Research on the swing and vibration restraint of the 2 MN deadweight machine frame


ACTA IMEKO | www.imeko.org December 2020 | Volume 9 | Number 5 | 97 

ACTA IMEKO 
ISSN: 2221-870X 
December 2020, Volume 9, Number 5, 97 - 99 

 
 

RESEARCH ON THE SWING AND VIBRATION RESTRAINT 

OF THE 2 MN DEADWEIGHT MACHINE FRAME 
 

Lin Shuo1 

 
1 Fujian Province Institute of Metrology, Fuzhou, China, linshuo1001@126.com  

 

Abstract: 

In this paper, the new 2 MN deadweight force 

standard machine of FJIM (Fujian Metrology 

Institute) with an optimal 20 kN frame with swing 

and vibration restraint is detailed. Load transient 

response is considered in the design of column, and 

a mass damper is used to make the frame stable in 

3 s. These methods help the machine reach the 

uncertainty it claimed. 

Keywords: deadweight machine, vibration, 

damper, dynamic 

1. INTRODUCTION 

A new 2 MN deadweight force standard machine 

(DWM) was installed in 2018 at Fuzhou, China, and 

as shown in Figure 1 in the structural sketch. The 

measuring range of the machine is (20 ~ 2000) kN 

with a stated uncertainty of 0.002 % (k = 3). All the 

masses are made of stainless steel. This machine is 

the largest DWM in China while the maximum 

range was 1.1 MN before. It has finished the 

comparison with NIMTT China and NPL UK, and 

the results give confidence in the claimed 

uncertainty of the machine. 

  
Figure 1: FJIM’s 2 MN deadweight machine 

Compared to PTB’s 2 MN machine [1], there are 

some significant differences and improvements. 

First is the lower limit of measurement; FJIM’s is 

designed to 1 %, so the measurement range is from 

20 kN to 2 MN to meet more types of calibration 

requirements. 

In order to achieve this goal, a three steps frame 

was used. The frame includes a 20 kN reverser, and 

two 10 kN columns joined by cones. They are able 

to be loaded independently. To minimise the effects 

of longitudinal vibration and lateral swinging, and 

to improve stability in the high accuracy 

measurement, the method of dynamic optimisation 

and the mass damper were used in the design. 

2. OPTIMISATION OF THE REVERSER 
AND THE COLUMN  

The conventional design method tends to 

consider only von Mises stress under static 

conditions [2]. For the purpose of weight reduction, 

the static stiffness is used to the maximum extent, 

then the dynamic stiffness might be not enough. 

Longitudinal vibration will appear when the masses 

loading at high speed under low dynamic stiffness 

of structure, and force fluctuation acts on the load 

cell which being tested and then be reacted on the 

instrumentation. 

The whole frame can be viewed as a quality-

spring-damper system. Short response time and 

short stabilisation time is the optimisation goal. 

FEM tool was used in the design of 2 MN DWM 

frame, and different mass states and loading speeds 

were set as constraints conditions. 

 
Figure 2: FEM analysis of reverser 

Through the optimisation design, the maximum 

stress value of the beam under the inverter is 

http://www.imeko.org/
mailto:linshuo1001@126.com


ACTA IMEKO | www.imeko.org December 2020 | Volume 9 | Number 5 | 98 

obviously reduced, the stress distribution range span 

is reduced, the overall stress is more uniform, and 

the working state is more stable. 

 
Figure 3: FEM analysis of column 

Local reinforcement was implemented on the 

triple column reverser. Under the guidance of 

simulation analysis, a step-shaped hollow column 

with a weight of 10 kN was produced, with a 

maximum diameter of 245 mm. Topological 

optimisation helps the structure to have higher 

dynamic stiffness when the damping coefficient of 

material is constant. The stress concentration 

coefficient was lowered by using arc transition 

angle at the same time. The results of transient 

analysis show the acceleration corresponding curve 

in Figure 4, and the longitudinal vibration stopped 

within 1 s when 2 MN masses load simultaneously. 

 
Figure 4: Transient of longitude vibration 

3. MASS DAMPER 

There are several factors that cause lateral swing 

including installation error, the eccentricity of the 

mass centroid and the eccentric or tilt on load cell 

being tested. All of the above factors will lead to the 

imbalance of the frame and column caused by the 

deviation of the centre of mass from the central axis 

during the loading and unloading process, which 

will ultimately affect the accuracy and stability of 

the machine. An ADAMS (a multibody dynamics 

simulation software) model was built up for kinetic 

analysis and is shown in Figure 5. 

