..

Balancing
.

Balancing with X-Viber

First check that dominating vibration is TOTAL I-FILTERED
caused by an unbalance '*

19.17 14^4
The I-FILTERED vibration should have the mm ^ rin^
same frequency or Hz as the running
speed of the machine 800Hz 2949RPM

Do riot change the position of the vibration transducer after the start of the balancing
procedure

Balancing using this method requires only three consecutive mai runs and changing the
balance status of the rotor
Only measurement of the vibration levelis needed.
~

Balancing will of course only reduce the vibration caused by unbalance

A balancing round will often be a good approach and a first attempt to find the reason for
increased vibration. Ifthe balancing attemptis riot successful, the cause can be loose
rotor paris etc

Ifthe machine speed is variable, be sure to choose the same speed during every trial run
Do riot search the speed that gives the highest vibration. Such speeds mostly show non-
.
linear results

Start the procedure by measuring on the bearings looking for high levels in major
directions. Choose a pointthat should have a good connection to a balancing plane where
you can putin a weight in the machine. You must use the same radius forthe trial weights
and the baiandng weights.

X-Viber manual 35

.,*. " '.~

p"*,,*.,.,*,. -~...-"-..-. ,

.
.

This is the procedure forthe Two-POINTBALANCING METHOD:

TRIAL RUN ,:

TOTAL I-FILTERED Selectthe running speed and choose the

measuring point. Measure and note the vibration
19.17 14^4- level and stop the machine. Note this vibration as
min grins reading A.
800Hz 2949RPM In this example the vibration is 14.64mm/s at
2949rpm

TRIAL RUN 2:
Putin a mai neightin the baiandng plane. Note the
P 10' location and size of the balandng neight. Use its
weight, volume or length (if you have a band material)
as a measure of the size of the trial weight

Callthe trial weight P in any measure propQrtional to the

weight

Measure and note the vibration level and stop the

machine
Note this vibration as reading 2
TOTAL I-FILTERED
13.01 958 In this example re added 409rams at the radius
mm srms
320mm in position O'.
The vibration becomes 9.58mm/s and the speed
800Hz 2952RPM was almostthe same 2952rpm.

TRIAL RUN 3:
Move the trial weight 180 degrees to a position opposite
to the first location.

Measure and note the vibration level and stop the

machine

Note this vibration as reading 3.

, ,'"*

TOTAL I-FILTERED
In this example re added 40grams at the radius
26.71 2134 320mm in position 180".
mm grins The vibration becomes 21.34mm/s and to speed
was almostthe same 2953rpm
800Hz 2953RPM

36 X-Vibermanual

,,
.

.
,

CALCULATION
We-now have alithe necessary machine infonnation to start the calculations

Compare the vibration levels reading 2 and reading 3.

*
Callthe highest of reading 2 and reading 3 for B and the smallest for C.

In our example reading 3 (2134) is the highest and ne call it B and reading 2 (958) is
the smallest and re call it C. ~

Draw a figure there you use a length measure proportional to the actual vibration level
measured on each reading A, B and C respectively. Use a scale that gives you as large a
figure as possible. The accuracy depends largely on the size of the figure

STEP I: DrawA as a horizontal line. Use a scale so you

make a large figure and good use oftha paper
A

B/2 STEP 2 Drawan arc centred at the left end of A having the
radius equal to 1/2 (HALF)the length of. B

STEP 3: Draw an arc centred at the tightend of A

having the radius equal to 1/2 (HALF) the length ofC.
B/2 Mark the crossing of the two arcs. Draw a line starting
from the right end of Athrough the crossing of the two
C/2 arcs and with the length proportional to C.

STEP 4: Join the upper tip of C with the left end of A so

that a mangle shaped figure is made. Mark this new line
C as D. The angle position of the balancing neightis tfre
D angle (d) in the left corner of the figure betneen line D
B/ C/2 and baseline A.

A In our example the angle (0) is measured to be 38'.

X-Viber manual 37

^~:~^,~~.~~~^,,,~,.=*,.:".,.~~,,~,*....--__.:
'"."!'W* -~..-., -...~ ... .

,,
.

.
 \
,

THE SIZE OFTHE BALANCINGWEIGHT

, The SIZE of the balancing weightthat you should put on the balancing plane is
proportional to the size of the trial neight with the same relation as the line A to D in the
figure. In other words:

trial weight P * the length of A .3

Balancing weightBW=---------------
the length of D

In our example will the balancing weight BW = 409rams x 14.64 I 75 = 789rams on the
same radius 320mm

You can measure A and D in the figure. You may soon realise that bestresults are
obtained then the triangle has approximately equal sides

THE ANGLE POSlnON OFTHE BALANCINGWEIGHT

B When we had the inalweightin position C re had a

lower vibration than in position B. The trial weight in
position C must therefore be on the proper half of the
rotor.
The balancing weight should be positioned (0) degrees
from the position of the trial neight C
a a

POS. 2 C POS. I The angle (0) can be measured on your rotorin either
the diredion against orwith rotation. You must make a
qualified guess and try one alternative. Ifthe vibration is
riot reduced the other location may be the better

TOTAL I-FILTERED
22.82 18.16 In our example we first placed 789rams 38 degrees in
the direction of the rotation
mm grins

800Hz 2954RPM The result was not good.

