it can handle higher viscosity fluids, up to 65 centipoise, or cP, in standard mode and 100cP in high laydown mode, which as Mills notes is “useful for industrial printing.” This capability to handle higher viscosity fluids comes directly from the through flow feature. Viscosity is a major limitation for most inkjet printing since the inks need to be fluid enough to be jetted from the nozzles. Most bulk piezo printheads are limited to around 25cP while many thin film printheads struggle with viscosity levels of 10cP or more. But many industrial applications require fluids with particular functionality, such as a higher pigment loading, that tends towards high viscosity levels, so this is a considerable advantage. 

Xaar is also working on faster print speeds, as Mills explains: “We have technologies that will take us to 48kHz which we will be launching relatively soon, and we have demonstrated and have heads working at 57kHz.” For the record, 48kHz would result in 40 percent increase in print speed over Xaar’s current 35kHz, while 57kHz would be 60 percent faster, though both of these are still being tested. 

Mills adds: “We have found a way to get the Xaar heads to print every drop so instead of printing with every third nozzle we can now print every single nozzle at the same time, which will take speeds to over 150kHz with all the through flow and higher viscosities.” This is still in development and will be a couple more years before it comes to market but it’s a significant improvement as it should allow Xaar to triple its print speeds. 

Xaar is also promising increased resolution as a result of new piezo materials and construction processes – much of which seems to have come from the work on the thin film piezo heads – that has doubled the resolution of the shared walls leading to an increase in the native resolution from 180npi to 360npi per row enabling 1440 npi printheads. This is also still in development and could be a couple of years off.

These new materials, together with new manufacturing processes, has allowed Xaar to increase the operating temperature of its heads, up to 200ºC, further opening up the range of fluids that Xaar heads can jet. The advantage here is that one way to jet viscous fluids is to heat the fluid to lower its viscosity, so the ability to work with higher temperatures at the printhead is one way that Xaar has been able to jet such higher viscosity fluids, as Mills notes: “If you can go to even higher temperatures then the range of materials you can print becomes significantly wider so there is real interest in the heads, for extending the material set that can be used in 3D printing.”

Xaar is also moving away from the use of plastic in creating its nozzle plates. Mills explains: “A plastic nozzle plate is really effective in a number of areas but in some markets the nozzle technology has moved to stainless steel and silicon and left Xaar’s technology behind. We have technology around glass-based and silicon nozzle plates and we will be implementing those across the product range.” This is still in development though Mills says that this has allowed Xaar to play with the geometry of its nozzles and to develop new silicon nozzle profiles that allow for significantly longer throw distances. That will ultimately be useful in coding and marking applications.

Aqueous and pigment inks

The other big problem that Xaar has suffered from has been its inability to work with aqueous inks, which meant that the growth in textile printing amongst other areas completely passed Xaar by. Mills points out that aqueous fluids are conductive and says that this creates issues internally with electrodes exposed to the ink. He adds: “We have now changed the way we manufacture the printheads so we can now drive the printheads with aqueous fluids. We have heads running with water-based inks and are now looking for beta partners to fully test that before launching it to the market.”

Mills says that there is growing interest in pigmented inks but that it can be challenging to run a head with pigmented ink. He says: “If you have got to keep it in suspension and stop agglomeration then you need a very stable suspension, and with Xaar heads there’s no opportunity for blockages so if you get bubbles forming from dissolved oxygen or you get agglomeration of pigments in your printhead or you’ve got a fast drying ink so you have problems with nozzle open time, then we solve all of those problems because of the Through Flow technology.”

He adds: “So if you want a high resolution, high speed head and you’ve got a nice dye-based, water-based ink, go and buy Kyocera because all day long it’s brilliant but if you’ve got a 40cP highly pigmented ink that dries pretty quickly then I’ll put our head against anything on the planet to beat it hands down.”

Mills continues: “There are a couple of ink companies out there in textiles who have created new fast drying pigmented base inks for textiles and they have concluded that they need a Xaar head because of the increased open time so you get the best performance out of that ink. And one of the major OEMs in Europe is now looking at building a range of products for textiles using this ink based on Xaar printheads.

