Si-Ware Systems introduced a chip- sized fourier transform infrared (FT-IR) spectrometer that has the potential to be integrated into consumer products. Si-Ware has said that its technology is

different because it doesn’t contain any discrete optics – apart from a chip-scale photodetector – which has enabled the miniaturisation and allows for high-volume wafer-level manufacturing. The NeoSpectra Micro consists of a

micro-interferometer integrated onto a silicon chip. Its size is 18 x 18mm and 4mm thick, and has a spectral range of between 1,150nm and 2,500nm. By using silicon MEMS processes,

the company has been able to take the same technology found inside its module instrument, the NeoSpectra, which is already being used for analysis in the oil and gas, pharmaceutical and agricultural industries. So, when integrated into consumer devices, the technology will be able to provide more detailed and useful information, Scott Smyser, executive vice president of Si-Ware Systems, said. The chip-sized spectrometer consists of an optical core module (OCM) containing the ASIC, MEMS chip, chip- scale photodetector, and the lenses all encapsulated inside a glass lid, along with two halogen light sources that are also integrated onto the board. ‘We’re doing that because we want to

make it something that is more easily plug- and-play for the consumer electronics market, but we can still take the OCM and sell that to perhaps more sophisticated designers that want to incorporate their own light sources [in systems] and do something different with the optics,’ Smyser remarked. Si-Ware expects to see its chip-sized spectrometer in consumer devices in 2018. The company is releasing development kits in the early part of the second quarter 2017, and engineering samples later on in the year, which will enable designers to create consumer use cases. ‘The plan is that we’ll be ready for volume production by the end of the year,’ he said. At Photonics West, Si-Ware showed a

particularly as most people are not used to taking measurements in this way. ‘We need to refine that [sampling method] such that we can measure more robustly if there is movement or when it is not possible to have direct contact,’ Smyser said.

Difficult data Indeed, huge challenges remain in making sure the sampling process is robust enough so consumers can produce valid data. Although spectrometers are now being used by less conventional users – such as by those working on a manufacturing line – consumer spectrometers will most likely be purchased online, with users following an instruction manual rather than receiving any level of training. ‘Most consumers are not scientists or chemo technicians, and don’t necessarily understand the process of taking readings repeatedly or with variables kept the same,’ said Ocean Optic’s Langston. ‘So you’ve really got to reduce and simplify the process of the measurement to something anybody can do, and to make sure they can do it without making mistakes, or without creating any invalid data.’ This is one such problem that TellSpec encountered during the development of its spectroscopic food scanner that was due for launch in 2014. TellSpec had planned to crowdsource

“Most consumers are not scientists or chemo technicians, and don’t necessarily understand the process of taking readings repeatedly”

its data to build up databases. However, problems associated with getting consumers to scan properly have caused the company to re-think this approach. ‘Customers are not careful enough with scanning, so the data produced is not adequate enough to be used for a good robust detection,’ commented TellSpec CEO Isabel Hoffman. ‘Our experience has been that whatever is asked of consumers, to scan properly is a huge process

and tricky to get right. So, we have built our databases ourselves, internally, by hiring lab technicians that can do it correctly and carefully.’ This was just one of many barriers

prototype smartphone case containing its spectrometer chip, which was able to detect gluten and caffeine in flour and coffee grains. Although the results were accurate, the samples had to be in contact with the case and held securely for a short period of time. This is one of the things that will need to be worked on for consumer devices,

22 Electro Optics March 2017

TellSpec experienced during the development of its food scanner. The initial challenge in 2014 was miniaturisation, noted Hoffman, but since working with Texas Instruments the company has managed to work with this. ‘Soon after, we had questions about how to manufacturer these small spectrometers, and with who should be our manufacturer. We needed to choose a manufacturer that understood a

The NeoSpectra Micro from Si-Ware is small enough to be incorporated into consumer devices, such as inside a smartphone case or integrated into a mobile phone

start-up, but at the same time could scale to 100,000 units in a month,’ Hoffman explained. ‘Then came the scalability of our cloud detection engine, to go from a small number of users to eventually millions. And finally, how do we distribute these units, and what model of distribution do we use? These are some of the barriers we had to deal with.’ Having overcome these major challenges,

the company remains confident that it will be able to launch its Generation 1 scanner in the near future. Along with crowdsourcing, there are

several cases for building up databases, something that is hugely important for enabling proper analysis of data and to give consumers meaningful information. The method used depends on the application, according to Langston. ‘At Ocean Optics we’re mainly working

through our OEMs as well as developing our own solutions in a number of targeted markets. I think what we’ll see is that it’s more a community of niche users, whether the data is handled by the likes of Ocean or a company that Ocean is supplying technology into, or perhaps even a third party – it could be as simple as a DropBox,’ he said. ‘Also, companies working with customers

that are engaged in applications are becoming more aware of how they can build more of an ecosystem model, or what we call a “platform model”, where you can take advantage on both sides. ‘It will depend on the specific business model of the application in question,’ Langston said. EO

@electrooptics |

Si-Ware Systems

Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44