chapter 4 vehicle testing - virginia tech

37

Chapter 4

Vehicle Testing

The purpose of this chapter is to describe the field testing of the controllable dampers on a Volvo

VN heavy truck. The first part of this chapter describes the test vehicle used in the damper field

testing. The second part provides a background on how the dampers were mounted on the test

vehicle. Next, the configuration of sensors used to control the dampers and acquire the test data,

as well as the data acquisition system are described. Finally, this chapter describes the field tests

that were performed along with an analysis of the results.

4.1 Test Vehicle Description

The test vehicle used in this study was a Volvo VN series, heavy truck with L4 cab, as shown in

Figure 4.1.

Figure 4.1. Volvo VN Heavy Truck, Model 770 with the Test Trailer

The test truck includes a 48 ft box trailer that was unladen for our tests. The tractor weighs

44,000 pounds and the legal limit on the gross vehicle weight for this vehicle is 80,000 pounds.

38

4.2 Damper Installation on Test Vehicle

The Volvo VN heavy truck has six dampers at the primary suspension, as shown in Figure 4.2.

Figure 4.2. Location of Primary Suspension Dampers on Test Vehicle

Of the six dampers, four were replaced with controllable MR Dampers, as shown in Figure 4.3.

Figure 4.3. Location of MR Dampers on Test Vehicle

We selected to place four dampers on the truck, because we did not have enough dampers for all

six locations, and our past experience had shown that we can get most of the benefits of

39

semiactive dampers by placing them on only two of the four locations at the drive axle. The four

MR dampers are shown installed on the front and rear axles in Figures 4.4 and 4.5, respectively.

fron

t

driver passenger

Figure 4.4. MR Dampers Installed on the Front Axle

rear

passengerdriver

Figure 4.5. MR Dampers Installed on the Rear Axle

The four MR dampers were wired to the system controller, which was located in the sleeper cab

of the test vehicle. The wiring between the controller MR dampers was installed such that they

can be used for repeated tests, while the MR dampers were installed for quick exchange with the

truck’s stock dampers.

40

4.3 Sensors and Data Acquisition System

In order to control the dampers according to the skyhook policy described in section 2.1.4, it is

necessary to sense the velocities at each end of the controllable dampers. The accelerometers

used for this, shown in Figure 4.6, were manufactured by PCB Piezotronics, and had a sensitivity

of 100 mV/g and were used in conjunction with a PCB 584 series 16 channel signal conditioner.

Figure 4.6. PCB Accelerometers Used for Field Testing

Eight accelerometers, one at each end of each damper, were used to capture the data needed to

control the four dampers, as shown in Figure 4.7. Four of the accelerometers were mounted on

the truck’s frame rail, near the top of each of the four MR dampers and four were located on the

front and rear axles, approximately below the accelerometers on the frame rail.

frame rail mountedaccelerometer

axle mountedaccelerometer

Figure 4.7. Rear Passenger-Side Accelerometers

41

Additionly, three accelerometers were used to capture data for evaluating vehicle ride quality.

These accelerometers were arranged in a triax configuration, as shown in Figure 4.8.

Figure 4.8. Accelerometer Triax

The accelerometer triax was located at the B-post, directly behind the driver, 35 inches above the

cab floor, as shown in Figure 4.9, in order to measure the vibration transmissions to the cab in

the vertical, lateral, and fore and aft directions.

Figure 4.9. Accelerometer Triax Location

The signals from the eight accelerometers located on the frame and axles of the truck were

multiplexed to allow the signal to be simultaneously recorded and used in the control of the

dampers. The acceleration signals were integrated by the controller to find the velocity at each

42

end of the dampers. Based on the sign of the relative velocity across each damper, the controller

supplies either a zero or a three amp current to the damper, according to the on-off skyhook

control policy that was discussed earlier.

All eleven channels of accelerometer data were sampled at 6000 Hz and recorded for the

duration of the test using a sixteen channel SONY DAT recorder model PC216Ax, shown in

Figure 4.10.

Figure 4.10. SONY DAT recorder

All channels of the DAT recorder were tested with known waveforms before and after the tests,

to ensure proper functioning of the recorder channels.

4.4 Field Testing

The effect of the dampers was investigated with respect to both transient and steady state

dynamics. In the transient dynamic tests, the truck was driven over the speed bump shown in

figure 4.11 at 6-7 mph.

