Of all the machine tools, the power press or punch press has over the years presented one of the most formidable challenges to the acoustical engineer. There is no single noise source present but rather many noise sources. Some of these noise mechanisms can be traced to easily identifiable singular causes, while others result from a complex combination or sequence of events.
In addition, the noise paths are, in some cases, straightforward and direct, while other are insidiously confounding. However, from recent research programmes, the major noise sources or mechanisms have been identified and some noise reduction measures developed. As has been our convention, we shall first identify the major sources of noise and the character of the noise and then consider those measures or design guidelines for control.
Principal Areas of Power Presses:
ADVERTISEMENTS:
The punch press can be separated into four principal areas or functions with respect to the emanation of noise.
In somewhat rank order of intensity they are as follows:
1. Die space area.
2. Press equipment and controls.
3. Press structure.
4. Material handling.
ADVERTISEMENTS:
In the die space area, the noise originates from:
1. The metal-to-metal impact of the tool or stripper plate.
2. Part breakthrough or fracture.
3. High-velocity air to eject parts.
ADVERTISEMENTS:
Press drive equipment and controls such as gears, pneumatic control vents, motors, etc., are also major contributors but usually only on a cumulative basis.
With respect to press structural members, significant noise which is difficult to control is often radiated from the frame, die sets, pitman flywheel cover-guards, etc.
Finally, there is the metal-to-metal clatter associated with material handling which cannot usually be ignored. Here we have part-to-die-set impact at load, part-to-part impact or part-to-bin impact after ejection and finally scrap collection. Although these areas or functions have been ranked in order generally, it must be emphasised that any of these basic sources or mechanisms can, in any given blanking or forming operation, be dominant.
More often, however, it is the accumulation of all sources that renders significant noise level reduction difficult to achieve. Now, of the sources listed, specific control measures for all but the metal-to-metal impact and fracture which occurs in the die area. As such we shall focus attention mainly in the die area. However, a few supplementary comments which are unique to presses will also be presented.
ADVERTISEMENTS:
Noise emanating from the die area generally originates from the following mechanisms or cause in somewhat sequential order:
1. Stripper plate (if used) impact with the workpiece.
2. Tool impact with the workpiece.
3. Breakthrough (fracture).
4. Restrike or knockout.
5. High-velocity air to eject parts.
The high velocity and subsequent impact of metal parts and scrap against tote bins is often a significant contributor to press room noise. In fact, where large presses (600 to 2000 tonnes) and corresponding large parts are being handled, the metal-to-metal clatter associated with parts handling can be the dominant source of noise. Consider, for example- the clatter a sheet of steel (soon to be a truck fender) makes when it is placed (thrown) into the press die.
Next comes the rather modest crunch of the forming operation and then again, the impact noise as the scrap and fender are thrown into a metal tote bin or stacked. Another example is the noise made as small parts (several ounces) are blown out of the die onto the side of a metal tote bin and subsequently fall to the bottom, impacting other similar parts.
Material handling, like die area noise, is not easy to control, but listed below are two basic guidelines:
1. Treat areas where parts impact the tote bin with rubber or polyurethane impact pads.
2. Minimise the parts or scrap drop distances.
Press Room Machine Layout:
Finally, significant reduction in noise exposure can be accomplished by following some common sense guidelines in press room layout.
As a first consideration, isolate (when possible) typically noisy equipment from quieter operations. In particular, it may be possible to wall off high-speed blanking operations, thus reducing exposure to adjacent manual press operators and inspectors. The wall may be of simple stude construction, acoustical panels or flexible acoustical curtains.
Obviously, the application of these press room layout guidelines are easiest to implement and more effective when facility expansions or new machine installations are in the planning stage.
In summary, press room noise control must be considered as one of the most challenging areas of noise control. There are emphatically no simple gimmicks.
The measure and guidelines presented herein will, in most case, provide measurable noise level reduction and in many cases will significantly lower operator exposure. It must, however, be emphasised that a coordinated effort among the tool designer, manufacturing engineer, setup and the maintenance department is essential for effective results.
Hydraulic Pumps:
The varieties of hydraulic pumps are far too numerous to list. However, there are four basic types which are commonly found in industrial applications: centrifugal, screw, reciprocating and gear. In all cases, discrete parcels of fluid are taken in at the inlet, compressed and recombined at the discharge. Ideally, the flow would be steady with no fluid or pressure pulsations.
