Greg Blackman explores the laser processing methods used in manufacturing electronics, along with a novel way of assembling tiny electronic components using light
While printed electronics might be a vision of the future and where a lot of R&D work is targeted, building capacitors and assembling printed circuit boards still accounts for the majority of the electronics market. According to the UK Electronic Systems Community report, ‘A blueprint for UK economic growth’, the UK electronics sector contributes around £78 billion (5.4 per cent) to the UK’s GDP.
And there is plenty of activity and government support in the sector, at least in the UK: in a round of funding that began in mid-February, the UK Technology Strategy Board is to invest £4.75 million in collaborative R&D studies in manufacturing electronic systems, registration for which closes on 2 April of this year. Also, at the beginning of 2014, UK Secretary of State for Business, Innovation and Skills, Vince Cable, announced £3.6 million of funding for manufacturing research from the Engineering and Physical Sciences Research Council (EPSRC), one of the projects for which is for research into a novel way of assembling electronic components.
Laser-cutting and welding forms a part of electronics production; a portion of Lasered Components’ business, a laser and water-cutting job shop based in Essex in the UK, is to laser-cut electronic components. Most of this work, according to Dan McGinty, sales manager, involves cutting small batches of prototype parts – a lot of the mass production work takes place in Asia where the components are punched out, he said.
Short runs, however, are best cut with a laser. ‘If you want to make an alteration to it [the cutting pattern], just change the [CAD] drawing and away you go. Anything you can draw, the laser can follow,’ McGinty commented. ‘Punching is fast, but you get a lot of waste. We can choose our sheet size and cut a nest of parts to minimise waste.’
Lasered Components runs three Bystronic lasers: two CO2 and one fibre laser. It’s the fibre laser that is used to cut the copper sheet, as the laser cuts materials less than 3mm thick faster than a CO2 system. McGinty said that a lot of these components will be 0.5mm thick. He said one of the biggest issues is to keep the part still in the machine, because it is so thin and it is cut under air and gas pressure. The components are therefore processed as a whole sheet to minimise movement.
The laser is good for both cutting and welding electronics because it can process small parts very precisely. Diode laser systems provider Dilas supplies diode lasers for selective soldering of electronic components. ‘Diode lasers are used for selective soldering because the laser power can be controlled precisely by an analogue signal and the heat input into the material is very localised,’ stated Steffen Reinl, a product manager at Dilas.
Dilas’ laser processing heads contain a pyrometer as an option, which allows the user to create a closed-loop system to control laser power. The temperature can be set in the software, which then controls the laser power.
‘We compete with standard selective soldering processes like soldering by iron, hot air or induction heating. Laser soldering is much more selective and the process does not damage or input heat into nearby components,’ Reinl said.
According to Reinl, the main applications for diode lasers are either soldering very small electronic components, in the range of a few tenths of a millimetre, or processing heat-sensitive electronic parts that cannot be soldered by conventional methods. A good example is LED chips for car manufacturers, which are selectively soldered with a laser, or soldering PCB boards inside a temperature-sensitive plastic housing, since the laser inputs very little heat into the surrounding material.
Assembling with light
The latest computers and mobile phones are now smaller and slimmer than ever before and this is down to the ability to manufacture extremely small electronic components – Japanese company Murata Manufacturing has produced a monolithic ceramic capacitor 250µm in length. Monolithic ceramic capacitors are part of many electronic products; there are around 400 to 500 in the latest smartphones, for instance.
Two hundred and fifty microns, however, is a lot smaller than the current smallest standard size for a capacitor of 400µm, and the smaller these components get, the more difficult it is to assemble them using traditional robotic methods.
Dr Steven Neale, Royal Academy of Engineering research fellow at the University of Glasgow in the UK, is investigating a novel and radically different approach to assembling discrete electronic components, which uses light to move the parts around. The project received funding from the EPSRC in January of this year, and will start in July and run for 18 months.
‘The current smallest standard is 400µm long,’ commented Neale. ‘Every time they [components] get smaller it is more difficult for a robot [to handle]... whereas it becomes easier for our optical manipulation [at smaller sizes].’
The current assembly method uses a robot equipped with a vacuum nozzle, which picks components from a tape and places them onto a circuit board. Neale’s method is based on opto-electronic tweezers, which use light to control an electrical force that then moves the particles. This is not to be confused with optical tweezers where the light itself, in the form of a laser beam focused to an intense spot, manipulates microscopic objects.
A laser can manipulate particles, but it doesn’t have a lot of force and would struggle to move objects larger than 1µm in diameter. The electronic components this technique is looking to move are hundreds of microns across. By patterning light onto a device with a layer of photoconductive material, the light is converted into a pattern of conductivity, and an electrical field moves the particle.
According to Neale, the forces achieved with this technique are in the range of tens of piconewtons, which can move particles around 100µm at tens of micrometres per second (hundreds of micrometres per second has been reported). Robots would be faster than this, but the advantage of the technique is that it’s able to handle smaller components.
‘Tens of microns, and even single microns and below, is what we would normally do with optical manipulation,’ said Neale. The ability to handle smaller components reliably will distinguish this technique, he added.
Where the robot falls down is when it mistakenly thinks it has picked up a part when it hasn’t, or when it fails to handle the components correctly, Neale said, adding that, as the components get smaller, the optical technique should be able to place them better than the robots. There’s also a possibility of the technique being able to place several components at the same time, which the robot cannot, because of the small size of the circuit boards.
Opto-electronic tweezers have been around since 2005, and 90 per cent of the research is in the biomedical area, for manipulating cells for instance. ‘This project will be manipulating these, what are to me, very large components – they are hundreds of microns [in diameter] – but to a pick-and-place machine it is a really small component,’ explained Neale. He added that there is this gap between objects millimetres in size, which are easy for a robot to handle, and those measuring single microns, which can be manipulated optically with lasers. ‘There is this area in between that we’re going to explore with this project.’
At the moment the group at Glasgow University is patterning the light with a data projector. ‘[The technique] is nice and straightforward in that way,’ said Neale. Laser tweezers are a lot more complex.
Any patterned light source works. Higher light powers do provide higher forces, and so a laser with an SLM to pattern the light would be a very good light source for this as well, according to Neale. However, it would be a more expensive and a more complicated system.
The technique is still very new and still very much in the domain of academic research. ‘I’ve shown I can move one of these components with this technique. The next hurdle is to actually assemble a circuit,’ Neale said, adding that this is certainly feasible within the timeframe of the project.
Following that, the next step would be to get the industrial companies excited about the technique. That would hopefully drive the companies to make smaller components, Neale believes.
‘The question is: are there going to be smaller components coming out? I think there would be if there was a mechanism to assemble them. It’s no good companies making components that are 100µm long if nobody can assemble them.
‘What tends to drive it [the reduction in size] is a new standard coming out. Murata in Japan are trying to drive a new standard by bringing out this range of components [250µm long]. When there is a new standard, everybody can then develop their assembly techniques to work with it.’
Once a standard is in place, all the different manufacturers build their components according to the document, and the robots are built to handle the parts. ‘It really drives [the industry] once a new standard comes out and everybody knows what the next generation will be,’ said Dr Neale.
If these components can be made smaller, then so can devices like mobile phones and computers, but also weight-critical devices like micro-satellites can be made lighter. Opto-electronic tweezers have still some way to go until they are used to assemble electronics in anything outside research, but new assembly techniques are going to be needed if components keep getting smaller and optical manipulation is one technique showing potential.