Tài liệu hạn chế xem trước, để xem đầy đủ mời bạn chọn Tải xuống
1
/ 47 trang
THÔNG TIN TÀI LIỆU
Thông tin cơ bản
Định dạng
Số trang
47
Dung lượng
3,37 MB
Nội dung
ODS & Modal Case Histories Barry T Cease Cease Industrial Consulting February 20th, 2009 ODS & MODAL CASE HISTORIES BARRY T CEASE, CEASE INDUSTRIAL CONSULTING FEBRUARY 20TH, 2009 INTRODUCTION What is ODS analysis and why do we need it? What is Modal analysis and why do we need it? When should either technique be used? Example of how to collect ODS & Modal data (test unit) CASE HISTORY#1 – ACCEPTANCE TESTING OF AHU FAN Equipment & problem description Route data, coastdown data & determination of “offending frequencies” Modal analysis of fan, motor & base Conclusions & recommendations CASE HISTORY#2 – ACCEPTANCE TESTING OF WATER PUMP Equipment & problem description Route data results versus standards & determination of “offending frequencies” ODS analysis of pump, step 1 (baseline) ODS analysis of pump, step 2 ODS analysis of pump, step 3 Conclusions & recommendations QUESTIONS & CREDITS “Modal Testing”, Robert J. Sayer, PE, Vibration Institute 31st Annual Meeting, June 19th, 2007 “Applied Modal & ODS Analysis”, James E. Berry, PE, 2004 “Machinery Vibration Analysis 3, Volume 2”, Vibration Institute, 1995 “Mechanical Vibrations, 2nd Edition”, Singiresu S. Rao, 1990 What Is ODS? • • • • • ODS stands for operating deflection shape ODS analysis generates a computer model of your machinery that depicts its motion while running at operating speed & load. You literally “see” how your machine is moving as it operates. This modeling can be extremely useful to illuminate an otherwise elusive solution to machinery vibration problems First, a CAD model of the machine or mechanical system is created (structure file) Second, detailed & meticulous vibration measurements are made on the machine typically during normal operation. These measurements consist of both the amplitude & phase of vibration at one or multiple frequencies of interest all referenced to a common point Finally, these field measurements are imposed on the model to generate visible animations of the model/machine at the distinct vibration frequencies of interest (typically the “offending frequencies”) What Is Modal Analysis? • • • • Modal analysis identifies the frequencies & shapes your machine “likes to vibrate at” (natural frequencies) and compares these to the normal forces present on the machine to see if a match exists that produces an undesirable resonant condition If a resonant condition is identified, common solutions involve the following: force reduction (ie: reducing the vibration forces present in the machine), tuning of the mechanical system (ie: adding or reducing mass or stiffness to the system at the right spots), or force “movement” (ie: changing the machine speed as possible to avoid the condition) The actual process of modal analysis is similar to that of ODS analysis except measurements are made while the machine is not running typically using a force hammer and one or more sensors. The hammer provides the input (force) and the sensor(s) measure the response (motion) at multiple points on the machine These modal measurements are then processed thru a technique known as curvefitting and then like ODS measurements, imposed on the model to produce animations that are analyzed Vibration Spectra vs Modal Data PLOT 1: Vibration data measured during normal operation Dominant vibration at 1,789 cpm or 1x RPM of machine (“offending frequency”) PLOT 2: Modal data measured while machine down Note the strong response at 1,837 cpm which is near 1x RPM When Should ODS or Modal Analysis Be Used? • • • When standard vibration analysis techniques have failed to determine the exact problem When resonance is suspected An ODS or Modal job begins best with a determination of the “offending frequencies of vibration” usually made using standard, route vibration spectra Example: Collecting ODS Data From CMS Test Rotor Kit • • • • • Machine operating Determine reference point (typically use route data point with strong vibration at all “offending frequencies”) First roving point collected at reference point (ie: 1Y:1Y) Continue collecting other points all along machine at predetermined points Both the total number of points collected as well as the point locations are key to how accurate the model animation will represent reality (ie: spatial aliasing). Example: Collecting Modal Data From CMS Test Rotor Kit • • • • • • Machine not operating Determine reference (driving) point. Like ODS analysis above, we want to use a point with strong vibration at all “offending frequencies”, but for modal analysis, we must be even more “picky” by applying the impact & measuring the response at many points until good representation of all offending frequencies is found (“driving point”) First roving point collected at driving point (ie: 1Y:1Y) Usually, we rove around with the sensor(s) and apply impact at the driving point, but this isn’t necessary. We could also rove around with the hammer with similar results although getting a good impact at all points is typically difficult Continue collecting other points all along machine at predetermined points Like ODS analysis, both the total number of