“The biggest challenge in a textile ink is that to get the nozzle open time you’ve got to put in a humectant to the ink and therefore you have to dry the ink on the substrate. And if you have a whole heap of humectant in the ink then you have to slow the machine down in order to dry the ink so the humectant directly affects the output of the machine, whereas if you can achieve the same nozzle open time without putting a humectant in the ink, which is what Xaar’s heads can do because you are directly flowing the ink over the back of the nozzle, take the humectant out and like for like you have a machine that runs faster and consumes less power. So we think that as there is a trend towards more pigment-based inks in textiles, that we will have a strong value proposition for water-based markets, particularly in that textile area”

He estimates that it will take Xaar 12 to 18 months to be confident about this, explaining: “We have got multiple heads running the inks for thousands of hours in the labs now that look extremely good. But as we all know there is a big difference between doing it in the lab and doing it in the real world so we will check it in the real world first before we launch the products. So we have got a few OEMs and their ink companies working with us to test it fully before we launch it.”

Taken together, all of these initiatives should be enough to put Xaar back in the game. But since most of these features appear to be sensible evolutions of Xaar’s existing bulk piezo technology, the obvious question is where did they come from?

Mills answers: “In terms of the speed, it was all technology that we had. Some we have created in the last nine months but the majority is stuff that came from either the thin film technology – a huge amount comes from that capability – but you’ve got things that were done over the last 10 years, sat in peoples bottom drawers, and then things that we acquired.”

Mills was reluctant to talk about what kind of technology Xaar had acquired other than to say: “It helps us with resolution, the resolution we can print with and some of the things that help us get to an aqueous printhead.”

From the ashes, a fire shall be woken, 

A light from the shadows shall spring…

 J.R.R. Tolkien The Fellowship of the Ring

New customers

Mill’s approach is obviously working as Xaar has picked up 30 new customers since the start of the year. He says that a lot of them are in glass printing, adding: “Because glass inks are even more challenging than ceramic inks so we are finding that we have got most of the major OEMs who are getting into glass printing are using Xaar heads.” Another area is electronics, as he notes: “So it’s interesting how some of the everyday electronic devices that people have now, have parts of them that are printed with Xaar printheads, or functional materials that are printed with Xaar printheads, so electronics companies.”

He says that Xaar also remain strong in the coding and marking sector, adding: “And other areas in relation, you could list as automotive, parts of cars that are currently painted or decorated in conventional fashions that people are looking to use inkjet as an alternative. Things like interior panels, where you have trim on doors.”

Mills says that some of these customers are working in niche areas but that most can be classed as industrial printing, explaining: “The reason why they come to Xaar is because of the capability to print challenging inks. So for us, when we are looking at an application, if somebody comes through the door and says they are really interested in your printhead and they’ve got a really difficult ink, we know we are going to win that one, mainly because we are the only head that can print it. If they come with a 6cP dye-based water-based ink, well we are going to be up against Kyocera and Dimatix Samba and the chances are that we are not going to win that one. So we focus on the areas where we know we have a competitive advantage, and the good news is the huge amount of opportunity that’s out there.”

To some extent, the ImagineX concept is really more about rebranding Xaar as a company that OEMs can work with. Most of the features that we’ve covered here are the logical evolution of Xaar’s existing technology, albeit benefiting from the work done on the 5601 head and some additional investment. But if Xaar can follow through on all the further advances in its roadmap then the company could be in a very strong position. This can only be good for the industrial manufacturing world, given that Xaar is the only major printhead manufacturer that’s truly independent.

In the meantime, we can see some of these features that have already come to fruition in the recently launched 2002 printhead. I’ve covered this story here, which also includes other technologies, such as Xaar’s Tuned Actuator and AcuChp, which improve the plug and play implementation of the heads. 

So far we’ve only covered the printheads themselves, but in the last part of this story, which I’ll publish in the next couple of days, I’ll look at some of the ancillary products that go around the heads and Xaar’s overall strategy going forward.

The first part of this story is The shadow of a dream, and the third part is Under the sun.