Faster solenoid valve gives food for thought
A longer-life, quick-response solenoid valve designed for sorting applications could have implications for the technology as a whole. Paul Fanning reports.
High speed optical sorting machines rely upon rapid response pneumatic ejector valves, or 'jet valves', to intercept and remove foreign or defective items from a stream of material moving along a conveyor or falling under gravity from a chute. Materials may vary from grains of rice and wheat through to peas, coffee beans, nuts or increasingly, waste materials for recycling.
Such machines rely on optical/vision systems to identify reject items. An eject valve, usually forming part of an array of such valves, is switched on to create an air pulse for a period of time just as the part passes the appropriate eject nozzle. To ensure accurate rejection of only the selected items, the pulse timing needs to be very fast, repeatable and reliable.
Nearly all jet valves employ a direct-operating solenoid, to provide the fastest possible response. For many years, these types of valve had a typical response time – the time period between the application of a trigger signal and the pressure reaching a defined level downstream of the valve – of about 20ms. Manufacturers have recently managed to get this figure down to about 10ms, but it is still the case that standard solenoid valves cannot provide the performance required for high-throughput applications.
The Mean Time Between Failure (MTBF) figures for standard jet valves are one area of concern. Most valves on the market today that are designed for sorting-type applications have an operational life of about 1 billion cycles. This sounds quite high until you consider that a modern, high-throughput application will have a large number of solenoid valves installed side by side on banks of manifolds. After a while, successive valve failures can lead to the machine being out of service for untenable amounts of time. Valves with a much higher operational life rating would make it cost-effective for users to consider replacing entire manifolds as part of their planned maintenance schedule, reducing machine downtime significantly.
Steve Sands, Festo's automation specialist, outlines the problem: "Traditional solenoid valves normally have a round armature inside the coil and you get friction in two places: within the solenoid itself you get metal rubbing on metal and, if you use a rubber compound soft seal in the poppet at the bottom of the valve, after a few million cycles you start to get hardening off and you start to lose the repeatability of the timing."
Another problem is heat, or rather, its removal. A typical solenoid valve can have a power consumption of 10W during its 'on' time, which despite the fact that it is probably only operating with a duty cycle of 50% or less, means that a significant amount of power is dissipated in the solenoid. The problem is exacerbated by the fact that many users elect to over-drive the solenoids to secure a faster response time.
In contrast, the MHJ series of jet valves from Festo has a typical switching time of just 0.7ms and also offers high short-term and long-term repeatability of 0.1ms. A typical service life of 5 billion operations is claimed, although 10 billion has been achieved.
Based on a novel, short-stroke solenoid and a patented damping plate, the design has only one moving part and dispenses with dynamic seals and lubrication. The valves have a power consumption of 7W, with the option of adding holding current control to reduce this to 2W after the initial turn-on period, and can be triggered by standard 5V PLC-level signals.
Given the number of solenoid valves employed, their cost is a critical factor. Says Sands: "Because they use so many of these in their machines, this isn't a product where you can afford to be pushing up their costs. They have to be designed in the same order of costs as standard valves."
The system was developed using extensive CAE modeling of the magnetic, temperature and flow characteristics. "You can only really achieve this through computer modelling," says Sands. "If you leave this to life testing, it just takes too long and you'd never get through the cycle. Your time to market would be absolutely horrendous. I think it's one of our most extensively modelled products."
While the product was developed with a very specific end market in mind, its success in operation has opened up other possibilities. Says Sands: "The spin-off, of course, is that whereas with other solenoid valves we talk about a life of something in the region of 20m operating cycles, a product that has now gone beyond 5bn operating cycles obviously that starts to give us the data feedback that we can start to consider and use in other product developments."