points collected as well as the point locations are key to how accurate the model animation will represent reality (ie: spatial aliasing). Case History#1: Acceptance Testing Of AHU Fan Equipment & Problem Description • • • • • • Newly installed AHU Fan operating at medical facility Vibration acceptance testing required for all rotating equipment at facility Fan OEM contacted for vibration specifications - maximum acceptable vibration at 0.35 ips-pk Isolated, center-hung, centrifugal fan driven thru v-belts by a 4-pole induction motor operating on a variable speed drive Entire machine supported by 4-ea spring isolators mounted on floor arranged per diagram at right Two spring isolators are also mounted between the fan frame and wall to counter fan thrust 4-ea Floor Isolators 2-ea Wall Isolators Fan Motor INITIAL DATA & FINDINGS, PART • • • • • • Initial vibration data was collected on both fan & motor at 100% speed and overall levels were compared to OEM specifications Because this machine operated on a variable speed drive with normal operation anywhere between 50 and 100% full speed, coastdown data was collected between this speed range Unfortunately, this machine failed to stay within OEM specs both at 100% speed and at many points between 50 & 100% speed Maximum vibration levels occurred not at 100% speed, but at lower speeds suggesting possible resonance problems “Offending speeds/frequencies” were identified from coastdown data at approximately 1,500, 1,800 & 1,900 cpm Field observations noted the entire machine visibly “jumped” when the machine speed was set to 90-95% and motion at the motor outboard isolator seemed worst Vibration @ 100% Speed Maximum Vibration Level Fan Speed @ Max Vibration OEM Vibration Spec Motor, Outboard, Horizontal 1.289 n/a n/a 0.35 Motor, Outboard - Vertical 1.475 n/a n/a 0.35 Motor, Inboard - Horizontal 0.955 n/a n/a 0.35 Motor, Inboard - Vertical 1.027 n/a n/a 0.35 Motor, Inboard - Axial 1.205 n/a n/a 0.35 Fan, Inboard - Horizontal 1.929 3.11 1,903 0.35 Fan, Inboard - Vertical 0.605 0.45 1,495 0.35 Fan, Inboard - Axial 0.257 n/a n/a 0.35 Fan, Outboard - Horizontal 0.797 2.60 1,492 0.35 Fan, Outboard - Vertical 0.672 0.65 1,805 0.35 Fan, Outboard - Axial 0.258 n/a n/a 0.35 Measurement Point 10 ODS 10/2/08 – MOTION @ 3xRPM (10,800 cpm) • • • • • Pump maximum vibration at this frequency occurred at the pump, inboard, horizontal measurement (PIH) at 0.16 ips-pk Again, notice how the piping motion dwarfs that seen at either pump or motor Note how excessive motion of the recirculation line (1st Mode) is “pulling the pump” Note how excessive motion of the suction line is also “pulling the pump” Note the excessive motion in the short section of discharge piping between the recirculation line & pump discharge 33 ODS 10/2/08 – MOTION @ 4xRPM (14,400 cpm) • • • • Pump maximum vibration at this frequency occurred at the pump, inboard, horizontal measurement (PIH) at 0.36 ips-pk Again, notice how much more the piping is moving (vibrating) compared to either the pump or motor Note how motion of the recirculation line at this key frequency is by far the most and resembles a possible 2nd mode Note how motion at the suction piping remains high as well 34 ODS 10/2/08 – MOTION @ 8xRPM (28,800 cpm) • • • • Pump maximum vibration at this frequency occurred at the pump, inboard, axial measurement (PIA) at 0.05 ips-pk Again, notice how motion of the piping dwarfs that seen at either motor or pump Note the excessive motion of the discharge piping here Note how motion at the recirculation line is relatively small when compared to earlier frequencies 35 INSPECTION RESULTS 10/2/08 • • A close inspection of the piping found a broken discharge pipe hanger just above the horizontal pipe run in the ceiling It was unknown how long this hanger had been broken, but it’s absence no doubt added flexibility to the discharge piping run 36 CONCLUSIONS & RECOMMENDATIONS, 10/02/08 CONFIGURATION 1) 2) 3) 4) 5) Remove the recirculation line if possible. The addition of the new recirculation line has had a negative effect on machine vibration levels due to multiple suspected resonances occurring there Like the recirculation line above, motion of the discharge piping at multiple frequencies is having a negative effect on machine vibration. Add additional support to the discharge piping at the points where high motion is observed in the ODS animations If possible, try adding support from at least two additional points Repair or replace the broken discharge hanger found at the ceiling Provide additional support (if possible) under the suction piping as excessive motion continues here No soft foot records were identified from the alignment job performed since 9/16/08 on this machine. Please perform another soft foot and alignment check on this machine, make corrections as necessary and document results 37 PUMP & PIPING CONFIGURATION 10/16/08 1) 2) 3) 4) 5) 6) A new, larger standpipe was added under the suction piping for better support A new support was added to the discharge piping at the nearby wall A new stiffening “connection” was added between the discharge & suction piping above the recirculation line The recirculation line was not removed The broken discharge hanger was not repaired or replaced Pump