Figure 4.11. Speed Bump Test

43

Driving over the speed bump induced large amplitude oscillations in the test vehicle, which were

then damped out by the suspension system. This test was performed repeatedly to increase the

accuracy of the data, which was collected for four cases:

1. The MR dampers on the truck, operated according to the on-off skyhook control policy

outlined previously.

2. The MR dampers on the truck, continuously operated in their off (zero current) state.

3. The MR dampers on the truck, continuously operated in their on (three amp current)

state.

4. The original passive dampers in place (i.e., stock dampers).

The above represent one semiactive case and three passive cases. The three passive cases

represent hard damping (MR dampers in their on state), soft damping (MR dampers in their off

state), and medium damping (stock dampers).

The steady state portion of the test consisted of driving the test vehicle along a straight,

level road at a sustained highway speed of 55 mph. In this case, the input to the suspensions is

the road input at the tires. The steady state tests were conducted for cases 1, 2, and 4 of the

transient tests. Case 3 was not performed with a steady state input as the expected performance

can be extrapolated from the measured performance of cases 2 and 4.

The data resulting from the field tests of the dampers consists of ten-second segments of

eleven channels of data sampled at 6000 Hz, resulting in approximately 660,000 data points for

each data segment. Each data set needed to be heavily processed in order to extract the

necessary information. Each of the data points in the data set is a voltage, which from the

sensitivity of the accelerometers and the gains of the signal conditioner, can be converted into

acceleration. Each data channel maps to an accelerometer in one of the eleven positions

previously outlined.

In order to discuss the separate accelerometers, it is necessary to label each of the

accelerometer positions. To facilitate this, each wheel of the truck was given a letter as shown in

Figure 4.12.

44

Figure 4.12. Accelerometer Position Convention

As shown in Table 4.1, accelerometers are then referred to by pair of letters referring first to the

wheel letter shown above, and second to either T of B corresponding to either frame-mounted or

axle-mounted accelerometers, respectively. The triax accelerometers are referred to by the

direction in which they are measuring.

Table 4.1. Accelerometer- Channel Assignments for Field Testing

Channel Position Measurement1 AT driver side front, frame accelerometer2 AB driver side front, axle accelerometer3 BT passenger side front, frame accelerometer4 BB passenger side front, axle accelerometer5 ET driver side rear, frame accelerometer6 EB driver side rear, axle accelerometer7 FT passenger side rear, frame accelerometer8 FB passenger side rear, axle accelerometer9 Y B-post roll

10 Z B-post heave11 X B-post pitch

4.5 Transient Data Analysis

The transient or speed bump data was looked at in both the time and frequency domains, but the

main analysis was carried out in the frequency domain. In each ten-second data set, the truck

hits the speed bump with the front wheels at about two seconds into the set. The ten second data

set is long enough for the vehicle oscillations to damp out by the end of the data set.

45

4.5.1 Time Domain Analysis of the Transient Data

The time domain analysis of the transient data was performed with respect to both acceleration

and displacement, with the first steps of the data processing being the same for both. The first

steps included decimating the data by first passing it through a lowpass filter, and then

resampling it at a lower frequency. The low pass filter used for the decimation was a 30 point

finite impulse response (FIR) filter, shown in Figure 4.13.

Figure 4.13. Frequency Response for 30 Point FIR Filter Used in Decimation

This filter was chosen because of its low pass-band ripple, and steep attenuation at higher

frequencies. The data was then resampled with a decimation factor of 60 (i.e., every 60th point

was used), moving the new Nyquist frequency to 50 Hz. The next step was to apply a digital

filter to the decimated data in order to eliminate both high frequency noise and low frequency

drift. A Chebyshev bandpass filter was created with a bandpass of 1 to 15 Hz, steep attenuation

on either side of the passband, and unity magnitude within the passband. The low end of the

bandpass was chosen to be 1 Hz to match the low end of the useful range of the accelerometers.

Figure 4.14 shows the ideal filter in red and the actual filter used in blue.

46

0 5 10 15 20 25 30 35 40 45 500

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

frequency (Hz)

mag

nitu

de

Figure 4.14. Frequency Response for Chebyshev Filter Used

The filter was applied to the data in first the forward direction and then the data reversed and the

filter reapplied. This eliminated phase distortion and modified the magnitude by the square of

the filter magnitude. The passband magnitude of the filter used is unity with small ripples to

eliminate the effect of the filter magnitude. This filtering was accomplished using the MATLAB

.m file “filtfilt”.