In practice, however, the fluid flow and discharge pressures are not steady but contain periodic components due to the pump compression mechanism. The compression mechanisms may include vanes, pistons, gears, screws, etc. It is from these pulsations that the major sources of pump noise originate. Additional broadband noise is often present due to mechanisms such as cavitation, turbulence, etc.
These broadband sources are usually secondary in magnitude and related to poor pump design or applications.
Characteristics of Pump Noise:
With respect to the magnitude of pump noise, little can be said because of the number and diversity of basic pump designs.
It is pretty well acknowledged by manufactures that reciprocating pumps with odd numbers of pistons produce less flow pulsation amplitude or ripple. For example – a nine-piston pump generates one- fifth as much ripple as an eight-piston pump of equal displacement. It is for this reason that most piston pumps are made with an odd number of cylinders. The added bonus is of course generally less noise.
In many cases, most of the noise from pumps is radiated from the case of the pump itself. As such one of the most effective noise reduction measures is to totally enclose the pump. Often the enclosure design is rather simple since most pumps do not require air for cooling purposes and therefore only strict attention to detail is required for acoustically sealing the plumbing and shaft penetrations.
With respect to the plumbing penetrations, rubber grommet like seals provides a good acoustical seal. In addition, if the rubber is soft, durometer 40 to 60, good vibration isolation is also obtained. With respect to the shaft, most pumps are driven directly from the motor. As such a slot can usually be cut to allow the shaft to penetrate the enclosure.
Naturally, the clearance should be minimal. With the pump itself enclosed, noise associated with the plumbing must be given priority. Here the noise, which is fluid-borne, radiates from the pipes themselves, usually at the compression frequency and harmonics thereof. Reactive silencers designed to reduce noise at the fundamental compression frequency are quite effective. However, their range of effectiveness is rather narrow and performance falls off rapidly for even small changes in speed.
One other interesting design approach is to split the flow equally and then recombine. Now if the difference in path lengths is one-half wavelength, the fluid pulsations are 180° out of phase when they recombine and they cancel. This device is called a Quincke tube and is often used in aircraft and shipboard hydraulics.
The limitations are clear- the device can only be tuned to a single frequency, say the fundamental and integer odd harmonics. For the even harmonics, the pulsations are in phase when recombined and thus there is reinforcement.
In summary, devices to reduce fluid-borne noise are somewhat limited in overall noise reduction. As such, pipe wrapping or lagging methods are often the only practical approach for significant noise reduction.
There is another source of noise related to the plumbing which always must be considered. The fluid pulsations and motion of the pump will cause the mechanically coupled pipes to vibrate. As such they often impact against bulkheads, other adjacent pipes, walls, etc., with a resultant clatter. This pipeline vibration can be controlled by using an abundance of clamps and supports with elastomeric or rubber spacers between the pipes, bulkheads, walls, etc.
In addition, a short length of flexible hose should be installed near the pump inlet and discharge to provide isolation between the pump and the pipes. In high-pressure systems, 3000 psi and up, two sections of flexible hose installed at 90° are strongly recommended for vibration-sensitive installations.
Finally, the vibration of the pump itself can transmit high-level vibratory energy to the supports, mounting table and beyond. This structure-borne noise can be controlled by utilising good-quality isolation mounts. One rule of thumb which applies directly to pumps is to select a mount such that the system natural frequency (fundamental) is at least 0.5 times the shaft rotational speed. At this point, transmitted energy at the shaft rotational frequency is sharply reduced and the energy transmitted at the compression frequency and harmonics will usually be more than 30 dB down.
Guidelines for Control of Pump Noise and Vibration:
Guidelines for control of pump noise and vibration can be summarised as follows:
I. Isolate Pumps:
1. Use flexible couplings between pump and motor shafts.
2. Use adequately sized hose between pump pressure port and start of pipe runs. A two or three- foot length is generally sufficient.
3. Avoid mechanical contact between suction and return lines and fluid reservoir.
4. Isolation mounts between the pump and motor and the reservoir or machine frame are essential.
1. Mount all directional and flow control valves to the machine base or frame. Avoid in-line mounting. Smaller two-port solenoid, in-line flow and in-line check valves can be considered rigidly mounted when installed directly next to a clamp point.
III. Minimise Fluid Resistance:
1. Use adequately sized components and lines.
2. Avoid drilled junction blocks.
3. Use bends in the conduits rather than right-angle fittings. Straight fittings may be used to facilitate disassembly and service.
