SR handling--where do we go from here? (Process Machinery).
Baling of the crumb rubber requires high platen forces and long drive strokes. Generally, platen forces at compression are capable of 100-125 tons and piston strokes run about 30 inches. All of this activity (filling, compresson and release) needs to occur in less than 15 seconds.
Historically, all balers have been hydraulic, due to the forces required. In the last ten years, advances in motor efficiencies, over torque capabilities and electronic variable frequency drives have offered the possibility of electric drive balers. The advantage of electronic drives is that a more controlled compression rate/ force curve can be obtained and the maintenance of the drive is less. Of course, the rate/force curve could also be improved with a serve hydraulic drive.
Sage Automation has made initial studies of such a baler. Conclusions are that such a baler is practical up to a platen load capacity of about 80 tons. This would then be feasible for much of the rubber produced today. No investigation has been made into any potential quality improvements or cost comparisons between hydraulics and electronics.
One would think that something as intuitively simple as conveying systems would not have much to offer in improving the plant operation. However, conveyors tend to be high maintenance items in any manufacturing organization. As a rule, conveyor manufacturers quote their year warranties based on 2,000 hours per year. For a 24 hour per day operation, this year warranty works out to be about three months.
Improvements in conveyor technology pay off. Replacing the motor gearbox-chain sprocket-bearing combination with a motorized pulley reduces failures and maintenance calls dramatically. Moving from simple relay controls and motor starters to variable frequency motor drives and PLC controls reduces start/stop load peaks and adds flexibility of operations.
While the initial cost of these types of conveying systems is large, the payoff comes in less maintenance cost and less human involvement to compensate for equipment downtime. As with many technology improvements, some side benefits show up. For example, chain and sprocket wear on a belt conveyor used with weigh cells can cause irregularity in the tare weight due to vibration and settling time. Going to a motorized pulley here offers the added advantage of improving the process information.
Generally, all synthetic rubber plants measure automatically for weight, temperature and metal inclusion on all bales. In addition, manual inspection is made for surface stains, evidence of irregular blending, wrapper quality and viscosity. Often, these inspections (always for viscosity) are made by sampling, rather than by 100% inspection. In addition, with the lower head count on finishing lines today, some inspections can be missed should the operator have to address an operational issue elsewhere.
Weight, temperature and metal inclusion measurement equipment is widely used in other industries. This equipment is constantly being improved due to the large customer base addressed. These improvements have evidently been satisfactorily incorporated into rubber plants, and no operational deficiencies have come to our attention in this area. Inclusion of these data into a process data base can be done as part of the overall plant operations.
Automatic vision inspection
Automatic vision inspection is a next logical step for inprocess inspection. Vision inspection has been used in many industries since the early 1970s. Initially, vision looked for discrete errors, such as a missing or askew labels. Improvements have been continuously made in this field, although surface inspection has offered particularly difficult challenges.
Vision systems require illumination to see the item being inspected. Illumination of non-flat surfaces can create shadows, which are indistinguishable from dark stains. Friable bales tend to have surface holes, and they tend to look the same as a surface inclusion or an unblended area. These problems can be addressed with careful lighting design, color analysis and classification of visual defects. These problems are not trivial, and require site and process specific development work.
Automatic viscosity inspection
Automatic viscosity measurement is the Holy Grail of inprocess inspection. Viscosity is difficult to measure even under laboratory conditions; the tests tend to be a closely controlled process with a load-time curve developed. In the rubber industry, the standard is the Mooney viscosity. This is the basic measurement for rubber processability.
Due to the sampling nature of the Mooney test, hours can elapse between a process change and catching that process change. With lines running in the range of 13,000 lbs./hour, even a half-day delay is a lot of unloading, grinding and reblending.
It would be ideal if a true viscosity test could be made in-process. Even if there was a way to compress, spot-heat and auger a portion of the bale, all of this has to he done in under 15 seconds. The only practical tests that might shine light on processability changes are some form of stress-strain tests. For example, could you treat the whole bale as a very large Williams plastometer sample and measure its compressive stress-strain curve? Would the bulk characteristics correspond to the small sample characteristics?
It is simplistic thinking to expect that a single test would accurately track the Mooney test. After all, everyone has had since 1927 to find such a test. However, it is not unreasonable to expect that, over a limited process type, a loose correlation could be made between one or more in-process tests that would track changes in the Mooney values. Again, we are back to the situation where site and process specific development work is required.
This is not a trivial problem, but the potential payback in plant productivity is immense.