The effect of applying this type of filter was experimentally verified by testing a known

signal. The known test signal, shown in Figure 4.15, was a decaying 4 Hz sine wave.

0 1 2 3 4 5 6 7 8 9 10-2

-1.5

-1

-0.5

0

0.5

1

1.5

24 Hz test signal

time

sign

al

Figure 4.15. Test Signal Used for Validating Filters

47

The test signal was “hidden” by combining it with both a decaying 20 Hz sine wave and a

decaying 0.8 Hz sine wave, as shown in Figure 4.16.

0 1 2 3 4 5 6 7 8 9 10-2.5

-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

2.5test signal "hidden" with 0.8 and 20 Hz signals

time

sign

al

Figure 4.16. Test Signal “Hidden”

The combination of these three waves was put through the filter shown in Figure 4.14, using both

standard filter techniques and the zero-phase forward and reverse filtering that filtfilt applies.

The results of putting the “hidden” test signal through the filter in Figure 4.14 using both

standard filtering and zero-phase forward and reverse filtering are shown in Figure 4.17. The

MATLAB .m file for this purpose is included in Appendix 1b.

48

Figure 4.17. Effect of Applied Filters to a Known Signal

Though the zero phase forward and reverse filtering induces greater discrepancies at the start of

the data set than standard filtering, it is more effective at preserving the transient character of the

data. At this point the data processing differs depending on whether it is acceleration or

displacement that is of interest.

4.5.1.1 Acceleration Data Analysis

The acceleration data was mean-zeroed and plotted versus time to obtain the summary

information. The summary information consists of:

• the global acceleration maximum during the ten second data block

• the local acceleration maximum immediately following the global maximum

• the corresponding times of the two acceleration maximums

• the slope of the decay between the first and second acceleration maximums

• the RMS acceleration for the time period of one second before the first acceleration

maximum to two seconds after.

49

The MATLAB .m file that was used to compute this information is included in Appendix 1c. A

sample plot of the acceleration data for channel 10 (cab acceleration in the vertical direction) is

shown in Figure 4.18 with a line connecting the first and second peaks used in the summary

information.

0 1 2 3 4 5 6 7 8 9 10-5

-4

-3

-2

-1

0

1

2

3

4filtered acceleration of channel 10

acce

lera

tion

(m/s

2)

t ime (sec)

Figure 4.18. Sample Plot of Acceleration Data for Channel 10

Plots of this for the all channels tested in each of the four test scenarios (MR dampers with

skyhook control, MR continuously on, MR continuously off, and original dampers) are included

in Appendix 2. Values of both the maximum and RMS acceleration were averaged across like

data sets for each channel. There were nine data sets taken in which the test truck, equipped with

MR dampers and skyhook control, was driven over the same speed bump. The average peak

acceleration amplitude and average RMS acceleration for each of these nine sets of data were

averaged together. The results of this are shown in Figures 4.19 and 20.

50

Average Peak Acceleration Amplitude

0

2

4

6

8

10

12

14

AT AB BT BB ET EB FT FB Y Z X

Accelerometer Position

Acc

eler

atio

n

P o s i t i o n M e a s u r e m e n t

A T d r i v e r s i d e f r o n t , f r a m e

A B d r i v e r s i d e f r o n t , a x l e

B T p a s s e n g e r s i d e fro n t , f r a m e

B B p a s s e n g e r s i d e fro n t , a x le

E T d r i v e r s i d e r e a r , f r a m e

E B d r i v e r s i d e r e a r , a x l e

F T p a s s e n g e r s i d e r e a r , f r a m e

F B p a s s e n g e r s i d e r e a r , a x le

Y B - p o s t r o l l

Z B - p o s t h e a v e

X B - p o s t p i t c h

Figure 4.19. Acceleration Results: Average Peak Acceleration Amplitude for the Test Vehicle

w/MR Dampers and Skyhook Control Policy

Average RMS Acceleration

0

0.5

1

1.5

2

2.5



RM

S A

ccel

erat

ion

P o s i t io n M e a s u r e m e n t

A T d r i v e r s i d e f r o n t , f r a m eA B d r i v e r s i d e f r o n t , a x l eB T p a s s e n g e r s i d e f r o n t , f r a m eB B p a s s e n g e r s i d e f r o n t , a x l eE T d r i v e r s i d e r e a r , f r a m eE B d r i v e r s i d e r e a r , a x leF T p a s s e n g e r s i d e r e a r , f r a m eF B p a s s e n g e r s i d e r e a r , a x l eY B -p o s t ro llZ B -p o s t h e a v eX B -p o s t p i t ch