IV. Secure the Fluid Conductor System:
1. Use resilient cushion clamps.
2. Clamp spacing distance should not be more than 60 diameters. Where severe vibration is present, clamp spacing distance should not be more than 30 diameters.
3. Provide clamps as close as possible to each side of a bend.
Noise from Electrical Equipment:
Noise from electrical equipment such as motors, generators, transformers, ballasts, etc., is characterised generally as a discrete low frequency hum. Narrow-band spectral analysis will show, in most cases, a fundamental frequency at twice the line frequency, i.e., 2 × 60 =120 Hz or 2 × 50 = 100 Hz and integer- ordered harmonics thereof. The question is often asked, ‘Why twice the line frequency’? To see this, one must consider the origin of the noise.
When there is an increase in magnetic flux density in ferromagnetic materials, a mechanical strain called magnetostriction results. Most iron alloys increase in length when magentised; nickel alloys, on the other hand, decreases in length. For example – a transformer core excited by alternating current will experience an increase in flux density twice every cycle. Hence, the magnetostrictive displacement and corresponding noise occur at twice the line frequency and harmonics thereof.
These dimensional changes, though small, often less than 1 part in 10000, can provide a high level of acoustical energy when mechanically coupled to a large radiating surface such as found in transformers.
Another discrete source, commonly referred to as pole attraction, occurs dominantly in rotating electrical equipment. Here rotors or stators are displaced due to applied magnetics stresses. The effect again at twice the line frequency is most pronounced in slotted core structures or in dual-pole single- phase motors or generators.
Thus, with the major noise sources defined and characterised, we can now consider the major electrical components and methods of noise control. It should be emphasised that other noise sources such as cooling fans, bearings or pumps and noise due to vibration may well exceed the noise of electrical origin.
The noise from transformers is almost exclusively due to magnetostricive effect in the cores. As such the magnitude of the noise increases steadily with flux density in the core up to saturation.
With respect to noise reduction, few guidelines are available. Extensive studies have shown that little noise reduction is available at the source, that is, at the core. Core material variations, lamina shape modifications, core clamping and internal barriers yielded only relatively small noise reduction.
In short, the noise levels associated with currently available transformers represent, more or less, the state of the art in noise control. As such, partial barriers or total enclosures are the only practical measures available to the acoustical engineer.
For large power distribution transformers located outdoors, partial barriers of masonry construction have been used successfully. However, since the noise is dominantly low frequency, 120 to 480 Hz, rather high walls are often required to control defracted sound. In critical noise-sensitive neighborhoods, a total building-type enclosure should be considered.
Until recently, relatively large transformers were also present in rectifiers, welding units, etc. Now advances in solid-state rectifiers, which are virtually silent, are making the older types obsolete.
The variety of types, sizes and models of electric motors and generators is enormous. However, since 1969, National Electrical Manufacturers Association (NEMA) has provided lists of acoustical power levels not likely to be exceeded for several common more types. It must be emphasised that the overall power levels are A weighted and include the noise associated with cooling fans which in many cases is dominant.
Note in particular the sharp increase in acoustical power as shaft rotational speed increases from 1800 to 3600 rpm, which is characteristic of fans. Therefore, since air-cooling fans are generally the major source of electric motor noise and electrical noise is negligible we shall focus attention on the control of cooling fan noise.
For the cooling fans of totally enclosed fan-cooled (TEFC) motors up to 150 hp, absorptive-type silencers can be used effectively.
Typically an overall noise reduction of 6 to 10 dBA can be expected from these silencers. For larger motors, however, total enclosure with carefully designed low-pressure-loss inlets and exhausts is the only practical approach to significant noise reduction. With total enclosure, there is virtually no lower limit to acoustical isolation.
Shaft misalignment or rotor unbalance can also create excessive vibration and corresponding additional noise. Since these are maintenance items, little more need be mentioned. Finally, careful attention to vibration isolation should always be a design consideration for a motor installation regardless of size or shaft rotational speed.
Vibratory Bowls:
A very common noise source in manufacturing areas is the vibratory feeder bowl. Such feeders are used extensively in assembly operations, with automatic machine tools and as screeners or sorters, etc.
