EXPERIENCE WITH TORSIONAL VIBRATION AT

RECIPROCATING COMPRESSORS
Dr.-Ing. Johann Lenz*, Dr.-Ing. Fikre Boru
KTTER Consulting Engineers GmbH & Co. KG
Bonifatiusstrae 400, 48432 Rheine, Germany
[email protected]

Abstract
Reciprocating compressors are unavoidable classical solutions in the field of
natural and process gas compression with the ability to function over a wide
range of operating conditions. The dynamic design of the reciprocating
compressor is complicated due to the large number of conditions that have to
be satisfied. Since high torsional dynamic stress is often not recognised until
damages appear, it is advisable to conduct a detailed torsional vibration
analysis when planning a new drive train or modifying an existing one. In this
paper, the different measures to influence the torsional behaviour of
reciprocating compressors are presented with the help of four case studies.

THEORETICAL BACKGROUND

The working principle of the reciprocating compressor leads to torsional

loading in the crank shaft. This torsional loading is then transmitted to the
optional flywheel, the coupling and the driving motor. Besides the working
principle of the compressor, a lot of factors influence the torsional loading of
the drive train. The excitation loads responsible for the torsional fatigue
loading are explained below.
The piston rod force FP can be calculated from the cylinder gas forces and the
simplified dynamic forces of the piston and piston rod masses as follows:

s(t)

Piston rod

Kolben
Piston

FFp K

ppeA
ppi i

uerer
External
volume
Zylinderraum

Figure 1:

innerer
Internal
volume
Zylinderraum

Simplified cylinder

Piston rod force FP(t):

F p (t ) = m p a p (t ) + p i (t ) Ai p e (t ) Ae

(1)

where
m p = piston and piston rod mass
ap

= piston acceleration

pi e
Ai e

= internal, external cylinder pressure

= internal, external piston area

From the piston rod force and the dynamic force of the crosshead and the
reciprocating mass of the connecting rod, one can calculate the radial (Fr) and
tangential (Ft) forces (as a function of the crank shaft angle) acting on the
crosshead pin. The tangential force component is responsible for the torque
loading of the crank shaft.
Figure 2 depicts the piston rod force and the resultant moment load of a
single crank shaft of a typical slowly rotating natural gas reciprocating
compressor for one complete crank shaft revolution.

150

Crank shaft moment [kNm]

20

Piston rod force [kN]

100
50
0
-50
-100
-150
-200
-250

Figure 2:

50

100

150

200

250

300

350

15

10

-5

50

100

150

200

250

300

350

Piston rod force and crank shaft moment, variation of a typical natural
gas piston compressor as a function of crank angle

Moment [kNm]

In order to analyse the resulting torsional vibration, the spectrum of the

dynamic components of the torque is generated as shown in figure 3.

Harmonics [-]

Figure 3:

Spectrum of excitation torque

The torque spectrum is composed of a number of harmonic components

which are multiples of the compressor rotational speed. The value of these
harmonic components depends on the operating condition of the compressor.
The torque spectrum presented above acts on the torsionally flexible drive
train. The drive train is composed of a number of elements having moment of
inertia, torsional damping and torsional stiffness properties.

Before installation of reciprocating compressors a torsional vibration study

should be conducted in order to check if the allowable loadings are not
violated by the planned operation of the compressor. Besides the allowable
load levels of the drive train components, the allowable maximum vibration
amplitude and the allowable safety margin between resonance and excitation
frequency have to be checked.
The results of the torsional vibration analysis depend on the quality of the
equivalent physical model and the consideration of the different conditions,
which may arise during operation of the compressor. The model should be
accurate enough so that it is sensitive to the smallest changes, for example a
different coupling manufacturer. In practice, it is advisable to check the
torsional vibration level by measurements during commissioning of the drive
train.

2
2.1

CASE STUDIES

Problems after retrofit with an active suction valve unloader

Since the retrofit of an active suction valve unloader to a reciprocating

compressor led repeatedly to failures of its coupling, a basic investigation of
the torsional vibration was conducted on an equivalent drive train to
determine the influence of the unloader. The analysed reciprocating
compressor is a 2-stage, boxer-type compressor with 2,100 kW coupling
power having a loaded operating speed range from 600 rpm to 1,000 rpm. In
order to broaden the range of the volume flow, an active unloader was
retrofitted on both stages of the drive train. An equivalent torsional physical
(figure 4) and mathematical model was generated. The first two eigenvalues
and their corresponding eigenforms are shown in the Campbell diagram
(figure 5).
Driving torque

Throw 1 to 4

Crank shaft

Motor shaft
Coupling

Figure 4:

Torsional physical model of the investigated drive train

Torsional eigenfrequency f [Hz]

2nd ef @ 183.75 Hz

1st ef @ 98.15 Hz

Rotational speed n [rpm]

Figure 5:

Campbell diagram as well as 1st and 2nd torsional eigenform at the drive
train

The suction valve unloaders influence the excitation torque spectrum of the
drive train. Figure 6 shows the torque spectra for different settings of the
unloaders. The dynamic response of the drive train was analysed for the
unloader settings. In figure 7 the torque amplitude in the coupling element is
given for an operation between 400 rpm and 1,100 rpm.

