It is sometimes, but not always possible, to significantly reduce the amount of noise and vibration emanating from a machine tool and make it run more smoothly. e.g. a rumbling noise from the headstock might not be caused by worn roller bearings (or gears) and noises from the motor or drive system (or other parts) may not originate from where they appear to be coming. Causes are varied and sometimes, unfortunately combine to make the job of tracking them down even more difficult. Even if the machine is useable, it's often the case that vibrations are transmitted to the workpiece in the form of a wavy or uneven finish (though this is distinct from that old bugbear "tool chatter", a quite different matter).

Well worn machines are more likely to vibrate than new ones. An accumulation of wear in bearings, bushes, motor shafts, belts and gear teeth may result in noise and vibration that cannot be eliminated by other than a complete rebuild.

A light machine not bolted securely to its stand or the floor will obviously tend to vibrate more than one correctly aligned and fastened down. Even very heavy machine tools have been known to "walk across the floor" when a cyclic vibration builds up out of control.

Workpieces:
Any rotating workpiece that is not perfectly cylindrical will set up an imbalance - and hence vibration. However, even with what appears to be a perfectly symmetrical job it might be that it has a "heavy spot" caused by material being denser in one part than the rest (called "inhomogeneities") or even one hole drilled near its periphery. Obviously, the faster the rotational speed the greater the effect.

Motor, pulleys and associated areas
Single-phase motors are inherently vibration prone and mounting one (where possible) on rubber blocks - or one those handy rubber studs with a thread at each end as used on some car exhaust systems - can work wonders.

If the machine is making a "cyclic" vibration or noise - rising and falling in volume and pitch - this may be caused by worn V-belts, or by badly jointed flat belts of inconsistent width and/or thickness. A good test in this case is to run the machine at a middle speed for 30 minutes to warm it up thoroughly and see if the consequent heating of the belts softens them enough to ease (if not eliminate) the problem.

Worn V-belts - as the worn and unworn sections arrive at a pulley they alternatively drop into and ride up the pulley grooves so raising and lowering gear ratio-  and hence the speed of rotation. These conditions are ideal for setting up sympathetic vibrations that can have many undesirable results - from tool marks on the work to a severe shaking of the whole machine that causes mechanical damage. Fitting new belts - especially new T-link segment belts which we can supply - can often result in a transformed machine.
V-belts are normal sized by the width across the top viz:
Z Section 10 mm
A Section (very common) 13 mm
B Section 17 mm
C Section 22 mm

As worn flat belts can cause the same problem, new ones can be supplied in a custom-made form to your exact requirements. (phone for help if required 01298-871633 from overseas: +44-1298-871633

Belt tension: belts must not be too tight. Try slackening slightly, a little at a time, to see what happens.

Twin or multiple V-belt drives: these can be a notorious cause of noise and vibration (especially with 2 belts). The belts must be matched exactly for length. Check the installation by first running with no belts, then just one before fitting the second or third, etc.

A link belt that is running in a pulley insufficiently deep allowing the link pins to hit the bottom of the groove

A link belt with de-laminated links flapping

A link belt with an inserted section of another make - or a newer or older section let in (difficult to spot)

Loose motor-mounting nuts and bolts - or a slack motor-mounting platform.

Loose or damaged motor-cooling fan or fan cowling

Belt and other covers touching the motor or motor mounting area - gently pressing against any suspect covers and listening for any change of note might show which are at fault

Loose, broken-up or misaligned motor rubber resilient mountings. The rubbers are often pressed over a hexagon boss on the end of the motor and retained by a two-part strap over the top.

Loose stand doors or air-trunking from the motor

Loose or misaligned motor or spindle pulley (very common). The pulley should be keyed to the shaft and locked by one or more grub screws on top. It's not uncommon to discover that the alignment of the keyway in the motor shaft and pulley do not coincide properly. The grub screw should bear down firmly on the key - which itself should be seated correctly in its slot. Check that the key is not so far back that it is partially lifted by the radiused end of the keyway.

Worn or damaged motor bearings. Ball bearings are especially susceptible to shock loadings that indent the tracks. Often caused by heavy-handed methods used to remove or fit pulleys; to check remove the belts and feel for up and down play and rough turning when spun by hand.

Electric motors are often out of balance from new (few have the necessary dynamically balanced parts for true, smooth running). Although this slight imbalance might not matter on a new lathe - the vibration is absorbed by the relative tightness of everything else - once the machine is well used and tolerances slackened, this can cause problems.

Motor and driven pulleys running out of line - laterally as well as vertically. Check with a steel straight edge. This is especially important with flat belts

Out-of-balance pulleys - not generally a problem with ones in aluminium but it has been known for larger cast-iron pulleys to have been poorly manufactured. To see if a pulley is out of balance arrange for it to be spun by hand between centres and mark the lowest point. Spin again and, if the same point is at the bottom, that where metal can be removed by filing of drilling a small hole. Repeat the process - spinning, removing a little metal - until the pulley stops at random. When it does, it's balanced

Just changing the motor can sometimes make a difference. A particular motor can get into a sympathetic vibration with some other part of the machine - whilst another (even of the same make and type) has been known to run smoothly.

Other possibilities:
Changewheels (the drive to the leadscrew) should have equal amounts of backlash. Some gears tightly in mesh and some loose can cause problems..

Worn or incorrectly adjusted spindle bearings - this may also have damaged the spindle. Roller bearings often appear "as new" when inspected by the naked eye. However, using a strong magnifying glass might reveal what looks like black spots on the bearing tracks--this is a sign that the hardness is breaking up and as the rollers pass over them vibrations can be induced.

Incorrect spindle bearings for the application

Spindle bearings damaged due to incorrect fitting technique

Wrong end loading on the spindle bearings - so many different types are used that reference to the maker's instructions might be the only way to discover what to do. However, most standard taper-roller types with no separate thrust race only need a very slight preloading to eliminate end play - typically 1-foot lb (1.4 Newton-metres) or thereabouts as specified, for example, on the Boxford lathe. If in doubt, start low and work up...

Loose, worn or broken keys in the pulleys and gears running on the main headstock spindle shaft

It may be that the machine has always been noisy and vibrated badly - something caused by an out-of-balance motor armature. This is not as uncommon as might be thought - and one reason why some grinding machine makers (where smooth running is very important) will only fit motors whose armatures have been dynamically balanced.