Basically, the feeder bowl consists of a cylindrical bowl with a spiral track attached inside of the bowl. The bowl is coupled through leaf springs to an electromagnetic base exciter unit. The angular vibratory motion of the bowl then causes the parts, generally small, to move or, in a sense, to march, in an almost military fashion up the spiral tract to the exit ramp.
The major source of noise is the part-to-bowl and part-to-part impact. The noise is broadband in spectral character with peak levels typically in the range of 2000 to 10000 Hz. Shown in Fig. 22.5 is an actual octave band spectral analysis of the noise from a small vibrator bowl with the microphone located at approximately 1 m.
There are several approaches to noise reduction for these devices and the most straightforward is total enclosures for vibrator bowls. Note the transparent upper section for good visibility and the access door at the top. Partial enclosures, such as transparent plastic lids, are relatively ineffective, with noise reduction seldom exceeding 3 to 5 dB.
Application of polyurethane or rubberlike damping material to the inside bottom of the bowl and the track has, however, been shown to be effective. The damping material acts as a cushion between the parts and the bowl and significant noise reduction of up to 8 to 10 dBA has been reported. One novel approach, which is often applicable, is to add water or solvent to a depth of approximately 1 inch in the bowl.
Measurements evaluating this approach showed an overall noise reduction of 12 dBA. Further, only minor adjustments to the excitation level or vibratory amplitude were required. To maintain part movement, it should be emphasised that damping material applied to the exterior of the bowls has been shown to be ineffective, with noise reduction often less than 2 dB.
Intensely high noise levels in excess of 124 dBA are common in routing operations of both wood and metal. The noise is discrete in character at the router tool impact frequency with intense integer-ordered higher harmonics usually present. The presence of higher-ordered harmonics is due primarily to resonant conditions associated with the dimensions of the workpiece itself.
The actual impact of the tool to workpiece is at shaft rotation frequency or 2, 3 or 4 times shaft rotation frequency, depending on the configuration of the tool bit. Hence, the strong higher sixth, twelfth, twentieth, etc., harmonics are effectively radiated resonances associated with the workpiece itself.
In some sheet metal routing operations a Numerical Control (NC) positioning approach has been used successfully. With this approach remote monitoring and operator isolation in a control room has proved practical and effective.
A novel approach which is applicable for some metal routing is to perform the actual machining underwater. Here the actual cutting is done beneath the water surface at a depth of 1 to 2″. In this setup the workpiece is clamped under approximately 1.5 inch of water. Overall noise level reductions of more than 15 dB were achieved with no loss in production quality. Further, as an added bonus, the life of router bits increased fivefold due to cooler operating temperatures.
Planers or Surfacers:
The intense noise from wood planers or surfacers originates from the impact of the rotating knives and the workpiece. As such, strong discrete tones are present at the knife blade impact frequency and integer higher-ordered harmonics thereof. Further, at idle (not cutting) as the knives pass stationary surfaces such as the chip breaker, intense noise of aerodynamic origin is also produced at the knife passing frequency and higher harmonics. Noise levels in excess of 110 dBA are common.
Progress in the control of planer noise has, however, been encouraging. Planers lend themselves rather readily to total enclosure. Here conventional acoustical panels or even stud wall construction has been used successfully. One note of caution – noise radiated from the workpiece itself during the planning operation is considerable, if not dominant. Hence, a lengthy inlet tunnel and exit tunnel are required for optimum noise reduction. These tunnels can be of plywood construction and should include absorbing materials on the interior surface.
The most encouraging breakthrough in planer noise control is the use of continuous helical cutter heads. With these heads the cutting edges are in continuous contact with the workpiece, eliminating periodic impact. Overall noise reduction of more than 15 dBA has been reported with these helical cutter heads. In addition, the discrete aerodynamic noise at idle essentially vanishes, further reducing operator noise exposure.
Tool Squeal or Screech:
A major noise source in machine shops is the shrill screech that often occurs during metal cutting on lathes, drills etc. The spectral character of this noise source is almost exclusively discrete and often nearly a pure sinusoidal tone. It is easiest to understand this noise mechanism itself by considering the special case of its generation applied to lathes.
Shown in Fig. 22.6 is an illustration of a lathe tool cutting a chip as the part moves with linear velocity V. Now, as the tool ‘plows’ along, the tool is deflected, as shown by the dashed lines, in the direction of the moving part. This deflection is due to steady forces required to force the tool through the metal part. In addition, there are dry frictional forces between the tool and part, which also tend to deflect the tool. These forces, it turns out, are not steady.