TOTAL -FILTERED We then moved the 7891ams 38 degrees in the

5.22 164 direction against the rotation
min s rms
The result was very good.
800Hz 2952RPM
Ttre vibration became that lowthattrre unbalance
vibration became the second highest peak and ne
hade to select the 2-FILTERED to find the same
frequency as the running speed.

The balancing may be stopped then the highest radial direction is below 3 minis ms.

38 X-Vibermanual

~~..~...
. .,..,...,. ,.~~ . . ." ... " ...,.~ . . ~ .....-"
,grey. ,

. ,,

.
.

MISCELLANEOUS
If you prefer to calculate the baiandng weight, this is the forrnula:
32_CT
,
By,
15'*,-? cos a =
^;^5~ 24'a'*A,

=.

TROUBLE HINTS
The most difficulttask in balancing is to guess a suitable size of the trial weight because
,
we do not know the unbalance sensitivity of the machine. The wiiole baiandng procedure
depends on the changes in vibration levelthat ocerir then vre add a trial neight. Ifthe trial
neightis too small compared to the unbalance we can riot measure any changes in the
vibration levels and the measurements A, B and C become almost equal. The triangle in
our figure becomes flat

Ifthe maiweightis too large compared to the unbalance the measurements B and C
become very large compared to the measurement A and the mangle in our figure
becomes very high and narrow.

This produces an uncertainty in our calculations

,

,
Ifthe triangle can riot be fomied well, use the results to guess a better size of the trial
weight ortry a location of the trial weight a quarter turn away from the first position

Ifthe mangle can not be formed at alithere are often two majorreasons:
Something is loose. Check the fit between the rotor and shaft. Check boltjoints
Is dirt gradually falling off during each run?
The vibrations in these cases are not caused by an unbalance. You are trying to balance a
machine where the unbalance is very small and there the vibration comes from other
faults. E. g. misalignment, gearbox, cavitations in pumps etc

X-Viber manual 39

^,*,,."eL, q*^
.,^ ,

, ,,

.
.

This is the procedureforthe THREE. POINTBALANCING METHOD:

The three point method gives you both the size and the location angle (a) without
guessing the position of the balancing weight. Be sure to choose the same speed during
alitrialruns and the same radius for alitrial weights and the balancing neight

Step ,.
e.

TOTAL I-FILTERED With the rotor operating at normal speed, measure

17.44 1362 and note the original vibration level as Re
min grins
In our example we measured the vibration at the
800Hz 2946RPM running speed I-FILTERED to be : R, = 1362 min/s

B=, 20"
Draw a circle with a radius proportional to R, and
then use the same scale forthe rest of the procedure.

R=, 3 62 Stop the rotor. Mark three points on the rotor,

approximately 120 degrees apart where we later will
A= 0"
add trial neights. Callthese points A, B and C. These
three points do not need to be exactly 120 degrees
apart butthe accuracy decreases the more you leave
equalspadng. You have to measure your chosen
angle position as accurate as possible.
C=240'

Step 2
TOTAL I-FILTERED Put in a trial neight in position A.
12.56 920 Start the rotor and measure the vibration level R2
mm grins
In our example we added 40 grains at 320 mm
800Hz 2945RPM radius and the vibration became R2 = 9.2 mints.

B =9 2

Draw a circle proportional to R2 with the centre in

position A.
+ A
The red circle

40 X-Vibermanual

.....,.. .,,.. ."

,,
.

.
.

Step 3
TOTAL IFILTERED Move the same trial neight to position B.
18.91 1436 Start the rotor and measure the vibration level R3.
min grins
In our example we moved the 409rams at 320 min
800Hz 2947RPM radius to po^ition B and tire vibration became R3 =
14.36 mints

R=, 4 36
Draw a circle proportional to R3 with the centre in
B position B.

The green circle

A

Step 4.
TOTAL I-FILTERED
Move the trial weight to position C.
26.33 21n3 Start the rotor and measure tire vibration level R4
mm grin^
2948RPM In our example ne moved tfre 40 grains at 320 min
800Hz
radius to position C and the vibration became R4 =
21.03 min/s.

Draw a circle proportional to R4 with the

centre in position C

The blue circle.

R=21 03

X-Viber manual 41

<4"~",,*,*.~~"""~'~~" * . -..-, -. -.,-...... . - -.~.... ~ .-... . ~ . ...

,,
,

.
 .
,

Step 5

Alithe three circles R2 , R3 and R4 intersect in a

*
common point D.
Draw a line from the centre of the circle R, to the point
D. Callthis line R5. Measure the length of this line
using the same scale as before

In our example R5 = 7.52 and it is the vibration level

R5 caused by the trial weight P if P was the only
CL unbalance in this rotor

Measure the angle or in the drawing. This is the

angular position of the balandng from position A

Step 6
Calculate the size of the balancing weight using the formula:

TrialWeighix LengthqjRl
Boloncihg17eighi
Lengihq/R5

In our example the baiandng neight will be:

40 x 13,62
BIP = 72.4groms
7.52

The balancing weight 72.4 grains was placed 38

TOTAL -FILTERED degrees from position A.
8.03 278
The result was very good.
mm srm^
The vibration became that lowthatthe
800Hz 2955Rpjvj
unbalance vibration became the third highest
peak and ne hade to seled the 3-FILTERED to
find the same frequency as the running speed

The balancing may be stopped when the highest radial direction is below 3 min/s mis
If you want to further improve the balancing status you must start horn tile beginning and
use a smaller trial weigt.

42 X-Vibermanual

. , ~~....

,,
.