Purely mechanical solutions can still offer many advantages over systems that rely on digital electronic control. Tom Shelley reports

Ingenious mechanics are still often the best way of achieving optimum performance and reliability.

Mechanics require no batteries, suffer from no computer viruses and can usually be relied on to work – even after long periods of non-use.

They are often the cheapest way of achieving a goal, especially when made of plastic, and with the possibilities offered by micro-fabrication can often be made very small.

One of the first places to start looking for mechanical ideas is the IBM website ( www.patents.ibm.com ), which is a vast searchable database. It is available online entirely for free. While it is a good source of information, it must be remembered that the purpose of a well drawn up patent is to fully protect intellectual property rights while making it as difficult as possible for potential competitors to understand it. The database is used most by engineers and inventors looking to see if their idea has previously been invented and patented by somebody else.

One whole set of ideas has come to our attention from Taiwan. While it has yet to achieve the reputation for quality products achieved by the UK, Germany or Japan, nobody can say its companies are slouches when it comes to getting low cost products to market. Most of the latest batch involve clutches and gears, and many originate from the Department of Mechanical Engineering in the National Taiwan University in Taipei.

One example is a very compact, three-speed automatic gearbox. It consists of three pairs of gears on parallel shafts, all of which are constantly in mesh. Different drive ratios are selected by engagement and disengagement of two torque clutches set at different levels, and two one-way clutches. (Its operation can be understood from the diagram.) Also on offer is a very compact, two-stage worm reduction mechanism, which all sits inside a single ring gear. The developers see it as being particularly suitable for fine adjustments of instruments and watch drives.

The developments are available from the agent, Giant Lion Know How giantleo@ms24.hinet.net

Key to the door

Clever use of a clutch is the key to an automatic door mechanism made by the small drives division of Lenze in Germany.

The package consists of a dc motor, worm gearbox, electromagnetic clutch and position sensor mounted together. The clutch is an 'energise to engage' design, meaning that the drive releases if power fails. Thus in emergency the doors can be operated manually. Normally such drives are connected by toothed belt to the main door drive belt. By combining the pulley for the door drive belt into the clutch, an extra set of shaft and bearings is avoided with consequent cost savings. The clutch has customised output parts with two plastic gears, bearing mounted back to the gearbox shaft. The large gear drives direct to the door toothed drive belt without need for any additional pulleys. A second narrow gear drives to the integrally mounted door position sensor. The clutch is nominally oversized for torque, so in normal operation remains locked up and creates a rigid output connection from the motor to the doors. The motor takes full control of acceleration and deceleration. When supply to the clutch is lost, it releases with zero residual torque making manual door operation possible.

Another, possibly even cleverer piece of integrated clutch design comes from Abssac. The company has devised a single machined item that comprises wrap spring clutch, lever arm, bearing mount and positional stops.

The application called for incrementally rotating a six-position carousel using a pneumatic cylinder. The wrap spring is supported on a ball bearing, which is mounted on one end of the spring and held in place by a retaining ring. A lever arm extends from the spring, which attaches to the pneumatic cylinder. A notch in the wrap spring encompasses a stop pin, which limits rotational travel to the required 60 degrees. The carousel support shaft is concentric within the wrap spring and is gripped by it. As the pneumatic cylinder is extended, it rotates the wrap spring in the ball bearing and the carousel rotates with it to the next position. When the pneumatic cylinder retracts, the wrap spring releases its grip and rotates without the carousel to the original position. The clutching action of the wrap spring thus provides incremental unidirectional motion.

Coming up with a simple mechanism, such as the above, does not necessarily mean simple development. Mechanical Dynamics, which produces the Adams mechanical analysis software, cites the example of a device for a child seat retractor mechanism which allows passage of webbing and then grips it.

The device was developed by TRW Automotive’s Seat Belt Systems Division in Washington, Michigan. It is made of injection moulded plastic and has to hold a child’s car seat securely in place in the event of a vehicle crash or sudden stop. When operating properly, a planetary gear is driven round a central rotating sun gear and a stationary outer ring gear as the seat belt webbing is pulled through. When a specified amount of webbing has been extracted, the planetary gear toggles a spring loaded rocker arm that lifts a lever to engage the locking mechanism of the retractor. As the webbing is then retracted to tighten it around the child seat, the locking mechanism stays engaged to prevent further extraction of the webbing.

It was found that the mechanism would not operate properly in the final stages of life cycle testing, and the rocker arm often failed to toggle completely into the locking position. Initially, the engineers suspected gear wobble as the source of the problem and considered tightening up gear tolerances and adding shims to the rocker arm. TRW engineers, working closely with Mechanical Dynamics consultants, created a baseline model of the assembly in Adams. A special algorithm simulated tooth-to-tooth contact forces and associated friction between parts. Multiple simulations were then run with different values of sun gear tolerance, planetary gear axial slop, spring stiffness, rocker arm to spring friction and rocker arm extension.

The study showed that planet wobble and sun gear operation were not significant contributing factors. Instead, friction between spring and rocker arm was found to be critical, correlating with a high incidence of failures during the dust contamination of life cycle testing. Rohit Tangri, manager of TRW’s Global Seat Belt Simulation Group, says: "Low toggle force in combination with increased friction at the rocker arm-to-spring interface – due to contamination – made the rocker arm prematurely stop during the toggle action and not adequately engage the lever." The problem was eventually solved by re-designing the rocker arm with a modified cam surface including a new arc radius, angle and centre position which would maintain gear contact longer as the arm toggled over the spring, providing a firm snap. Samples of the re-designed parts were created by stereolithography and successfully run through life cycle testing. Tangri says: "Getting the re-design done quickly was probably the most important gain. Without Adams, we would have been shooting in the dark. Instead of finding the problem in two months, it may have taken a year or more."

Back to Power & Transmission