Reading a spectrum graph for vibration analysis


by Ask1122
Tags: analysis, graph, reading, spectrum, vibration
Ask1122
Ask1122 is offline
#1
May15-10, 08:34 AM
P: 52
Hi all,

I am confused with the amplitude at 1x, 2x, 3x frequency. The 1x, 2x and 3x is the harmonic (or mutiples) of the original frequency, meaning that if the shaft is turning at 1500rpm, then 2x will be the amplitude at 3000rpm and 3x will be the amplitude at 4500rpm right? But how will the amplitudes at 3000 and 4500rpm be measured as the shaft is only turning at 1500rpm?

And how does one interpret this kind of graph, for e.g.: why does an unbalanced shaft wil have high amplitude at 1x frequency, while misalignments will have high amplitude at both 1x, 2x and sometimes even at 3x frequency?

Thanks.
Phys.Org News Partner Engineering news on Phys.org
A smart prosthetic knee with in-vivo diagnoses
Old tires become material for new and improved roads
Students take clot-buster for a spin
nucleus
nucleus is offline
#2
May15-10, 11:46 PM
P: 171
The graph has the amplitude on the Y axis and frequency on X axis in cps. It may show several frequencies, therefore you need to know the speed of your machine or measure it with a tach. In your case the frequencies will be interested in are 25, 50 and 75 Hz and also speed 12.5.


Misaligned or bent shaft will show up at 1X, but that is not the only problem that will show up at that speed. This is an Unbalance. Misalignment with high axial vibration will show up at 2X. This is caused by mechanical looseness. At 3X usually a combination of misalignment and excessive axial clearance (looseness) can occur. There are many other possible problems but a good book on vibration analysis from hands on perspective, not just theory is IPT's Rotating Equipment Handbook:
http://www.iptbooks.com/index.php?op...d=48&Itemid=60
Ask1122
Ask1122 is offline
#3
May16-10, 12:24 AM
P: 52
Thank you for your answer, but would you please explain why do we worry about 2x and 3x the frequencies? Because the shaft is only turning at 1500rpm, it will not ever reach 3000 and 4500 rpm, so why are the harmonics important, and how do we even measure the amplitude of the frequencies that can never be reached (in this case, the 3000 and 4500rpm) in the first place?
And I want to understand "why" does unbalanced shaft spikes at 1x frequency while misalignment causes spike in both 1x and 2x frequencies?

sophiecentaur
sophiecentaur is online now
#4
May17-10, 06:54 AM
Sci Advisor
PF Gold
sophiecentaur's Avatar
P: 11,382

Reading a spectrum graph for vibration analysis


I think that harmonics could be caused by non-linearities - rattling, etc.. That's true for oscillating systems in general.
This could be of more concern than a slight imbalance as the load on the bearings in that case could be more benign.
brewnog
brewnog is offline
#5
May17-10, 02:10 PM
Sci Advisor
PF Gold
brewnog's Avatar
P: 2,793
The harmonics are important because (for instance) your shaft at 1,500rpm (25Hz) could be supported by a bracket which has a natural frequency of 50Hz, in which case it will be excited by the shaft motion, resonate, fail, destroy your machine, and kill everyone nearby. Or something.

Imbalance is at 1x because, well, it is. Imagine a big overhung weight on your shaft; it will move up and down once per revolution.

You would obviously do well to look at some basic vibration theory.

http://en.wikipedia.org/wiki/Vibration
S_Happens
S_Happens is offline
#6
May17-10, 10:08 PM
PF Gold
S_Happens's Avatar
P: 301
Vibrations at harmonics that are multiples of the rpm can be caused by misalignments for a couple reasons. It obviously depends a lot of the type of equipment. Misaligment could cause vibrations due to clearance issues (contact) and if you're talking about a semi-open or open pump impeller there could be multiple times per revolution that contact is made. That's a very simple example and maybe not what you are looking for, but you weren't very specific.

I have the book by J.P. Den Hartog that is referenced in the wiki article, and it's a good book, but being that it's quite old it can be difficult to read. The best one I've come across is probably...
http://www.amazon.com/Fundamentals-R...7&sr=8-1-spell

It's quite pricey, but maybe you can find a copy somewhere to borrow. Just the first few chapters would do answer a lot of questions for you.
Ask1122
Ask1122 is offline
#7
May17-10, 11:07 PM
P: 52
Quote Quote by brewnog View Post
The harmonics are important because (for instance) your shaft at 1,500rpm (25Hz) could be supported by a bracket which has a natural frequency of 50Hz, in which case it will be excited by the shaft motion, resonate, fail, destroy your machine, and kill everyone nearby. Or something.

Imbalance is at 1x because, well, it is. Imagine a big overhung weight on your shaft; it will move up and down once per revolution.

You would obviously do well to look at some basic vibration theory.

http://en.wikipedia.org/wiki/Vibration
Thank you for the replies guys, but a question in terms of the explanation above, firstly, the bracket support can have a natural frequency of 45Hz, and the bearings holding the shaft can have a natural frequency of 35hz for examples, so wouldn't the frequencies in between the harmonics be important as well? And secondly, how did they even measure the amplitude of the harmonics when the shaft speed only stayed at 1500rpm?

Lastly, I can understand that imbalance is at 1x, your example was very easy to understand, but why does misalignment have frequency at both 1x and 2x? When you have a misalignment, doesn't the shaft also only move up and down once per revolution?

thanks!
brewnog
brewnog is offline
#8
May18-10, 12:05 PM
Sci Advisor
PF Gold
brewnog's Avatar
P: 2,793
The vibration theory (pay attention to the bit about mode shapes) will help you understand how a shaft spinning at 1,500rpm can have a component at 3,000rpm and 4,500rpm etc.

See S Happens's post regarding harmonic frequencies for misalignment. Interesting things also happen, for instance, when you have angular misalignment. Here, a radial motion is translated to an axial motion, which can excite higher order resonances too.


Register to reply

Related Discussions
Relation between spectrum in algebraic geometry and spectrum in functional analysis Linear & Abstract Algebra 0
Reading frequency spectrum / Fourier Transform and Power Spectra Engineering, Comp Sci, & Technology Homework 2
Reading a pressure vs enthalpy graph Mechanical Engineering 4
Reading Variable Force Graph Introductory Physics Homework 4
PHYSICS OF MUSIC: Pitch is organ pipes & vibration spectrum Introductory Physics Homework 1