Introduction. Current Techniques

Transcription

1 SIMPLIFIED FIELD BALANCING One Correction Plane Or Two? An Application Paper for using CSI FAST Bal II & Ultraspec ICM Balancing Programs By: Edwin B. Lindskog Introduction This paper discusses a simple but often confusing aspect in the procedures utilized during field balancing. Often the question is asked at the start of a new balancing job: Can this machine be successfully balanced with just one correction plane, or will it be necessary to apply the use of two separate correction planes? The correct answer can sometimes be difficult to determine. Current Techniques Field balancing technicians have adapted their own guidelines or rules of thumb that apply when considering whether to perform single or two plane balancing. Typical of these is the simple diameter to width ratio rule. Usually it suggests that if the diameter of the rotor is greater than 4 times the width, perform single plane balancing. If the diameter is less than 4 times the width, perform two plane balancing. This 4:1 ratio rule has been generally regarded as more of a basic guideline. The 4:1 ratio rule has been proven unsuitable on many occasions. Often this results in the balancing technician starting the job over, after a successful correction or two, followed by many unsuccessful trim runs that typically improve vibration on one bearing, while increasing the vibration on the other. After a significant amount of balance runs and time has passed by, the decision is often made to compromise and accept higher vibration limits. Alternatively, the job may have to be started over again, this time defining two correction planes in place of only one. Individuals who wish to avoid the above potential frustration will simply set up most all their field balancing jobs utilizing two correction planes. This procedure however requires that an additional trial weight run be performed at additional cost. The decision to do this is not based on any specific proof or need of this additional correction plane at the time. More seasoned analyst/balancing technicians may use an alternative method to determine the number of required correction planes. The technician reviews the 1xRPM phase and amplitude data acquired at the machinery vibration measurement points. Calculation and/or graphical analysis of the data then helps to determine the required number of correction planes needed for the job. This method is perhaps one of the more accurate methods because it accounts for the specific machinery data that is likely attributed by the unbalance characteristics of the machine. However, the problem in this last method is in computing the phase and amplitude data for 4 or potentially more vibration measurement points. This is often not performed to the high degree of accuracy required. 1

2 ICM Balancing Programs Some ICM (Influence Coefficient Method) balancing programs have the ability to provide the user with estimated or predicted residual vibration data. This information can be reviewed before any calculated balance weights are installed. Because of this ability, ICM programs with this feature can aide the user in determining whether one or two (or more) correction planes are required to perform a satisfactory balance job. Many are not aware of this feature, or the powerful and useful information it provides to the balancing technician. You need to know The need to know whether the particular balance job requires one or two (or perhaps more) correction planes can prove most beneficial to the process of performing successful field balancing sessions. Specific unbalance characteristics and rotor stiffness play a major role in determining the number of correction planes needed to perform a successful balancing job. The analyzer with the ICM program has the ability to compute the specific machinery vibration data and provide the user with estimated residual vibration data. The user then views this information and decides if these predicted residual values would be acceptable for this machine. If not, the user can then define an additional correction plane(s) and continue on with the next trial weight run without loosing any previous balancing data or time. Consider the following: 1) Many large pieces of rotating equipment can not be readily started and stopped without a considerable amount of time between balance runs. Knowing that the job could be successfully performed utilizing only one correction plane would save a substantial amount of time in the completion of the balancing job. 2) The time it takes to perform the balancing procedure often means a substantial loss in production for many facilities. This lost time may result in severe financial penalties and lost revenues. Each unnecessary balance run increases downtime and lost production in many cases. 3) Access to the outboard side of many airhandler units as well as other pieces of equipment is often difficult or almost impossible without disassembling part of the machinery cabinet or cutting access holes. These areas also may or may not contain certain hazardous environmental conditions making it unsuitable or undesirable for human presence at the time. The more easily accessed correction plane on the rotor may provide the ability to reduce vibration in both monitoring planes to an acceptable level. Knowing if this were possible prior to the attachment of any balance weights would reduce much time, man power, and possibly resolve some potential safety issues. 4) A balance job that requires the use of two correction planes will have limited success if the user has defined only one correction plane. A couple unbalance characteristic cannot be corrected by placement of correction weight on only one plane of the rotor. Many compromises are ultimately made because the technician could not reduce the vibration in one monitoring plane without increasing the vibration in an other. A second defined correction plane would have avoided these compromises. 5) Somebody s reputation is at stake. The customer or department leaders are looking to the balancing technician to provide a precise and relatively quick turn around time in curing the vibration problem on their troubled piece of machinery. If there is an obvious competency 2

