Sign up Calendar Latest Topics Donate

  Author   Comment  

Avatar / Picture

Sr. Member / Supporter
Posts: 599
Reply with quote  #1 
A major focus of my vibration training material is to convey the difference between vibration as a diagnostic tool, and vibration as a hazard. That faults that cause vibration are generally separate from faults that are caused by vibration. I had a shower thought that I think provides a good analogy.

Your body temperature is generally 98.6F/37C. When you have a slight fever of 101/38.5, the fever is not a risk to you, but it does indicate that there is something else going on inside of you, i.e. an infection, which generally is something that needs your attention. So the fever is a great indicator of a hidden fault. However, when that fever gets too high 104/40, the fever itself becomes the imminent threat, and can often come from a relatively benign infection. At 107/41.5, the fever will cause brain damage and possibly kill you. In this case, it is not the infection that is killing you. It is your body's response to the infection.

I thought it was apt.

As an interesting sidenote, there were instances at the turn of the 20th century where syphilis was cured by infecting a patient with malaria, inciting a high fever of ~108 that killed the syphilis bacteria, then putting the patient in an icebath to control the fever, and using new medicines to cure the malaria.

Sr. Member
Posts: 163
Reply with quote  #2 
Interesting analogy

When you said, "....Vibration as a hazard" do you mean hazards to the human body (e.g. hand damage due to use of jack hammers, etc.)?


High vibrations (e.g. high unbalance forces, resonances, misalignment, etc.) which I consider the "ROOT CAUSE" problem resulting in other faults (e.g. looseness, bearing damage, etc.) which they themselves start to generate their own vibration signatures? 

Jim P

Sr. Member
Posts: 647
Reply with quote  #3 

In the end you can never explain everything we do in simple terms, but it's worth thinking about. I might (imperfectly) group the way we think about vibration into three groups:

  1. 1 – is it normal (compared to other similar machines or history).  If it’s not normal then that’s not necessarily a problem, but it deserves investigation.
  2. 2 – does the vibration  indicate a condition which is a threat to the reliability of the machine.  (rolling bearing defect is classic example).
  3. 3 – Is the vibration itself a threat to the reliability of the machine.  (I don't think it happens often except for maybe 1 - looseness which begets more looseness, and 2 - extreme high vibration levels)

Of course there is some overlap (anything in category 2 or 3 is probably also 1) and some room within each category (if you have a rolling element defect than it’s category 2, but how likely is it to threaten reliability and how soon... that we often don’t know).

Sometimes we don’t really know the cause from the vibration alone, so we fall back on 1 and ask the mechanics to take down the machine for a closer look.

But the fever analogy is simpler and more timely! 


Sr. Member
Posts: 670
Reply with quote  #4 
Very good analogy with the fever

Avatar / Picture

Sr. Member / Supporter
Posts: 599
Reply with quote  #5 

Originally Posted by ivibr8
When you said, "....Vibration as a hazard" do you mean...

Primarily a hazard to the machine. That means, vibration that itself causes fatigue or tribology-related faults.

I don't consider something like misalignment causing antifriction bearing damage to be in this category, because the vibration from misalignment doesn't cause the bearing damage. The increased lateral forces from the misalignment causes the damage. However, the vibration from misalignment can directly cause other faults.

Originally Posted by ePete
I might (imperfectly) group the way we think about vibration into three groups

I refer to group 2 as a causal fault and group 3 as a resultant fault. Causal faults have a sub category I call pseudo-faults (my colleagues hate the name) that cause vibration, but don't actually indicate a causal fault that itself is a threat to reliability besides the vibration it causes (e.g. most types of structural resonance).

I would separate group 1 completely. That doesn't look at the actually fault-vibration relationship, but just evaluates the amplitude. Every causal fault is always present in every machine to some degree. There is always unbalance. There is always misalignment. There is always vibration generated from the imperfections of a rolling element bearing. We can categorize them without considering the amplitude.

When we evaluate the amplitude, we can do 4 things:

  • Use absolute vibration limits from a standard
  • Use comparative analysis to compare to sister or similar reference machines
  • Use historical/trended limits to indicate rate of change of vibration
  • Use a high-effort analysis that considers every component of the machine and every known failure mode to generate a model that gives an ODS-dependent vibration limit specific to this machine.
Obviously the first 3 are quite easy. I am working on garnering support for the last one, so we can start rolling it out in 40 or 50 years.
Previous Topic | Next Topic

Quick Navigation:

Easily create a Forum Website with Website Toolbox.