Torque [kNm]

Discharge
volume

Harmonics [-]

Figure 6:

Torque spectra of the 2nd throw (2nd stage) of the 4-cylinder

reciprocating compressor operating at 750 rpm

Fatigue torque amplitude [kNm]

Coupling
Discharge
volume

Rotational speed n [rpm]

Figure 7:

Torque amplitude of the coupling for different settings of the suction

valve unloader, operation speed between 400 rpm and 1,100 rpm

Figure 7 shows that the dynamic torque of the coupling was considerably
higher for the 75 % setting than for 100 % (i. e. deactivated unloader). This
was due to the change in the 8th harmonic of the excitation, which interfered
with the 1st torsional eigenfrequency.
This example shows that - depending on the system damping and the
selected rotational speed - large dynamic torque may result from the
activation of the unloaders. Hence, it is always advisable to conduct a
torsional vibration analysis before retrofitting an existing reciprocating
compressor with an active suction valve unloader.

2.2

Resonance of the connecting shaft

Reciprocating compressors are often applied for liquefaction of natural gas.

Damage at the metal disc coupling of ten similar 2-stage reciprocating
compressors was registered after different operation life time. The
compressors with a coupling power of 350 kW were operated at a constant
rotational speed of 600 rpm. The drive train consists of an electric motor
connected to the compressor (with two double acting cylinders) by two metal
disc couplings, an interconnecting shaft and a flywheel (figure 8).

Figure 8:

Two-stage reciprocating compressor with metal disc coupling and

connecting shaft

For the preliminary investigation of the possible source of the problem, the
calculated torsional eigenfrequencies of the drive train were presented in the
Campbell diagram in figure 9.

Torsional eigenfrequency f [Hz]

nd

2 ef @ 116.6 Hz

st

1 ef @ 19.5 Hz

Rotational speed n [rpm]

Figure 9:

Campbell diagram with calculated eigenfrequencies (ef.) to determine

the possible resonance speed

It can be seen that in the neighbourhood of the operation speed of 600 rpm
there was an interference between the 1st torsional eigenfrequency (19.5 Hz)
and the 2nd excitation harmonic. The torsional vibration at the connecting
shaft was measured as shown in figure 10. Figure 11 shows the amplitude
spectrum of the measured torque.

Power supply

Telemetry unit

Reciprocating
compressor

Motor
Strain gauge

Shaft

Receiver

Figure 10:

Principle layout of a torque measurement system

Figure 11:

Amplitude spectrum of the torque for start-up of the compressor

During running up of the compressor, a large torsional resonance at about

20 Hz was recorded. To describe the torsional mode, a vector diagram of the
measured torque was used as shown in figure 12.

Figure 12:

Vector diagram of the measured torque of the drive train

From the vector presentation it can be seen that the main torsional
deformation occurred in the connecting shaft. As an easy corrective measure
the diameter of the connecting shaft was increased. This moved the
calculated 1st torsional eigenfrequency up to 26 Hz. Since this corrective
measure was implemented at all compressors, there have not been coupling
failures due to torsional vibration anymore.

2.3

Influence of damping

Expanding the operation speed range of two reciprocating compressors for a

natural gas storage resulted in a failure of the lubrication system (driven by
the crank shaft) of both drive trains after a short operation life time. A
measurement showed that torsional resonance vibration was the cause for
this failure. For this drive train the customer decided to run the drive train at
constant speeds of 750 rpm, 850 rpm and 1,000 rpm. Such decisions limit the
flexibility of the plant, hence other possible solutions are discussed below.
An alternative is to replace the metal disc coupling which is almost rigid and
has a very low damping, with an elastic coupling in order to torsionally
decouple the motor shaft from the crank shaft. Additionally, the elastic
coupling brings damping into the drive train. However, the application of
elastic coupling results in changing the dynamic property. Hence, a careful
torsional analysis is required before such an implementation.

