Machine vibrations tuned out with ease

Computer analysis, followed by judicious application of masses on rubber mounts, can do wonders for the better operation of mal performing high performance machine tools at modest cost. The basic idea of adding tuned vibration absorbers to remove particular resonant frequency vibrations has been with us for decades but it is their use in conjunction with refined measurement, and advanced software, which truly makes them cost effective, and allows machining to achieve new levels of precision. As explained by Brian Hardwick of AMTRI, "The theory behind these can be found in textbooks. They are secondary mass/spring/damper systems that are attached to the primary system. The idea is that they should be tuneable, so that their own natural frequency can be made to coincide with that of the machine. When this happens, the absorber removes the energy from the machine, so that the damper mass vibrates instead of the machine." This can be applied to new machines, which are found to have an undesirable vibration, without requiring a major redesign and rebuild, and to older machines, especially those which have been upgraded or otherwise pushed beyond what they were originally intended to do. The process used by AMTRI begins with physical measurement, to find exactly what the problem is. Vibration modes modes and the flexing behaviour of the machine structure are then identified and studied using modal analysis. The optimal solution to improving the performance of the machine may then be determined and applied. This might be the re-design of some part of the machine structure, using ANSYS FEA for further guidance, but very commonly a mass attached somewhere to the machine, either to the structure, or to rotating parts, will do the job at much lower cost and in much less time. The mass is attached via rubber pads, which act as both springs and damping elements. The damper may be coarse tuned by changing the type, thickness and surface area of the rubber pads. Fine tuning is achieved by adjusting the torque on the retaining screws. This changes the stiffness of the rubber spring elements and so the natural frequency of the device. When the natural frequency of the device is carefully tuned to coincide with the resonant frequency of the machine, maximum energy absorbtion takes place and the offending resonance can be damped virtually out of existence. Owing to the requirements of commercial confidentiality, three examples must suffice. One is the case of a roll grinding machine used in a major manufacturing plant in continental Europe. This was found to have a major problem with 'chatter', never a good idea when trying to grind rolls to achieve a high surface finish on the final product. Having established that the problem was a resonant frequency vibration, the solution adopted was to mount an absorber on a bracket on the wheelhead. The second is a titanium plate milling cutter operated by TIMET at Swansea. The main business of the company is to supply components to the aero-engine industry. Titanium is never the easiest material to mill, and in this case there was a resonant torsional vibration. The solution adopted was to mount an annular damper mass on the rear face of the 200mm cutter. AMTRI provided the conceptual design for the damper and Geoff Lawrence of TIMET turned it into a practicality by discovering that there was sufficient room to mount the damper on the rear face of the cutter, and produced the appropriate engineering drawings. As in many cases, this at one stroke improved: accuracy and repeatability, surface finish, metal removal rate, productivity and tool life. For our last example, we go right back to mid-eighties. In Hardwick's own words, "FMT won a contract to supply around 20 horizontal machining centres to a major UK engine plant at for cylinder head machining. It was a multi-million pound installation that involved an army of AGV's trundling cylinder heads from one machine to the next for different operations. During commissioning, Rover found that they could not achieve their required tolerances in circularity when machining the valve-seats on three of the machines - they were coming out oval! Our analysis traced the problem to a flexibility in the casting supporting the rotary table, leading to a resonance of the table/fixture/component assembly. The resonance meant that the fixture rocked to-and-fro in response to the rotating cutting force generated by the single-point boring tools. This led to greater tool push-off in the horizontal axis than in the vertical axis - hence oval valve-seats. It proved a severe embarrassment for FMT at the time, with serious contractual implications. "We found that a tuned vibration absorber fitted to the top of the fixture killed the resonance and cured the problem. However, the fixtures travelled from one machine to another, so every single fixture (all 45 of them!) was fitted up with a damper atop, each of which was tuned by us. The addition of the dampers raised the height of the fixtures by a few inches, and they even had to raise the height of the supervisory pulpit to allow the AGV's to pass underneath with sufficient clearance! The alternative, however, was too horrible to contemplate, as a major re-design would have been needed to be implemented on three of the machines." Adding a tuned absorber is not always the best solution and Hardwick cites examples where problems were solved by placing steel wedges beneath machine beds, and by changing the profile of a beam in the structure. Mark Twain, once said, "First get your facts right and you may then distort them as much as you please." In this case, one might say, first get your facts right, and you can then remove distortions as much as you please. Pointers * Use of careful measurement of vibration and finite element analysis enables to design of simple vibration absorbers which may prove to be a speedy and cost effective way of removing the adverse effects of a troublesome resonant frequency * Alternatively, the same procedure may lead to some other simple and cost effective way of solving the problem NOTE: AMTRI went into liquidation in September 2007, but the services and products described in the article continue to be available from UNIVIB Ltd: UNIVIB Ltd. Innovation Forum Salford University Business Park Frederick Road Salford Lancashire M6 6FP Tel: 0161 743 3531 Fax: 0161 743 3530 UNIVIB info@univib.com Geoff Lawrence ANSYS Star CD