soft foot corrections were made and documented. Machine alignment is documented to plant specs 38 OVERALL LEVELS, 10/16/08 • • • Both motor & pump overall vibration levels dropped significantly with the 10/16/08 modifications Motor overall vibration levels are now below plant spec at every measurement point Pump overall vibration levels are much better, but remain out of plant specs with highest levels being seen at the pump horizontal measurement 39 SPECTRAL DATA 10/16/08 • • • • Significantly reduced vibration levels at all offending frequencies is seen in all pump spectra Remaining vibration still occurring at 1x, 2x, 3x & 4x rpm (offending frequencies) Vibration at 4x rpm (vane-pass), although reduced, remains the dominant vibration frequency in most measurements Vibration at 3x rpm, although reduced, is the highest single source of vibration in all three pump measurements 40 ODS 10/16/08 – MOTION @ 1xRPM (3,590 cpm) • • • • • Pump maximum vibration at this frequency occurred at the pump, inboard, axial measurement (PIA) at 0.12 ips-pk Notice how motion of the piping is much greater than that observed at either the pump or motor Note how now both the discharge & suction piping are flexing in the axial (thrust) plane and are “pulling the pump” with them Notice how the motor is virtually “still” at this key frequency – any alignment or coupling problems are now unlikely here Both the addition of the new “stiffening connection” between suction & discharge as well as the continued existence of the recirculation line appear to have negative effects on machine vibration (transmission path) 41 ODS 10/16/08 – MOTION @ 2xRPM (7,180 cpm) • • • • Pump maximum vibration at this frequency occurred at the pump, inboard, horizontal measurement (PIH) at 0.05 ips-pk The newly installed discharge piping support at the wall was found both loose from the wall and the piping This looseness is at least partly to blame for the excessive motion of the discharge piping seen at or near the location of this new support Both the newly installed stiffening connection between discharge & suction piping as well as the recirculation line must be eliminated to reduce machine vibration levels 42 ODS 10/16/08 – MOTION @ 3xRPM (10,800 cpm) • • • • Pump maximum vibration at this frequency occurred at the pump, inboard, horizontal measurement (PIH) at 0.14 ips-pk Note the excessive horizontal motion of the discharge piping at this frequency with maximum deflection occurring somewhere between the recirculation line & discharge valve The recirculation line should be removed Horizontal bracing of the discharge line somewhere between the recirculation line & discharge valve may be necessary to eliminate this vibration. Only consider this modification after the glaring problems mentioned earlier are corrected and high vibration levels persist 43 ODS 10/16/08 – MOTION @ 4xRPM (14,400 cpm) • • • Pump maximum vibration at this frequency occurred at the pump, inboard, horizontal measurement (PIH) at 0.12 ips-pk Both the recirculation line as well as the newly installed “stiffening connection” continue their negative effects on machine vibration levels Again, notice how little both the motor & pump are moving when compared to the piping 44 ODS 10/16/08 – MOTION @ 8xRPM (28,800 cpm) • • • • Pump maximum vibration at this frequency occurred at the pump, inboard, axial measurement (PIA) at 0.02 ips-pk From the earlier spectral plots, vibration at this frequency is small when compared to the others Machine vibration levels at this frequency could be reduced by removing the “stiffening connection” between discharge & suction lines The suction pipe stand is vibrating excessively at this frequency in the horizontal direction 45 CONCLUSIONS & RECOMMENDATIONS, 10/16/08 CONFIGURATION 1) 2) 3) Remove the recirculation line & stiffening connection. Excessive motion (vibration) was seen at both the recirculation line & newly installed “stiffening connection” at many frequencies. Remove these two piping components to reduce machine vibration levels Tighten up the newly installed support between the discharge piping and wall Repair the broken discharge hanger located in the ceiling 46 QUESTIONS & CREDITS 1) “Modal Testing”, Robert J. Sayer, PE, Vibration Institute 31st Annual Meeting, June 19th, 2007 2) “Applied Modal & ODS Analysis”, James E. Berry, PE, 2004 3) “Machinery Vibration Analysis 3, Volume 2”, Vibration Institute, 1995 4) “Mechanical Vibrations, 2nd Edition”, Singiresu S. Rao, 1990 47 .. .ODS & MODAL CASE HISTORIES BARRY T CEASE, CEASE INDUSTRIAL CONSULTING FEBRUARY 20TH, 2009 INTRODUCTION What is ODS analysis and why do we need it? What is Modal analysis and why do we need it?... Example of how to collect ODS & Modal data (test unit) CASE HISTORY#1 – ACCEPTANCE TESTING OF AHU FAN Equipment & problem description Route data, coastdown data & determination of “offending frequencies” Modal analysis of fan, motor & base... Route data results versus standards & determination of “offending frequencies” ODS analysis of pump, step 1 (baseline) ODS analysis of pump, step 2 ODS analysis of pump, step 3 Conclusions & recommendations QUESTIONS & CREDITS Modal Testing”, Robert J. Sayer, PE, Vibration Institute 31st Annual Meeting, June 19th, 2007