To see this, note that when the tool is deflected exclusively due to frictional forces, it springs back toward equilibrium when the frictional forces are overcome. At rest, frictional forces again deflect the tool and the motion repeats itself. Now, when the tool moves in the same direction as the part, its velocity relative to the part is less than when it springs back or moves in a direction opposite to the part.
Hence, we have a varying or oscillating relative velocity AV between the tool and the part. Now the behaviour of dry frictional forces with respect to velocity is such that a negative slope or inverse relationship exists. Note that as the velocity increases, the frictional force decreases. One experiences this phenomenon when moving or pushing furniture across a floor. Intuitively, as the velocity, of, say, a heavy chair increases, the pushing generally becomes easier.
Finally, given this relative velocity difference between the tool and part, we have a plausible argument for the oscillating frictional forces acting on the cutting tool. Further, the oscillation is regenerative or self-exciting and thus we would expect the frequency of oscillation to be strongly related to the natural frequencies of the tool and holder system. Experimental data show this to be true and explains the discrete spectral character of the noise.
This form of regenerative self-exciting vibration is often referred to as stick slip and other common examples of stick slip with similar noise-producing mechanisms are the following:
1. Moving a violin bow across a string.
2. Fingernails drawn across a blackboard.
3. Scraping paint.
4. Auto brake squeal.
5. Subway car wheel squeal on a curve.
Returning to tool screech, from extensive measurements on vertical turret lathes over a wide range at different speeds and feeds, it has been shown that the screech frequency and harmonics remain constant. In short, the spectral character of tool screech is virtually independent of speed and feed.
Feed is the term commonly used to describe the rate of tool advance; it is also a measure of chip thickness. Note that this conclusion is reasonable when one considers bowing a violin. For example, it does not matter how fast one moves a bow across a violin string, say an E string; the resultant frequency of the tone is always the same, an E and harmonics. The amplitude of the sound level, however, generally increases as one applies more normal force or bows harder, which is also often the case with lathe tool noise.
Another more analytical way of looking at the stick-slip phenomenon is to return to and consider again the damped vibration of a mass spring system.
In summary, the stick-slip mechanism can also be considered, for purposes of analysis, an example of regenerative or self-excited vibration due to negative damping.
With respect to noise control, little research has been expanded in these areas of tool squeal. This is probably due to the fact that the surface finish of parts is usually not adversely affected by the small amplitude of the vibration. However, on turret lathes where the tool squeal is a common problem, significant noise level reduction in the range of 8 to 10 dB has been achieved by wrapping the tool or tool holder with a dense vinyl or mastic.
From this, one can reasonably conclude that for this situation the tool holder was the dominant radiating surface. As such the contribution of noise radiated from the part being machined can be considered secondary. This may not be the case for large disks or cone-shaped parts.
There is also experimental data that show that lubrication plays a small role in the stick-slip phenomenon. This is probably due to the manner in which lubrication is applied. Typically the lubrication is of a water- soluble type, flushed over the tool. As such, little lubrication is actually at the critical tool-to-part contact and probably actually acts more as a coolant than a lubricant.
Finally, it should be emphasised that from studies tool squeal occurs most often while machining hard-to-machine nickel or tantalum alloy steels with machinability rating indices less than 20 per cent. Suggested areas of research into this little understood phenomenon include tool shape, tool rake and tool holder damping, etc.
Tool squeal should not be confused with tool chatter, which is usually due to poor tool design, dull tools, improper rake or chip thickness variations. Tool squeal associated with drilling operations is also common for hard-to-machine alloys. The noise mechanism is, however, probably stick slip in origin and again occurs almost exclusively on hard-to-machine nickel alloys.
Printing Pressroom Noise Control:
Noise levels in large, high-volume printing pressrooms typically exceed 90 dBA. The sources of noise are many, but general mechanical noise associated with gears, bearings, motor bearings, etc., is often the dominant source. As such, enclosure is often the only practical noise reduction measure. For large presses, adjacent workroom-like total enclosures have reduced operator exposure to acceptable risk of hearing loss criteria.
In addition, ancillary equipment such as folders and scrap ‘hoggers’ can be successfully enclosed. By installing enclosers resultant sound levels at the operator and inspection stations were below 90 dBA in actual installations. In addition to the enclosures, reverberation control utilising hanging ceiling baffles has provided measurable noise reduction at the periphery of the pressroom.