Different initial conditions, including starting 

angles, weights, and mass positions were analysed, 

while the time to stop swing is the goal of simulation. 

The unbalanced state and swinging will stop under 

the action of friction and air resistance. An angle of 

0.5 % was set in analysis and it took 10.6 s until 

becoming stable at 2 MN. 

 
Figure 5: Factors of swing and ADAMS model 

Table 1: Stabilisation period under different loads 

Load / kN Time / s 

150 2.6 

900 5.4 

1500 6.2 

2000 10.6 

 

In the process of weight configuration, the 

position of the weight should be considered in the 

budget. For the same loading weight, different 

loading positions are selected to analyse the 

influence of loading positions on the tilt and swing 

of the central column. As the results in Table 2 show, 

the closer the loading position is to the top joint, the 

smaller the vibration amplitude and the shorter the 

stabilisation time. 

Table 2: Stabilisation period under different position 

Load / kN Distance / mm Time / s 

150 

1500 6.1 

3000 9.4 

5000 15.8 

7000 19.5 

 

Although the tilt and swing of the column can be 

minimised by careful processing and installation, 

this phenomenon cannot be completely eliminated. 

Therefore, it is necessary to design a structure to 

reduce the swing amplitude and shorten the 

stabilisation time [3]. The simulation results show 

that the stabilisation time t is negatively correlated 

with the suspension mass and the friction coefficient 

of the ball hinge, and positively correlated with the 

rod length. 

After several times of optimisation, the optimal 

result of mass damper was designed as a 115 kg 

mass with a 200 mm column, and a spherical hinge 

with friction coefficient of 1.3. The structure and the 

simulation result are shown in Figure 6. Obviously, 

the stabilisation time under 2 MN and 0.5 % tilt of 

frame is shortened to less than 2 s compared to 10 s 

without mass damper. 

http://www.imeko.org/


ACTA IMEKO | www.imeko.org December 2020 | Volume 9 | Number 5 | 99 

 
Figure 6: Mass damper and the simulation of swing 

4. TEST RESULT 

Tests of static deformation and dynamic 

response was tested using an HBM 2 MN load cell, 

a vibration sensor, and several laser displacement 

sensors, which are shown in Figure 7. 

  
Figure 7: Dynamic response test system and mass 

damper 

The maximum deformation of the frame under 

2 MN is less than 5 mm, and the uneven 

deformation less than 0.1 mm/m. The longitudinal 

vibration stopped in 0.2 s which hardly reaction on 

the DMP-41 indicator. A 5 mm eccentric was set on 

the load cell, and the swing caused by misalignment 

stopped within 3 s. Figure 8 shows the swing 

displacement at the end of the column, and the 

stability is restored after one period of vibration. 

 

Figure 8: Test of swing time and result 

5. SUMMARY 

FJIM’s 2 MN deadweight machine was 

optimally designed so that the first step of 20 kN can 

be reached. The paper presents the methods for 

guaranteeing high accuracy force measurement. 

The longitudinal vibration can be restrained by the 

structure of column, and the lateral swinging of 

frame was reduced by a mass damper. Test results 

show that all those advantages are effective. 

With the help of these innovations, the machine 

gets the uncertainty of 0.002 % (k = 2) and with 

excellent stability. The accuracy has been verified 

by international comparison. It is of positive 

significance to the application of force measurement 

up to 2 MN. 

6. REFERENCES 

[1] M. Peters, D. Peschel, “PTB’s new 2 MN 
deadweight force standard machine”, in Proc. of 

19th IMEKO TC3 Congress, February 2005. 

[2] N. Vilajic, A. Chijioke, R. Seifarth, “Stress Analysis 
of Conical Contact Joints in the NIST 4.45 MN 

Deadweight Machine”, Journal of Research of the 

National Institute of Standards and Technology, 

vol. 121, pp. 222-237, May 2016. 

[3] L-K. Wang, W-X. Shi, “An adaptive-passive 

retuning device for a pendulum tuned mass damper 

considering mass uncertainty and optimum 

frequency”, Structural Control and Health 

Monitoring, pp. 135-141, vol. 26, May 2019.

 

http://www.imeko.org/