3 problem, or if the vibration is unable to be reduced to an acceptable level, they are likely going to take note of this. Consideration as to who performs the next balance job will be carefully thought over. Other conditions may prove beneficial in usage of this recommend procedure. Two Identical Machines, Two very different balance jobs. Two examples are presented here that show two different situations, even though the fans (Forced Draft Fans #1 and #2) are of identical construction. The #1 F.D. Fan requires the use of only one correction plane to achieve acceptable results during the balancing process. The #2 F.D. Fan is set-up to utilize only one correction plane at first, just like the #1 F.D. Fan. The Estimated Residual Vibration data viewed in the ICM Balancing Program indicates that one correction plane will not reduce the vibration amplitude within the specified tolerances defined. The example continues on to show how to define the second balance plane and complete the job. as an overhung blower, the following rule applies: If the highest 1xRPM readings are on the drive side (inboard or M1 ) bearing, place the #1 trial weight on the inlet side of the fan wheel. If the 1xRPM vibration is highest on the fan side (outboard or M2 ) bearing, place the #1 trial weight on the backing plate side of the fan wheel. In either case, this becomes the #1 Correction Plane, or C1 (see Figure B). 3) Reduce the chances for human error. Make it a habit to physically label the correction planes on the machine, C1 and C2 with some type of marker. It is also a good idea to label the angular location of the trial weight locations on the rotor, T1 and T2. The time it takes in labeling these locations is minimal and may help you to avoid some costly mistakes and confusion later on. Here are some basic ground rules that should to be applied during these balancing procedures. 1) Always take two radial measurements per bearing, each separated by approximately 90 degrees on the component you are attempting to balance. For a horizontal fan, this would mean one vertical and one horizontal reading on both the inboard and outboard bearing locations. 2) Unless circumstances don t permit it, always place the #1 trial weight on the correction plane closest to the bearing that has the highest 1xRPM data. This becomes Correction Plane #1 or C1, (see Figure A). Exception, if the machine is of an overhung configuration such Figure A 3

4 Start of Balance Job, F.D. #1 Fan Figure B 1) Define the balance job as usual. #1 Forced Draft Fan The 1xRPM vibration amplitudes and phase readings were acquired on both the inboard (drive side) and outboard (fan side) bearings in the horizontal and vertical directions. The M1 and M2 signify the measurement planes. The C1 signifies the #1 correction plane. This correction plane was defined when the #1 trail weight, T1 was placed on the fan. This plane was chosen because the 1xRPM vibration was highest on the inboard (fan side) bearing (see figure C). Remember the previously mentioned rule about placement of trial weights? 2) Enter the following information: 1 Weight Plane, 2 Measurement Planes, and 4 Measurement Points. Enter the Tach, Sensor, and measurement Point information on the remaining set-up screens as usual. 3) Take the Reference Run data. Figure C 4

5 4) Here is the Reference Run data for this machine. 7) Here is the Trial Run 1 data for this machine. 5) Take the Trial Run 1 data. Note: the trial weight is placed on the inlet side of the fan wheel because the 1xRPM vibration is highest on the inboard (drive side) bearing. This plane has now been defined as C1. 8) Proceed to the Tolerance Check/Trim screen after viewing the information in the Correction Weights screen. 9) Press the Enter key for Applied Weight. 6) Enter the trial weight amount and the angular location in the analyzer as usual. 5

