Sign up Calendar Latest Topics Donate
 
 
 


Reply
  Author   Comment   Page 2 of 2      Prev   1   2
Mniroshan

Member
Registered:
Posts: 30
Reply with quote  #16 
Quote:
Originally Posted by RustyCas
Mniroshan, as Vibe-Rater states, when analyzing bearing defect frequencies, the process is greatly simplified when frequencies are stated and displayed as “orders” rather than cpm. I find when looking at fundamental defect frequencies that velocity works best because it gives higher amplitudes at lower frequencies. I use acceleration when looking at higher order (multiples) of defects, or high frequency peaks that are modulated (side-banded) by bearing defects, as acceleration units show higher amplitudes at high frequencies. (While looking at spectra in your software, you should be able to switch back and forth from V to A very easily)

As Danny states, the +/- 75 rpm “band” in the table is too wide for bearing defect analysis. For a 1500 rpm machine, that is +/- 0.05 orders (75/1500) which is not enough resolution. Also, when collecting data the resolution must be high enough to distinguish between 3X and 3.08, or 5X and 4.92, for example.



Thank you for you idea. I will keep it on mind because still I am new.
Mniroshan

Member
Registered:
Posts: 30
Reply with quote  #17 
Quote:
Originally Posted by Vibe-Rater
Hi Mniroshan,

Apologies for being curt. It is certainly not what this board is about and welcomes newcomers.

It all goes back to relating to shaft speed. Faults are either synchronous or non synchronous. That is either an integer multiple or non integer multiple of shaft speed. And is why you should think in orders which is a relative frequency unit.

After all we do not refer to our daily temperature in Kelvin - which is an absolute temperature unit. We prefer to use Celsius (or Fahrenheit) which are both relative units, relative to the freezing point of water. So vibration speak almost everything is relative to shaft speed.

For DGBB - deep groove ball bearings 63** dimension series with 8 rolling elements will produce outer race frequencies of 3.08x and BPFI of 4.92x. (non synchronous being the point) Regardless of shaft speed. So you don't really have to remember anything else. As an aside 3.08 + 4.92 = 8, the number of rolling elements.

So your spectrum appears to show a 1x.  Then the second highest peak could be a bearing frequency but it is not easy to tell if it is synch or non synch. Maybe if it is a pump it could be vanepass? 5x?

You are very welcome here and please get back to us all with more questions.  There is a great shortage of vibration people in the world so please continue to be involved. rgds


HI
Thank you for your great support. Pump Impeller has more than 5 vanes. So I dont think it is vane passing frequency.
Mniroshan

Member
Registered:
Posts: 30
Reply with quote  #18 
Quote:
Originally Posted by Vibe-Rater


So the data is from a motor so I would say all is normal.  They would normally have 63** at drive end and 62** at NDE unless pulley reduction and an NU3** at the drive end.) 1x you will always see, usually no matter how small. The next highest peak I think might be transmitted vane pass from the pump. All good. An extra display in acceleration units will be helpful because it will reveal early bearing characteristic frequencies better.  I don't like calling them fault frequencies because by inference they mean a fault. Often bearings will show characteristic frequencies but not necessarily indicate trouble.

It is easy to identify frequencies - more difficult to determine severity and deterioration rate, both of which are really important. rgds


Thank you for your idea.
Mniroshan

Member
Registered:
Posts: 30
Reply with quote  #19 
Quote:
Originally Posted by MarkL
I'd suggest looking at all of Jasons videos on youtube if you're starting out, they are what urged me to be more than a data dog. Got me interested in the analysis a whole lot more than I was at the beginning of my Pdm career.
A very good taster to taking the full courses with Mobius.

link to Mobius Youtube channel.


A lot of the later stuff is promoting the conferences but go back to the beginning of them and there are hours of excellent content.




Thank you for your help.
tomcd3

Member
Registered:
Posts: 62
Reply with quote  #20 
Excellent explanations for us PdM/VA newbies. Thx.

We in the bearing industries, don't like the term defect frequencies as it implies a manufacturing defect, which is usually not the case. Fault - like the San Andreas - doesn't imply an issue is present, only an impending one. Esp. since a demod or spectral energy/acoustic emmission signal can detect a crack before it becomes a spall.

Here is a great example discovered by a power plant using VA last year. 
20181129_161314558_iOS.jpg   

Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Easily create a Forum Website with Website Toolbox.