Figure 4.20. Acceleration Results: Average RMS Acceleration for the Test Vehicle w/ MR

Dampers and Skyhook Control Policy

The test in which the truck with the MR dampers being operated continuously in the on or three

amp state was driven over the speed bump was repeated five times. The peak acceleration

amplitude and RMS acceleration from each of these five data sets were averaged together. Since

the quantities to be compared from one test case to another are averaged across data sets, the

51

number of data sets from test to test does not need to be the same. The results for the five data

sets are shown in Figures 4.21 and 22.


0

2

4

6

8

10

12

14



Acc

eler

atio

n




B T p a s s e n g e r s i d e fro n t , f r a m eB B p a s s e n g e r s i d e fro n t , a x l e









w/MR Dampers Operated in the On State


0

0.5

1

1.5

2

2.5



RM

S A

ccel

erat

ion




B T p a s s e n g e r s i d e fro n t , f r a m eB B p a s s e n g e r s i d e fro n t , a x l e








Figure 4.22. Acceleration Results: Average RMS Acceleration for the Test Vehicle w/MR

Dampers Operated in the On State

There were four data sets in which the test vehicle was driven over the same speed bump with

the MR dampers on the truck and being operated continuously in the off or zero amp state. The

results for the four data sets are shown in Figures 4.23 and 24.

52

Average Peak Accelerat ion Ampl i tude

0

1

2

3

4

5

6

7

8

9

10

AT AB BT BB ET EB F T FB Y Z X

Accelerometer Posit ion

Acc

eler

atio

n



A B d r i v e r s i d e f r o n t , a x l eB T p a s s e n g e r s i d e f r o n t, fra m e

B B p a s s e n g e r s i d e f r o n t, a x leE T d r i v e r s i d e r e a r , f r a m e

E B d r i v e r s i d e r e a r , a x l eF T p a s s e n g e r s i d e r e a r , f ra m e

F B p a s s e n g e r s i d e r e a r , a x leY B - p o s t ro ll

Z B - p o s t h e a v eX B - p o s t p i t c h


w/MR Dampers Operated in the Off State

Average RMS Accelerat ion

0

0.5

1

1.5

2

2.5

AT AB BT BB ET EB F T FB Y Z X


RM

S A

ccel

erat

ion



A B d r i v e r s i d e f r o n t , a x l eB T p a s s e n g e r s i d e f r o n t, fra m e

B B p a s s e n g e r s i d e f r o n t, a x leE T d r i v e r s i d e r e a r , f r a m e

E B d r i v e r s i d e r e a r , a x l eF T p a s s e n g e r s i d e r e a r , f ra m e

F B p a s s e n g e r s i d e r e a r , a x leY B - p o s t ro ll


Figure 4.24. Acceleration Results: Average RMS Acceleration for the Test Vehicle w/MR

Dampers Operated in the Off State

There were six data sets in which the test vehicle was driven over the same speed bump with the

truck’s original dampers in place. These data sets serve as a baseline with which to judge the

53

effectiveness of the MR dampers. The averaged maximum acceleration and averaged RMS

acceleration are shown in Figures 4.25 and 26.

Average Peak Acceleration

0

1

2

3

4

5

6

7

8



Acc

eler

atio

n














with Original Dampers in Place


0

0.5

1

1.5

2

2.5


Channel

RM

S A

ccel

erat

ion













Figure 4.26. Acceleration Results: Average RMS Acceleration for the Test Vehicle with

Original Dampers in Place

54

4.5.1.2 Displacement Data Analysis

After the data was decimated and filtered, there were transients at the start and end of the data

that existed as artifacts of the digital filtering. This effect can be seen looking at the filter test

signal shown earlier in Figure 4.16. While the data was being analyzed in terms of acceleration,

this effect was unimportant, however since looking at the data in terms of displacement requires

the data to be integrated twice which amplifies these errors, corrections must be made. In order

to correct for this error, the value of the acceleration of the first and last one second of data was

set to zero, and then the data was again mean zeroed. To integrate the data, each set was put

through a 1/s integrator block (corresponding to multiplying each frequency component by 1/jw)

using the MATLAB command LSIM. The data, which is now velocity, was re-filtered using the

filter shown in Figure 4.13, and again mean zeroed. Finally, the data was again integrated using

an integrator block.

