Electro-optic Sensor Design (EOSD) technology expertise is historically linked to research and development conducted at the at the Defence Science and Technology Organisation (DSTO), Edinburgh, South Australia, which generated core technology for modern microbolometer design and fabrication and other optical sensor technology. The EOSD business owner was lead researcher in the following core technologies.

Microbolometer development

Bolometers developed at the DSTO from 1960 to 1980, showing the progression from traditional vacuum deposition of detectors and hand-machined packaging to microengineering fabrication and packaging. These developments represent core technology for modern microbolometer technology, and were introduced a decade before the coining of the acronym MEMS. The silicon microbolometer was first patented by DSTO in 1980.

Thermal Infrared Detection Aid (THIRDA)

THIRDA is a technology demonstrator sensor developed at DSTO during the period 1980 to 1985. THIRDA employed a linear microbolometer IR detector array and novel micro-scan to achieve optimum field coverage. The photograph (right) was taken at Uluru (Ayers Rock, Central Australia) during filming of the television series Beyond 2000.

The 20x1 format THIRDA detector array (below) was fabricated and packaged in an Australian CMOS foundry. This development led to production of nitrogen-filled packaging for unattended ground sensors and vacuum packaging for imaging arrays. 

Three THIRDA variants were manufactured: a long range IR intrusion sensor, a thermal imager, and a thermal pointer. The last was bore-sighted with a telescope or low-light-level image intensifier. These sensors form the core technology for modern microbolometer-based detection systems, preceding other well-known developments such as those based on the Honeywell microbolometer, patented after the THIRDA sensors were operational.

Infrared Intrusion Sensor (IRIS)

IRIS technology demonstrator being tested at Kata Tjuta (The Olgas), Central Australia.

IRIS is a development of the THIRDA technology demonstrator, manufactured in Australian industry. This patented technology incorporates a dither-scanned IRFPA platinum metal film bolometer or silicon semiconductor bolometer technology, the latter capable of detecting a human target at 1000 metre range with very low false alarm rate.
IRIS is a dual-use technology, some units being configured for defence applications such as unattended ground sensing (UGS), military perimeter and vital asset protection, others for industrial security. IRIS technology was been fully integrated with the UK Racal Classic system for perimeter protection (shown right) and with the US REMBASS system for remote surveillance. A ground line version was developed for fixed installations such as airport perimeter security, with a number of IRIS units connected to a central PC-controlled station. Off-the-shelf wireless data transmission equipment has been utilized for industrial security.

The IRIS 8x2 (optionally 32x2) focal plane detector array (left) was sealed in a standard production microcircuit package with a germanium lid coated externally with a diamond like carbon anti-reflection coating. The package was sealed on a production sealing line and soldered on a PCB as a surface mount component, a then unheard of process for an infrared detector.

Low Cost Uncooled Thermal Imager (LCUTI)

The Low Cost Uncooled Thermal Imager (LCUTI) (right) is another development of THIRDA technology. The imager employed a 128x128 pixel dither-scanned amorphous silicon detector array, extendable to 128x256 pixels, which at the time (circa 1990) was considered by competitors to have inadequate field coverage, but later re-invented as the now widely used 160x120 format. LCUTI employed an aspheric catadioptric optical system and biCMOS readout, with automatic bias compensation and pixilation correction.

Another achievement of the LCUTI program was the use of established microcircuit packaging, evacuated for performance enhancement (left).

The objectives of the LCUTI project were an IR imager capable of human detection at a range of 500 metres, weight less than 1kg, power consumption less than 1W, using off-the-shelf batteries, and low cost optics. Considered not feasible at the time, these objectives have now been achieved following large investment in uncooled microbolometer technology, particularly in the USA.
  1. K.C.Liddiard, “Evolution of Australian microbolometer and uncooled IR sensor technology”, Proc. SPIE 3894, pp.167-176 (1999).
  2. K.C.Liddiard, “Perspective of Australian uncooled IR sensor technology”, Proc. SPIE 4130, pp.208-217 (2000)

Contemporary Technology 

Following the establishment of a new microbolometer fabrication facility within the DSTO Microengineering Section during the 1990s a number of focal plane arrays were designed by EOSD and fabricated in support of collaborative projects and commercialisation. EOSD specified the processing specifications in these projects which continued for the next decade. The facility is based on amorphous silicon technology which was first introduced at DSTO in the 1980s.

The figure to left shows 16x16 and 320x240 focal plane array (FPA) chips fabricated on the same wafer. The FPA was integrated with a CMOS readout integrated circuit (ROIC) designed at the Swedish National Research Establishment (FOA, now FOI) having parallel 16 bit on-chip ADC now common in IRFPA. The FPA was designed for P-doped amorphous silicon, but a Si:N alloy was used in development.

The microbolometer detector design featured the patented DSTO/EOSD single level architecture with novel sandwich-gap conductor geometry and common contacts, the latter now used in large scale production. Imaging was achieved with both FPA formats. This project has historic links to later large scale FPA development in Sweden employing the same detector design.

  1. U.Ringh, C.Jansson, C.Svensson and K.C.Liddiard, “ CMOS RC-oscillator Technique for Digital Read Out from an IR Bolometer Array”, Proceedings of the 8th International Conference on Solid-State Sensors and Actuators (Tranducers’95) and Eurosensors IX, Vol.1, p.138, June 1995.
  2. K.C.Liddiard, U.Ringh and C.Jansson, “Status of uncooled focal plane detector arrays for smart IR sensors”, Proc. SPIE 2746, pp.72-79 (1996).
  3. K.C.Liddiard, U.Ringh, C.Jansson and O.Rienhold, “Progress of Swedish-Australian research collaboration on uncooled smart IR sensors”, Proc. SPIE 3436, p.578 (1998)
EOSD has more recently been engaged in commercial exploitation of Australian developed uncooled IR sensor technology with international clients. Both 160x120 and 320x260 FPA designed by EOSD and integrated on-chip with client designed ROIC have been fabricated in P-doped amorphous silicon.

The packaged 320x240 FPA chip shown at left produced IR imagery of comparable quality to a commercial vanadium oxide (VOx) FPA mounted in the same IR camera.

  1. K.C.Liddiard, J.P.Knauth, R.Decker, B.Altermus, B.Xu and N.Robinson, “Progress of ICC α-cell microbolometer development program”, Proc. SPIE 4820, p.183 (2002)

EOSD has conducted leading edge research and development on next generation passive infrared (PIR) security sensor technology. The patented mosaic pixel FPA technology (MP-FPA) has been demonstrated in a prototype fabricated at DSTO Australia.

The 4x4 format MP- FPA shown in the figure to right comprises 1mm x 1mm pixels with 100x100 sub-pixel microbolometers. This FPA has been the workhorse for sensor analysis and further development.

Left: Performance testing of MP-FPA technology in comparison to off-the-shelf production PIR security sensors.

Contact EOSD for further information