VA forms the
backbone of any condition monitoring program. One reason it is
so extensively used is because unlike
certain other CM techniques, VA is applicable to virtually all sections of a machine train
whereas some CM methods can only be carried out on the
motor, the gearbox or the fan, pump etc. as you can see from the diagram
What CM methods can we use
been doing vibration analysis long before anyone had heard of 'Condition
The old screwdriver trick is one of the longest
standing methods of CM however, it is very subjective. What
one person may consider excessive is another persons smooth!
But, using modern
instruments, vibration levels can be very accurately recorded and
compared to ISO and British standards for vibration levels.
However, VA is
most powerful when we can also compare it to previous
readings from the same point and trend how quickly a problem
analysis is also very
powerful at identifying exactly what type of fault is
developing on which component. How? well, all rotating and
reciprocating machinery will vibrate. The different
components will vibrate at their own rotational frequency.
All these vibrations are mixed up together in a complex
vibration waveform. Information about the condition of that machine is contained
within that waveform.
As a fault develops, so the
waveform will change and increase in amplitude. Vibration
analysis is a non intrusive method of collecting these
waveforms and using a mathematical process known as a fast
fourier transform or FFT which was originally invented by some
old French mathematician geezer (Joseph
1768–1830), this information can be broken
down into its component frequencies producing a spectrum
showing vibration amplitude against frequency. Analysing this
spectrum can give prior warning of component failures
thereby reducing the risk of unexpected breakdowns.
Three types of
radial reading are taken at each collection point on a machine,
usually in the vertical plane;
Velocity Spectra detect a wide range of faults such as
looseness, misalignment, mass imbalance and insecurity.
These spectras can be compared against the relevant ISO and
BS standards for vibration levels on rotating equipment.
They can also be converted mathematically in the software into
both acceleration and displacement spectras giving further
insight into the machine condition.
Band (HFB) acceleration readings indicate poor quality
lubrication and early stages of bearing and gear defect.
Signal Processing (ESP) extracts rolling element bearing
defect information from bearing housing resonance to give an
excellent indication of bearing condition. This is THE MOST
important technique available and should always be included
in any CM program as it identifies bearing defects well in
advance of a problem showing up in a standard velocity
spectra. Download our
VA Case Study presentation to see
how accurately it picked up a developing motor bearing defect.
In addition, an
axial velocity reading taken from each section
gives a more complete picture of machine condition.
In most cases, a
monthly VA survey is carried out by a CM engineer visiting site
and collecting data using a hand held vibration
analyser/collector, however, in some instances assets may be
deemed critical enough to warrant continuous monitoring
using permanently installed equipment which sends vibration
data back via the internet for analysis. This actually is
not as expensive as you may imagine, Please see our
page for further information.
readings are then trended
in the CM software to determine if any problems are
developing and if so, how quickly, and are graded according
The VA report will explain exactly what
problems are developing and suggest remedial action which
should be taken to prevent the machine from failing during
An example VA report can be found on our