Roger replies: This seems like as good a time as any to start a thread on vibration. It's a problem that all metal workers have to deal with, and it's a subject that I didn't have much of a "feel" for until I looked into it for awhile and read a few books. In fact, I think that it is so important to have a basic understanding of what vibration is, that I propose write a couple of short primers on the subject. I'll start with "Natural Frequency" and then do one one "Driven Vibrations" and finally one on "Vibration Control"....The emphasis will be on simple motor-type vibration, Natural Frequency of Vibration: Taking that motor that you asked about, yours is a common problem in metalworking machinery of any kind. Basically, you need to use a motor to turn something and when you do, either the whole machine - or some part of it - begins to shake. There are a couple different reasons for it to vibrate and the first place to start is by realizing that the whole lathe, as well as every single piece on it, has it's own natural frequency of vibration. You will also hear this called the "resonant frequency". As an example, suppose you were to dangle that motor from a string and then whack it with a hammer. I expect it would ring like a bell. The frequency (how many times a second something vibrates) of that ringing bell, or motor, is called it's "natural frequency. Everything has a natural frequency: bells, buildings, musical instrumentss, automobile suspensions, drums, buckets of water.....even the Earth has a natural frequency and will ring like a bell when struck. You can find the natural frequency of any object just by striking it and measuring the time period of the pulsations as it vibrates. The actual number that describes the resonant frequency is measured in Hertz and is defined as cycles per second. Think of it as counting how many times something happened and then dividing by how long a period of time it took to count them. Frequency is determined simply by how stiff something is versus how massive it is. The stiffer the object, the higher the natural frequency. The more massive the object is, the lower the frequency. In this way, a small bell will have a higher frequency than a large one because it is less massive. But if you were to make the small bell out of something that is not very stiff - like plastic - and make the large bell out of something that is very stiff; like bronze, or steel or even carbide, then you just might be able to make the large bell have a higher tone than the small one. Again, this applies to everything. Bells, boring bars, whole lathes, and even the pitch of a person's voice. Just so that the whole subject doesn't seem too simple, please let me introduce you to one of the more basic complications: an object will usually vibrate in more than one simple direction...and each different direction will have it's own natural frequency. An example of this would be a nail that is struck squarely on it's end. It not only vibrates from end to end, getting minutely longer and shorter as it vibrates, but it also vibrates in a radial direction so that the diameter also pulses. These are called the longitudinal mode and the radial or bell mode respectively. Since the nail is much stiffer radially than along it's long axis (stiffer means it takes more force to bend it that way)...then the natural frequency of the bell mode of our nail is much higher than the natural frequency of it's longitudinal mode. To add to the complications, these two modes are not the only ones possible, and objects also tend to vibrate at multiples of their basic frequencies (overtones). There are many directions in which an object can vibrate. Luckily, one of these directions is usually much preferred to the others and so we can build a bell (or nail!) with a nearly pure tone. So far we have used a single sharp hammer blow to start an object to vibrating at it's natural resonant frequency. Next we can talk about driving a vibration with repeated impulses. Roger Loving Vibrations - what drives them? Thanks to all of you for your encouragement on this subject! For awhile there I was worried that maybe I was just wasting bandwidth, but there seems to be a lot of interest after all. I'll start out by quickly reviewing and answering a couple of questions that have been asked. In the last primer I rambled on about how everything has a natural frequency of vibration, and that this natural frequency is higher for stiff materials and lower for massive ones. BTW, mass is mass; no other way around it. But we have two ways we can increase the mass of anything: we can either use more of the same material, which will make the lathe heavier, or we can switch to a material that is more dense which will also makes the lathe heavier! This is because as long as we are machining on the surface of the Earth we use the terms "mass" and "weight" pretty much interchangeably - which is real convenient for us, even if it does give the eggheads fits. Along with two ways to increase mass, we also have two ways to increase stiffness. And they are just as simple - although