Applied Vibration Analysis: Analyzing Pump Vibrations

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
[Music] step one familiarize yourself with the equipment this particular pump is Center hung between two bearings standing close to it you can tell the pump is operating very smoothly it is still cool and quiet it turns relatively slowly and the impeller has 8 vanes step 2 calculate important speeds and frequencies the one times turning speed of the shaft is always the first speed you'll want to determine in this case the shaft is turning at 9.14 Hertz the signature vibration we would expect from a pump is a vane pass frequency coming from the impeller since this impeller has 8 vanes we would expect a vibration frequency at eight times the turning speed of the shaft which equals 73 point one eight Hertz step three locate the one times in the spectral data let's look first at a spectrum taken up the outboard bearing and don't let the size of these Peaks fool you look at the amplitude scale it only reaches to point zero zero eight inches per second of velocity these are very very small vibrations we're seeing the graph has been scaled to make the Peaks more visible by moving the cursor around we find a peak at nine point one four Hertz and we have an exact match for the calculated one times step four identify signature vibration patterns the signature vibration we're looking for is the vane pass at seventy three point one eight Hertz there are several significant peaks to the right of the one x and with a little searching we can find an exact match at seventy-three point one eight Hertz step 5 identify other vibrations present when we mark the one x with the cursor the analysis software identified several harmonics this would suggest shaft movement but remember these are tiny low amplitude vibrations well within specs we also find a family of Peaks around 500 Hertz now there are a couple of things this could be first it could be that the bearing is breaking through its lubrication and we are seeing rub this seems unlikely however because the amplitude is so low second this pattern is often associated with pump cavitation exactly what is and what is not cavitation is a little complicated but cavitation involves bubbles in the liquid supply that explode on contact with the impeller long-term cavitation will leave pits in the impeller vanes cavitation can appear almost anywhere on the spectrum and it often takes on the shape of a mound much like this one however this would be a surprisingly low amplitude for cavitation let's zoom in and look at the peaks more closely they appear poorly formed and random measuring the distance between the peaks does not reveal a significant frequency such as the one x or vane pass frequency and the distance is not uniform between different pairs of Peaks if we look at the time domain we find low amplitude Peaks that are recurring but not entirely uniform if we measure the distance between two peaks we find a delta time of 1.9 5 3 milliseconds which converts to point zero zero one nine five seconds rounded to convert time to frequency take the reciprocal one divided by point zero zero one nine five equals 512 point eight two Hertz a number that certainly seems consistent with the peaks around 500 Hertz in the frequency domain these vibrations are real they're just very small this probably is a process vibration the pump is running so smoothly the vibration probe may be picking up traveling frequencies through the pedestal or it may be that we're seeing minor surges in the liquid supply now let's look at a spectrum taken at the inboard bearing the one x is at a higher amplitude than in the previous spectrum so there may in fact be a little shaft movement but amplitude is still very low and the one x harmonics have disappeared so the problem if we can call it a problem is very minor the vane pass frequency is also at a low amplitude and does not generate harmonics so let's look at the family of Peaks between 300 and 400 Hertz the pattern looks similar to a gear mesh frequency but there are no gears in this pump so what might this be well sometimes you get lucky and you catch a bearing defect in its infancy before the fault frequencies were used to seeing conform if you find a family of regularly spaced peaks at a relatively high frequency measure the space between the peaks in this case we find a delta frequency of nine point one five one Hertz which very closely approximates the one x normally this finding would tend to confirm a bearing problem but perhaps not in this case there is something else here notice that the tallest peak is found at three hundred fifty nine point seven five Hertz or very nearly 360 after you've accumulated a little experience you'll recognize this pattern right away anytime you find a peak at 36 Hertz and sidebands of turning speed look for a DC motor silicon-controlled rectifiers known as scr s in the motor control center convert the incoming AC voltage to DC voltage by performing a half wave rectification since AC voltage in North America is supplied at 60 Hertz and many DC motors have six poles a peak at 360 Hertz is very common let's confirm our finding in the time domain we find a pattern of recurring prominent Peaks measuring between the peaks we find a delta time of 110 point nine milliseconds which converts to point one one zero nine seconds taking the reciprocal one divided by point one one zero nine equals nine point zero two Hertz rounded which very nearly approximates the one times frequency the predominant vibration is the turning of the shaft nothing else the peaks between 300 and 400 Hertz are electrical vibrations not mechanical vibrations you'll learn more about them in the course on motors step 6 evaluate equipment and make recommendations for all practical purposes this pump has no problems all your pumps should run as smoothly as this one you want to continue monitoring it of course but for now there is nothing to fix you [Music]
Info
Channel: RedVectorOnline
Views: 6,480
Rating: 5 out of 5
Keywords: pump vibrations, pump problems, incorrect pump selection
Id: bn0pIh2GT74
Channel Id: undefined
Length: 8min 4sec (484 seconds)
Published: Fri Sep 14 2018
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.