6 12) This is the final data for the finished balance job. All measurement points are now well below in/sec peak. Also note the phase and amplitude similarities between the predicted data and the final data for this job. 10) This screen asks for the actual correction weight(s) and location(s). 1 st press the Page key to review the Estimated Residual Vibration information (see the note in upper right side of the screen). 11) Here is the Estimated Residual Vibration data for this machine. Note that all measurement points show predicted vibration levels well below the defined tolerance of in/sec peak. This data indicates this machine can be balanced by utilizing only one correction plane. Proceed to install the recommended correction weights in the proper location(s). Remember to place the weights in the defined correction plane, C1. This job was very successful in utilizing only one correction plane. It managed to reduce the vibration in all measurement planes to acceptable vibration levels. Two plane balancing doesn t require two correction planes in this instance. #2 Forced Draft Fan The 1xRPM vibration amplitudes and phase readings were acquired on both the inboard (drive side) and outboard (fan side) bearings in the horizontal and vertical directions. The M1 and M2 signify the measurement planes. The C1 signifies the #1 correction plane. The C2 signifies the #2 correction plane. The C1 correction plane was defined when that the #1 trail weight T1 was placed on the fan. While, the C2 correction plane was defined when the #2 trial weight T2 was placed on the fan (see Figure D). The machine was initially set-up the same way in the analyzer as the #1 F.D. Fan. It was later determined after viewing the Estimated Residual Vibration data that this fan couldn t be balanced within tolerance with the use of one correction plane. A second correction plane was then defined. None of the previous balancing data was lost in this process of defining the second correction plane. It only took a moment of time to determine this situation and make the needed changes in the program. Figure D Start of Balance Job, F.D. #2 Fan 6

7 4) Here is the Reference Run data for this machine. 1) Define the balance job as usual. 2) Enter the following information: 1 Weight Plane, 2 Measurement Planes, and 4 Measurement Points. Enter the Tach, Sensor, and Measurement Point information on the remaining set-up screens as usual. 5) Take the Trial Run 1 data. Note: the trial weight is placed on the inlet side of the fan wheel because the 1xRPM vibration is highest on the inboard (drive side) bearing. This plane has now been defined as C1. 6) Enter the trial weight amount and the angular location in the analyzer as usual. 3) Take the Reference Run data. 7

8 information (see the note in upper right side of the screen). 7) Here is the Trial Run 1 data for this machine. 8) Proceed to the Tolerance Check/Trim screen after viewing the information under Correction Weights screen. 11) Here is the Estimated Residual Vibration data for this machine. Note that some measurement points show predicted vibration levels above the defined tolerance of in/sec peak. This data indicates this machine can t be balanced by utilizing only one correction plane. A second correction plane must be defined. 12) Go back to the Job Definition screen on the main menu. 9) Press the Enter key for Applied Weight. 10) This screen asks for the actual correction weight(s) and location(s). 1 st press the Page key to review the Estimated Residual Vibration 13) Change the number of weight planes from one, to two weight planes. Press the Enter key. 8

9 14) The following message will occur. Disregard this warning and press the Enter key again to continue. Do not press the Main key to abort. Press the Enter Key on the remaining screens and proceed to the Make Measurement screen. 17) Here is the Trial Run 2 data for this machine. 18) Proceed to the Tolerance Check/Trim screen after viewing the information in the Correction Weights screen. 15) Proceed to the Trail Run 2 screen. All information in the Reference Run and Trail Run 1 screens have been retained. 19) Press the Enter key for Applied Weight. 16) Take the Trial Run 2 data. Enter the information for the trial weight #2. Note: this trial weight T2 is placed on the backing plate side of the fan wheel. This plane has now been defined as the Correction Plane #2, or C2. 20) This screen asks for the actual 9

10 correction weight(s) and location(s). 1 st press the Page key to review the Although this procedure for determining the Estimated Residual Vibration number of correction planes appears to be a bit information (see the note in upper tedious on paper, it is actually quite easy and is right side of the screen). quickly performed in real life situations. Please give it a try. There is no doubt that there is some serious value to utilizing this procedure, not only in time and revenue, but in improved machinery performance and life as well. 21) Here is the Estimated Residual Vibration data for this machine utilizing the second correction plane. Note that all measurement points now show predicted vibration levels well below the defined tolerance of in/sec peak. Proceed to install the recommended correction weights in their proper locations. Remember to place the weights for the correction plane #1 in the correction plane marked C1, and the weights for plane #2 in the correction plane marked C2. These examples were shown utilizing the CSI Ultraspec Balance program in the Standard Mode setting (this mode must be used for this procedure). The FAST Bal II program from CSI is virtually the same program and works the same way. The CSI, FAST Bal I program is a different ICM program and does not allow the capability to apply these procedures. Other non CSI Influence Coefficient Method balancing programs can or may be functioned in a similar fashion as explained here. There is at least 1 TEC-IRD ICM program that provides similar useful information. Please remember, two plane balancing doesn t necessarily mean two correction planes are needed to perform a successful field balance job. 22) This is the final data for the finished balance job. All measurement points are now well below the in/sec peak tolerance defined. Summary 10

11 11