The data, which is now displacement, was plotted and summary information extracted. The

summary information consists of:

• the global displacement maximum during the eight second data block

• the local displacement maximum immediately following the global maximum

• the corresponding times of the two displacement maximums

• the slope of the decay between the first and second displacement maximums

• the RMS displacement for the time period going from one second before the first

displacement maximum to two seconds after.

The MATLAB .m file that was used to do this is included in Appendix 1d. Figures 4.27-33 are

sample plots for seven of the eleven measurement positions showing displacement versus time.

These plots show a trend that the MR dampers operated continuously in their off state allow the

highest levels of displacement, and the MR dampers operated continuously in their on state allow

the lowest levels of displacement. Both the MR semiactive and original damper displacements

tend to be between these two extremes.

55

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02

0.03front passenger frame displacement

time (sec)

disp

lace

men

t (m

)

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02


time (sec)

MR dampersoperated withskyhook control

Originalpassivedampers

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02


time (sec)

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02

0.03body displacement (pitch direction)

time (sec)

MR damperscontinuouslyoff (soft)

MR damperscontinuouslyon (hard)

Figure 4.27. Front Passenger-Side Frame Displacement Sample Plots

56

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02

0.03front passenger axle displacement

time (sec)

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02


time (sec)

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02


time (sec)

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02


time (sec)





Figure 4.28. Front Passenger-Side Axle Displacement Sample Plots

57

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02

0.03front driver frame displacement

time (sec)

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02


time (sec)

disp

lace

men

t (m

)

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02


time (sec)

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02


time (sec)





Figure 4.29. Rear Driver-Side Frame Displacement Sample Plots

58

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02

0.03front driver axle displacement

time (sec)

disp

lace

men

t (m

)

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02


time (sec)

disp

lace

men

t (m

)

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02


time (sec)

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02


time (sec)





Figure 4.30. Rear Driver-Side Axle Displacement Sample Plots

59

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02

0.03body displacement (roll direction)

time (sec)

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02


time (sec)

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02


time (sec)

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02


time (sec)





Figure 4.31. B-Post Roll Displacement Sample Plots

60

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02

0.03body displacement (heave direction)

time (sec)

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02


time (sec)

(m

)

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02


time (sec)

disp

lace

men

t (m

)

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02


time (sec)

disp

lace

men

t (m

)





Figure 4.30. B-Post Heave Displacement Sample Plots

61

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02


time (sec)

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02


time (sec)

disp

lace

men

t (m

)

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02


time (sec)

disp

lace

men

t (m

)

0 1 2 3 4 5 6 7 8-0.03

-0.02

-0.01

0

0.01

0.02


time (sec)

disp

lace

men

t (m

)





Figure 4.33. B-Post Pitch Displacement Sample Plots

62

Plots of this for the all channels in each of the four test scenarios (MR dampers with skyhook

control, MR continuously on, MR continuously off, and original dampers) are included in

Appendix 3. Values of both the maximum and RMS displacement were averaged across like

data sets for each channel. There averaged maximum peak and RMS displacement for the nine

sets of data where the MR dampers were being controlled with the skyhook policy are shown in

Figures 4.34 and 35.

Average Peak Displacement Amplitude

0

0.005

0.01

0.015

0.02

0.025

0.03

0.035



Dis

plac

emen

t



Figure 4.34. Displacement Results: Average Peak Displacement Amplitude for the Test Vehicle

w/MR Dampers and Skyhook Control Policy

63

Average RMS Displacement

0

0.002

0.004

0.006

0.008

0.01

0.012



RM

S D

ispl

acem

ent

P o s i t i o n M e a s u r e m e nt



B T p a s s e n g e r s i d e f r o n t , f ra m e

B B p a s s e n g e r s i d e f r o n t , a xle





Y B - p o s t r o ll



Figure 4.35. Displacement Results: Average RMS Displacement for the Test Vehicle w/MR

Dampers and Skyhook Control Policy

The result of averaging the peak and RMS displacement for the five data sets where the test

vehicle was operated with the MR dampers continuously on are shown in Figures 4.36 and 37.


