Practical Solutions to Machinery and Maintenance Vibration Problems
Chapter 5, Unbalance
Section 12, Unbalance Tolerances
This is a difficult subject for the writer to comment on since his experiences cover not only what qualified theoreticians and professors have written, but also the very practical world of performing the corrections needed to obtain the final results. He has managed two balancing shops; trained several dozens of balancing machine operators and performed field balancing, vibration analysis and correction on vibrating machines in many parts of the world. He has participated in two balancing standardization committees. One was for aircraft jet engine balancing and the other, ultimately, produced ISO 1940, "Balancing Quality of Rotating Bodies." (ISO refers to "International Standards Organization").
The experiences in technically-oriented standardization committees, combined with extensive practical experience, has resulted in a very subjective opinion that treats balancing and vibration standards as only an approximate guide, rather than standards that are absolute. For example, when we think of a government or international standard for a yard or a meter, we are conscious of a very precisely measured bar of special materials kept in a carefully temperature-controlled room at the Bureau of Standards. The word "standard" implies high accuracy and a certainty that the number involved is quite definite.
Yet, most standardization committees are composed of all types of human beings with varying backgrounds, some representing high academic approaches at top universities and others learning primarily through practical experience. Still others have personal and company reputations to protect.
At this writing, for example, balancing standards are given in oz•in/pound of rotor weight or the equivalent in metric units. The oz•in are for each "correction plane" (planes at which the unbalance is measured and corrected). Yet so far, there is no consideration given for the angular positions of the usual two unbalances, relative to each other, or the distance between the two correction planes.
If, for example, the "within tolerance" residual unbalances in each plane are in-phase with each other, they will add to each other to create additional static unbalance. This was kept in mind when the balancing quality grade was determined. However, the residual unbalances may end up with many other possible phase relationships with respect to each other. If they are approximately 180° to each other, they form a couple. If the distance between planes is small, the resulting couple will be small; if larger, the couple will be larger. If the distance is quite large, the resulting couple unbalance can be very large (producing considerably more vibration than when the two residual unbalances were in-phase). Nothing in the standards takes this into consideration. Therefore, the following section entitled "Unbalance Standards For New And Rebuilt Machinery" should be carefully considered by those concerned about obtaining the smoothest running machinery for maximum machine life.
Most new and rebuilt machines are properly, dynamically balanced to National and International Standards. However, those standards were developed over 25 years ago for economic and competitive conditions that no longer apply for today's plants. The vibration programs that produce the greatest financial returns require precision balancing to considerably closer tolerances than are provided by the above standards. As most suppliers of new and rebuilt machinery are unaware of the precision work performed at some plants (or think it's unnecessary and not economically worth it), it is best for a multi-plant company to publish its own company-wide balancing standards.For example, balancing to the upper limit of a pump or compressor rotor using present International Standards, would result in an orbit diameter (shaft centerline around its axis of rotation) that is almost seven times larger than the orbit that results from the new API (American Petroleum Institute) standards. Obviously, a pump that has a centerline orbit that is 1/7 the size of another pump, is going to run considerably longer before having to change bearings or seals.
Most suppliers will resist balancing to closer tolerances or will want to drastically increase the price for the precision balancing you require. For more acceptance, it usually requires an all-company standard and way of working with suppliers. Remember, only the balancing segment requires extra time to obtain the smaller orbit of precision balance. For example, a pump impeller, to be balanced to the API standard, requires only two or three more balancing runs than it took to reach G6.3 or G2.5. (Usually less than 20 percent more time for the balance operation only.) It may, for example, take one hour to balance a 200# armature to the ISO tolerances of G6.3 and only 20 minutes to a maximum of ½ hour more to produce the API's considerably smaller orbit that would drastically extend machine life. It may be necessary to encourage (or insist) that more training will establish a precision balancing routine at very little extra cost, even if your own company has to provide training for the supplier.
Note: If closer balancing tolerances are to be maintained, there must be equal attention given to the precautions needed to avoid assembly errors (see section "Unbalance Due To Assembly Errors").
This textbook contains only part of the information in our Practical Solutions seminar.