Below, a principle investigation of the coupling effect is presented for a sixcylinder natural gas (reciprocating) compressor with an elastic coupling and
flywheel. A torsional finite element model was developed for the drive train.
Then, an appropriate metal disc coupling was selected. Figures 13 and 14
show calculated eigenfrequencies and eigenforms of the drive train with the
elastic (figure 13) and the metal disc (figure 14) coupling.
1st ef @ 5.48 Hz

2nd ef @ 34.1 Hz

Motor

3rd ef @ 58.33 Hz

Elastic
coupling

Crank shaft

Throw 1/2 Throw 3/4 Throw 5/6

4th ef @ 73.56 Hz

Figure 13:

Calculated eigenvalues of a six-cylinder reciprocating compressor with

elastic coupling

1st ef @ 52.54 Hz

2nd ef @ 110.48 Hz

3rd ef @ 145.31 Hz

Motor

Metal disc
coupling
Throw 1/2

Crank shaft

Throw 3/4

Throw 5/6

4th ef @ 206.96 Hz

Figure 14:

Calculated eigenvalues of a six-cylinder reciprocating compressor with

metal disc coupling

One can see that the elastic coupling gives three natural frequencies below
the 1st crank shaft natural frequency. These natural frequencies are torsional
vibrations within the coupling and hence do not exist in the system with metal
disc coupling. Using an elastic coupling the excitation of the eigenmodes with
large twisting within the coupling leads to a large power dissipation which may
lead to an overheating of the coupling elements resulting in coupling failure.
Hence, an additional control of the power dissipation is unavoidable for
systems with elastic coupling. The 1st crank shaft eigenfrequency of the drive
train with metal disc coupling is 52.5 Hz, whereas that of the drive train with
elastic coupling is 73.6 Hz. Figure 15 shows the Campbell diagram of both
assemblies to determine the possible resonance operation speeds between
450 rpm and 1,150 rpm.

Torque amplitude [kNm]

Elastic coupling ( x )
9xn

st

6xn

1 eigenfrequency 5.5 Hz
2nd eigenfrequency 34.1 Hz
3rd eigenfrequency 58.3 Hz

3xn

4th eigenfrequency 73.6 Hz

Metal disc coupling ( o )
st

1 eigenfrequency 52.5 Hz
2nd eigenfrequency 110.5 Hz
Rotational speed n [rpm]

Figure 15:

Campbell diagram of the drive train with elastic and metal disc coupling

In reciprocating compressors with variable operation speed, it is not possible

to completely avoid the excitation of all eigenfrequencies due to higher
harmonics of the rotation frequency. One way to maintain the torsional
amplitude within an acceptable limit is to use a coupling with damping
properties. The effect of different coupling damping on the dynamic torsional
response of the drive train (discussed above) is shown in figure 16.

Motor shaft stud

Torque amplitude [kNm]

9% Damping ratio
4.5% Damping ratio
1% Damping ratio

Rotational speed n [rpm]

Crank shaft stud

Torque amplitude [kNm]

9% Damping ratio
4.5% Damping ratio
1% Damping ratio

Rotational speed n [rpm]

Figure 16:

Comparison of the torsional loading of the motor shaft stud (top) and
crank shaft stud (bottom) for different damping ratio

The calculation results show that damping ratio has a crucial influence in
attenuating the torque amplitude level in the motor shaft stud. Unlike the
motor shaft stud, the torque level in the crank shaft stud is irresponsive to
variation of the coupling damping ratio. At the resonance speed of around
750 rpm in the crank shaft stud the 4th eigenfrequency is excited by the 6th
harmonics. Since the coupling elements are almost static (nodes) for the 4th
eigenform, the variation of coupling damping ratio plays an insignificant role
for this resonance.

2.4

Influence of dynamic absorber

There are different ways to reduce the dynamic torque at the resonance
speed of 750 rpm. One possible method is the installation of either an
undamped or damped torsional dynamic absorber directly on the crank shaft.
An undamped torsional dynamic absorber removes the 4th eigenfrequency but
results in two new eigenfrequencies, one above and one below the previous
value. As shown in figure 17, the dynamic torque level at the two new
resonance speeds (resulting from the new eigenvalues) is significantly lower
than the previous level. Including damping to the absorbers will further reduce
the torque level as illustrated in figure 18.
Crank shaft without dynamic absorber

Torque amplitude [kNm]

Crank shaft with dynamic absorber

Operation rotational speed

Allowable torsional amplitude

Rotational speed n [rpm]

Figure 17:

Torque level at the crank shaft stud with and without torsional dynamic
absorber
Crank shaft with dynamic absorber 1% damping ratio

Torque amplitude [kNm]

Crank shaft with dynamic absorber 4.5% damping ratio

Crank shaft with dynamic absorber 9% damping ratio

Operation rotational speed

Allowable torsional amplitude

Rotational speed n [rpm]

Figure 18:

Influence of damping on the torsional dynamic absorber on the torque

level

CONCLUSION

Torsional vibration at reciprocating compressors cannot be identified on site

with simple measuring methods like e. g. measurement of the casing
vibration. Often high torsional dynamic stress is not recognised until damages
appear. Hence, it is advisable to conduct a detailed torsional vibration
analysis when planning a new drive train or modifying an existing one.
Different tools to influence the torsional vibration behaviour of a reciprocating
compressor are possible. We recommend to check the results of the analysis
by a strain gauge measurement to ensure the safety of the plant.