0

0.005

0.01

0.015

0.02

0.025

0.03

0.035



Dis

plac

emen

t



A B d r i v e r s i d e f r o n t , a x l eB T p a s s e n g e r s i d e f r o n t , f r a m e

B B p a s s e n g e r s i d e f r o n t , a x leE T d r i v e r s i d e r e a r , f r a m e

E B d r i v e r s i d e r e a r , a x l eF T p a s s e n g e r s i d e r e a r , f r a m e

F B p a s s e n g e r s i d e r e a r , a x leY B - p o s t r o ll



w/MR Dampers Operated in the On State

64


0

0.002

0.004

0.006

0.008

0.01

0.012



RM

S D

ispl

acem

ent




B T p a s s e n g e r s i d e f r o n t , f r a m e

B B p a s s e n g e r s i d e f r o n t , a x le





Y B -p o s t ro ll

Z B -p o s t h e a v e

X B -p o s t p i tc h


Dampers Operated in the On State

The result of averaging the peak and RMS displacement for the four data sets where the test

vehicle was operated with the MR dampers continuously off are shown in Figures 4.38 and 39.


0

0.005

0.01

0.015

0.02

0.025

0.03

0.035

0.04



Dis

plac

emen

t








F T p a s s e n g e r s i d e r e a r , f ra m e

F B p a s s e n g e r s i d e r e a r , a xle

Y B -p o s t ro l l




w/MR Dampers Operated in the Off State

65


0

0.002

0.004

0.006

0.008

0.01

0.012

0.014

0.016


Channel

RM

S D

ispl

acem

ent








F T p a s s e n g e r s i d e r e a r , f ra m e

F B p a s s e n g e r s i d e r e a r , a xle

Y B -p o s t ro l l




Dampers Operated in the Off State

The result of averaging the peak and RMS displacement for the six data sets where the test

vehicle was operated with the original dampers in place are shown in Figures 4.40 and 41.

Average Peak Displacement

0

0.005

0.01

0.015

0.02

0.025

0.03

0.035



Dis

plac

emen

t



Figure 4.40. Displacement Results: Average Peak Displacement for the Test Vehicle with


66


0

0.002

0.004

0.006

0.008

0.01

0.012



RM

S D

ispl

acem

ent





B B p a s s e n g e r s i d e fro n t , a x l eE T d r i v e r s i d e r e a r , f r a m e







Figure 4.41. Displacement Results: Average Peak Displacement for the Test Vehicle with


4.5.1.3 Results of Time Domain Analysis of Transient Tests

There are two parts of the time domain analysis of the transient tests. The first part of the

discussion will deal with the acceleration data and the second part will look at the results derived

from the displacement data.

A comparison of the average peak accelerations as measured at the eleven measurement

points while the test vehicle is driven over the speed bump is shown in Figure 4.42.

67


0

2

4

6

8

10

12

14



MR ActiveMR 3AMR 0AOriginal

Position MeasurementAT driver side front, frameAB driver side front, axleBT passenger side front, frameBB passenger side front, axleET driver side rear, frame

EB driver side rear, axleFT passenger side rear, frameFB passenger side rear, axleY B-post rollZ B-post heaveX B-post pitch

Figure 4.42. Average Peak Acceleration Comparison

A comparison of the four test cases shows that the MR dampers controlled with the skyhook

control policy exhibit equal or greater levels of average peak acceleration than the original

dampers on all channels. The levels of average peak acceleration were significantly higher for

the MR active case than the original dampers for accelerometer positions AB, BB, EB, and FB.

As these positions all represent to measurements being taken on the axles of the truck, this result

was not unexpected, as the MR dampers can be softer than the stock dampers. For the

measurement positions measuring frame acceleration (AT, BT, ET, and FT), the levels of

average peak acceleration were similar for the MR active case and the original dampers, with the

original damper case exhibiting slightly better performance (lower acceleration). The results of

the tests where the MR dampers were controlled with either zero (continuously off) or three

amps (continuously on) of current tend to envelope the average peak acceleration values of the

MR active case at positions measuring frame acceleration. The results of measurements made at

the B-post in the y, z, and x directions (corresponding to roll, heave, and pitch respectively) show

that the MR active case accentuates the acceleration seen by the cab of the vehicle.