a little less obvious. The first one is just like we did for mass: switch to a different material. If we need our copper bell to ring with a higher pitch we use a stiffer material like steel instead of copper. The less obvious way to make an item stiffer is through geometry. As an example of this, a pound of steel formed into an "I" beam is stiffer than a pound of 18 guage steel as sheet metal. Think about that for a minute...Everything about the "I" beam and the sheet metal is the same except for the shape! We also discussed how it was that a single object could vibrate in different directions with a different frequency in each direction. And that the overall vibration of any object is simply the sum of all of the vibrations added together at the same time. Hmmmm, I think it's easier to refer to each the discrete types of vibration as a "mode"....or "vibrational mode"....it's all the same thing. To get back to what we were doing, if we sum all the different modes then this sum becomes the mode of the complete object. Even in the case of something that is a pure as a tuning fork, the total vibrational mode of the object is rich and complex. Now we are ready to drive those vibrations. One way you can get a vibration is with a sudden blow. Bells, anvils, and thunderclaps all work this way. What you get in that case is a a real complex vibration that has most of it's energy at the natural frequency of whatever it was that was struck. Vibrations driven by the energy of a single blow usually go away pretty quickly - which is lucky, since we can't do much to prevent them anyway. The more troublesome vibrations are the ones that are driven by a rythmical impulse....that is, some sort of repeated push. Examples of this would be a tire out of round, or a motor that is out of balance, a milling cutter with a chipped cutting lip, or any interrupted lathe cut. These examples all involve things that are rotating. It isn't necessary to use rotation to get a repeated push, but it is the most common way. Vibrations caused in this manner are called "driven vibrations" and they create a very powerful vibration AT THE DRIVING FREQUENCY. The "driving frequency" being how many times a revolution that the impulse occurs times the number of revolutions per second. Of course each one of these impulses is like a sudden hammer blow itself, so our machine is now shaking at it's own set of natural frequencies as well as at the driving frequency! The largest vibes of all occur when the frequency of the driving vibration matches the natural frequency of our object. This is called "resonance", and can literally build up until it shakes the object apart. A type of resonance also occurs when either the natural or the driving frequency is close to some natural multiple of each others frequency. Designers go to great lengths to make sure that resonance can't easily occur. Let me show you an example of how hard this is to do. Suppose you are designing a lathe and of course you are concerned about vibration. First you carefully balance the motor that drives your lathe. Then you do the same to every rotating part, not forgetting all of the gears as well as truing the shafts that they run on. Pretty expensive lathe at this point, but we're not done yet. We calculate the natural frequency of that lathe in every direction and make it real stiff so that the natural frequency is high enough that we could never rotate anything fast enough to excite the natural frequency of the lathe. But just in case that we do, we now add lots of mass so that any vibration that does sneak through will have to move so much metal that the total shaking is very low. Lastly, we make the entire machine out of cast iron because it has the capacity to damp out those vibrations quickly (more about damping later). Well, we have designed a really expensive lathe and yet there are still some sources of vibration that we just can't touch. Take that motor for example...the same motor that we balanced so carefully and at such an expense. That motor has a certain number of fields inside of it and it rotates because each one of those fields gives the armature a "kick" as it goes by. Suppose our motor has 10 fields in it and naturally runs at 28 rotations per second (about 1700 RPM). Our million dollar lathe will have a vibration at a frequency of 28 X 10 = 280 cycles per second (Hertz). And since it is an expensive lathe and we put a stout motor on it, it has plenty of energy left over to vibrate at the natural multiples of the motor frequency. We have particular problems at 560 Hz., 1120 Hz., and on up the frequency spectrum until hopefully the motor vibrational frequencys are pretty weak by the time we get to the natural frequency of our very stiff lathe. If this hasn't ruined the designer's day, he then discovers that even the best tapered roller bearing has it's own natural frequency...as well as having a certain number of rollers that have to rotate right along with the shafts. Since a bearing can't be totally tight or it won't spin at all, you can just bet that there are frequencies that will make those rollers chatter like an old dog's teeth on a frosty morning. Owners of machines with continuously variable speed control can usually find a speed that will make the machine jump up and down like it was rubber mounted. We also want to give some thought to the toolpost, cutter, and the stock that is being turned. That toolpost that is nice and rigid and is holding a cutter out in the air looks very much like a tuning fork to me. As the stock we are turning begins to rotate and impinge on the cutting tool we have a nearly perfect source of driven vibrations. If we really want to set up a vibration we can start with a piece of hex stock and turn it down to round.