Film wrapping of bales
To prevent rubber bales from sticking together in their shipping containers, they must be wrapped with a protective, leak-proof film. The film is somewhat intrusive in the down-stream processing; the customer is buying rubber, not film. Minimizing film and getting a complete seal are goals for this operation.
There have been several programs started in the last year to develop an improved bale wrapper. Two commercial bale wrappers and at lease one contracted bale wrapper have been developed.
Sage Automation became interested in this area due to customer complaints about poor quality wrap and excess film on the bale. Sage has refurbished older bale wrappers for some time now, and saw areas where the older designs could be improved for quality and reliability.
Sage's offering in this field uses a traverse wrapping action. The bale is mechanically stopped on one end, then mechanically pushed through the film for sealing. The bale is mechanically justified on two edges, creating a uniform seal with minimal wrap.
The future push in this area is driven by automatic depalletizing of bales by the end users. Excess film, stuck bales and long tails that are captivated by the next bale in the layer reduce effectiveness of downstream automation.
Loading of containers
Robots were first used in the early 1980s to package rubber bales. The first robots used were some of the first industrial robots, specifically six-axis Unimates. In the mid-1980s, gantry robots came into popular use due to their rugged construction and simple maintenance. These robots came to dominate the market, mainly due to their ability to survive crashing into out-of-specification containers. Various types of container clamps have been used to position the containers in their nominal location, minimizing crashes.
In the late 1990s, four-axis arm robots became available with large payloads and robust design. These robots can now be used to package rubber. Due to their arm design, they cannot reach down into the containers, but require a linear extension arm to their fourth axis. It is expected that they will offer good service, but there is no long-term experience with these robots in this environment as yet, especially on the survivability of the extension arm.
Historically, rubber bales were a loose fit into their containers. Typically, there was 1/2" or more spacing between the bales when packaged. The bales would creep in the container, filling the spaces. Usually, bales were stacked above the top of the container by the robot. After the load sat for a while, the bales would creep, fitting tightly into the container, and the top bales would settle down into the container.
The advent of returnable containers has led in some cases to tighter fits between the nominal bale and the container. Many bale lines now find that they have a nearpress fit between the bales and the container. When manually packaging the bales, the operator can twist the bales into the space and make them fit. When robotically packaging the bales using four-axis robots, this is not feasible.
One solution to robotic packaging of these bales is the use of a bale resizer. Original bale resizers were hydraulically powered. The current generation of Sage bale resizers is pneumatically powered. Elimination of a hydraulic system has been a benefit to this new generation of resizers. Another solution is the high compression bale clamp. This clamp will compress the bale while the robot loader is in motion.
Today, there are six-axis robots with the payload and robustness to survive in the rubber plant environment. In addition, some of these robots have inverted mounts so that they can reach directly into the bottom of the container without the need for an extension arm. These robots offer the ability to twist the bale when packaging it, similar to the dexterity of manual palletizing. This dexterity offers specific advantages when packaging friable rubber, which requires tightly fitting bale liners.
The goal here is to push for simpler, more forgiving container types. The reasonable expectation is that container types will evolve to meet the end user's needs, not the finishing engineer's needs. Being able to respond to the different packaging requirements requires highly flexible automation.
Storage and shipment
The next big area of automation in the finishing department is post-packaging. Currently, all rubber plants use fork truck picking of loaded containers, manual stacking for interim storage and manual loading of trucks and rail cars.
In the last ten years, consumer products companies have started automating this portion of their business. The use of AGVs (automatic guided vehicles), AS/RS (automatic storage and retrieval systems) and conveyorized truck beds has reduced head count, kept accurate track of inventory and reduced product damage. It is not unusual anymore to see a manufacturing plant where the role of the production staff is to monitor the equipment.
The biggest problem faced by rubber plants in implementing these types of changes is old plants that were never designed for continuing the production flow from initial process to final shipment. Significant plant improvements may need to be made to provide the flooring requirements for AGVs and to allow easy placement and retrieval of loads.
Yet again, we are back to the situation where site and process specific development work is required.
Even at today's levels of automation, there is still much room for process and productivity improvements in the rubber plant operations. Some of the improvements are simple to implement (buy new generation robots and bale wrappers, improve the conveyor systems). Some of the improvements are plant specific R&D developments (in-process vision and viscosity testing). And finally, some of the improvements are a little bit of both (non-hydraulic balers and post-packaging storage and retrieval).
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||synthetic rubber|
|Date:||Oct 1, 2002|
|Previous Article:||Patent News.|
|Next Article:||Avoiding errors in thermoset elastomer selection for wire and cable. (Tech Service).|