A comparison of the average RMS accelerations as measured at the eleven measurement


68



0

0.5

1

1.5

2

2.5





Figure 4.43. Average RMS Acceleration Comparison

The comparison of the four test cases show that the MR dampers controlled with the skyhook

control policy exhibit greater levels of RMS acceleration than the case with the original dampers

on all channels measured on the axles of the truck. This was also found to be true for channels

measured on the frame in the front of the truck. However, the levels of RMS acceleration at the

frame in the rear of the truck (ET and FT) showed the MR active case to be better (lower levels

of RMS acceleration) than the original case. The MR active case is shown to transmit less RMS

acceleration to the frame of the truck than the cases where the MR dampers were either

continuously on or continuously off. The results of measurements taken at the B-post in the y, z,

and x directions (roll, heave, and pitch) show that the MR active case accentuates the

acceleration seen by the cab of the vehicle versus the original dampers.

A comparison of the average peak displacement as measured at the eleven measurement


69



0

0.005

0.01

0.015

0.02

0.025

0.03

0.035

0.04





Figure 4.44. Average Peak Displacement Comparison

The comparison shows that the use of the MR dampers with the skyhook control policy

increased the vertical displacement of both the axle and the frame in the front of the test vehicle

(AT, AB, BT, and BB) as compared to the stock dampers. In the rear of the truck, the

application of the MR dampers with the skyhook control policy had little effect on the vertical

displacement of either the axle or frame compared to the original dampers. Measurements taken

at the B-post in the y, z, and x directions (roll, heave and pitch of the cab of the truck) show

increased motion with the MR dampers and skyhook control policy when compared to the

original dampers.

A comparison of the average RMS displacement as measured at the eleven measurement


70



0

0.002

0.004

0.006

0.008

0.01

0.012

0.014

0.016





Figure 4.45. Average RMS Displacement Comparison

The comparison shows that even though the MR dampers with the skyhook control policy had

higher levels of peak vertical displacement of the frame, the RMS displacement of the frame was

not increased. This points to higher levels of initial frame displacement as the truck passes over

the speed bump, but quicker dampening of the vibration in the MR dampers and the skyhook

control policy. The RMS displacement of the front axle was higher for the MR skyhook control

case than it was for the original dampers. The RMS displacements of the rear of the truck, both

axle and frame, were reduced in the MR damper skyhook control case.

4.5.2 Frequency Domain Analysis of the Transient Data

The transient data was investigated in the frequency domain as well. The first step in this

investigation mean zeroed the data. The next step involved creating an averaged fft of the

acceleration data for each data set. This was done by decimating each data set twenty times. In

each of the twenty sets the decimation started three elements later than the last set. This created

twenty ffts of the same data, which were then averaged together frequency by frequency. The

MATLAB m file that did this is included in Appendix 1-e. These averaged ffts were then again

averaged, this time across like data sets. The data sets included:

71

• Eight sets of data with the MR dampers were in place and controlled according to the sky

hook control policy

• Four data sets with the MR dampers were in place and powered continuously at three

amps

• Four data sets with the MR dampers were in place and not powered

• Five data sets with the stock dampers on the truck

A set of data (eleven measurement positions) for each of these four test cases is included as

Appendix 3.

In order to facilitate comparison between the four different test cases, the results of the

frequency domain analysis were looked at in terms of average peak intensity in four frequency

bands. The four frequency bands that were chosen for analysis are:

• 1-4 Hz

• 4-9 Hz

• 9-14 Hz

• 14-19 Hz.

These bands were chosen based on a study by M. Ahmadian [13], who correlates these bands to

different aspects of the truck dynamics. These correlations are summarized in Table 4.2.

Table 4.2. Summary of Frequency/Truck Dynamics Correlations

Frequency Band Truck Dynamics1-4 Hz rigid body modes of the truck frame (heave and pitch)4-9 Hz first bending mode of the truck frame

9-14 Hz wheel hop frequencies of the three tractor axles14-19 Hz second bending mode of the truck frame

For each of the four test cases, the average peak intensity was calculated for each of the for

frequency bands. The average peak intensity was defined as the sum over the frequency band of

the product of the magnitude of the acceleration at each frequency times the frequency width.

The MATLAB .m file used to calculate this information is included in Appendix 1f. This was

performed for each of the eleven channels of acceleration data captured. The average

acceleration peak intensity results in the 1-4 Hz frequency band, corresponding to the rigid body

modes of the truck frame, is shown in Figure 4.46 for the four cases tested.

72


Average Peak Intensity for Frequency Band 1-4 Hz

0.0000

0.0050

0.0100

0.0150

0.0200

0.0250

0.0300

0.0350

0.0400

0.0450

0.0500



Ave

rage

Pea

k In

tens

ity (

m/s

^2-H

z)

MR Active

MR 3A

MR 0A

Original


Figure 4.46. Average Peak Intensity in the 1-4 Hz Frequency Band

The average acceleration peak intensity results in the 4-9 Hz frequency band, corresponding to

the first bending mode of the truck frame is shown in Figure 4.47 for the four cases tested.