(BG) Almost as good a vibrator as the hex stock is when we machine something really thin - like a piece of thin wall pipe. The driving vibrations are generated by the rubbing (cutting) of the the pipe on the lathe tool, causing the tool and toolholder to shake at their natural frequency. You know that this is happening because you can hear it as a ringing or "singing" of the part..In turn, this vibration of the cutting tool marks it's own distinctive pattern across the surface being machined. Thes marks are like a fingerprint and will contain a wave-picture of all of the driven and natural vibrations. That ought to get us started on ways that vibrations are started. one of the key thoughts here is that it is very difficult if not impossible to keep a vibration from happening. About the best we can do is to plan to keep well away from resonance between driving and natural vibrations, and to find some way of defeating the vibes that do occur. Defeating the vibrations will be the next subject, and I plan to look at damping - both with an elastic material and with several types of "tuned dampers". I also got a question on the how to determine the damping coefficients of various metals. You can look this sort of thing up in engineering handbooks, but there must be a better way...some sort of very simple test... but offhand I can't think of what it would be. I'll try to post the next primer after the long weekend holiday. Enjoy ! Roger Loving The opposite of driving a vibration is damping it out. This takes us directly to the next topic, which is: Now that we have a pretty good under- standing of vibrations of what vibrations are, what do we do about them? All vibrations eventually wind themselves down. They do this because of damping. And the more damping that we add to a system the quicker the vibrations will quit. In fact, if we add enough damping, the vibrations will never even get started. An inevitable - but not very effective - form of damping is what takes place as a tuning fork gradually loses it's motion. In that case, the motion of the tuning fork is gradually converted to heat as the molecules in the fork slide past one another. When the fork finally quits moving, it is just a little warmer than it was when it started. In case you are wondering, the frequency of the tuning fork never changes! It's motion just keeps getting smaller at the very same frequency until you can't detect it anymore.Note: rubber/liquid damping, and counter vibrations: a tuned damper, and two identical lathes running opposite to one another.Keep tools short and stiff to raise the natural freq. of the toolpost system. __________________________________________________________________________________ > >You'd get "gear marks" on your work from a gear-driven lathe spindle??? > Not necessarily, but it's possible. If you look closely enough at the turned surface you can find wavy marks from every source of vibration that occurs while machining. By measuring the wavy lines (gearmarks) and back-calculating from the spindle RPM you can even figure out what caused some of these chatter marks. Think of Edison's first phonograph.....same principle. When doing some extreme micro-precision work of hard disks I was able to match one periodic vibration to a train which went buy every morning about a quarter mile away. Another -found by my R&D partner- had to do with morning sun shining through a set of windows....each time the sun came through the next set of window blinds it would shine on the tool for a few minutes and the slight heating was enough to make a very small but measureable difference. Not only can one find gear marks, but also things like the number of balls in a spindle ball bearing....each type of ball bearing has it's particular signature. These are the kinds of things which establish the limits for accuracy. They are real, but not the sort of thing to worry about. If I really wanted to see gear tooth marks I would consider that the gear teeth were like little hammers and figure out what frequency they were hammering at. Then I would set the tool post to just that frequency (resonance) by considering that it was a vibrating tuning fork (formulars are available). Then I would run a spindle RPM that was just "right to be wrong"......I'd bet I could get the gears to write their own signature that way Roger Loving __________________________________________________________________________________