0.0000

0.0010

0.0020

0.0030

0.0040

0.0050

0.0060

0.0070

0.0080

0.0090

0.0100



MR Active

MR 3A

MR 0A

Original




the wheel hop frequencies of the three tractor axles is shown in Figure 4.48 for the four cases

tested.

73



0.0000

0.0010

0.0020

0.0030

0.0040

0.0050

0.0060

0.0070



MR Active

MR 3A

MR 0A

Original




the second bending mode of the truck frame is shown in Figure 4.49 for the four cases tested.



0.0000

0.0005

0.0010

0.0015

0.0020

0.0025

0.0030

0.0035

0.0040



Ave

rage

Pea

k In

tens

ity (

m/s

^2-H

z)

MR Active

MR 3A

MR 0A

Original



The frequency domain analysis points to the effectiveness of the MR dampers and the skyhook

control policy. The average peak intensity of the measured acceleration is broken down into four

frequency bands. When these results are shown in terms of percent increase versus the case

where the truck was equipped with the original dampers, it is evident that there is a positive

74

effect of using the MR dampers with the skyhook control policy. A decrease in the average peak

intensity is shown by a negative percent increase. The results in the four frequency bands are

shown in Figure 4.50.


Average Peak Intensity of MR Active Case vs. Original Case (Transient)

-60.0000

-40.0000

-20.0000

0.0000

20.0000

40.0000

60.0000

80.0000

100.0000



Per

cent

Incr

ease

1-4 Hz4-9 Hz9-14 Hz14-19 Hz


Figure 4.50. Percent Increase of Average Peak Intensity: MR Active vs. Original (Transient)

In the front of the truck (AT, AB, BT, and BB) the use of the MR dampers with the skyhook

control policy significantly increased the average peak intensity of the acceleration of the axle in

both the 1-4 Hz and 14-19 Hz bands. The 4-9 Hz and 9-14 Hz bands showed a decrease in the

average peak intensity at these same positions. The rear of the truck (ET, EB, FT, and FB)

showed a significant reduction in the average peak intensity of the acceleration as measured on

both the axle and the frame of the truck. The acceleration measured in the cab of the roll, pith

and heave directions showed significant decreases in the average peak intensity in the frequency

bands 1-4 Hz, 4-9 Hz, and 14-19 Hz. The 4-9 Hz results in particular point to increased operator

comfort as the human body resonance typically falls in a 5-7 Hz range [12].

75

4.6 Steady State Data Analysis

The steady state data, like the transient data, was examined in both the time and frequency

domains. The time domain analysis of the transient data consisted of calculating the RMS

acceleration for three cases. The first of the three cases that was investigated was the truck

equipped with the MR dampers, and controlled according to the sky hook control policy. The

second of the three cases was the truck equipped with the MR dampers and operated with the

dampers continuously in the on or three amp state. The third case was the truck operated with

the original dampers in place. In each of the cases, the full ten second data block was used in the

RMS acceleration calculation. The result of this analysis is shown in Figure 4.51.

Position MeasurementAT driver side front,

frameAB driver side front,axleBT passenger side front,

frameBB passenger side front,axleET driver side rear, frame

RMS Acceleration for Steady State Data

0.0000

0.5000

1.0000

1.5000

2.0000

2.5000



RM

S A

ccel

erat

ion

MR Active

MR 3A

OriginalEB driver side rear, axleFT passenger side rear,

frameFB passenger side rear,axleY B-post roll

Z B-post heaveX B-post pitch

Figure 4.51. RMS Acceleration Results for Steady State Data

The steady state data was analyzed in the frequency domain in the same manner as the transient

data, with the exception that only three cases were investigated, the three cases being MR active,

MR 3A, and original dampers respectively. Another difference between the frequency domain

analysis carried out for the steady state and transient data is that in the steady state analysis

multiple data sets representing the same operating condition were not investigated as was the

case in the transient data analysis. A set of data (eleven measurement positions) for each of three

cases (MR active, MR 3A, and original) is included as Appendix 4. The average acceleration

76

peak intensity results in the 1-4 Hz frequency band, corresponding to the rigid body modes of the

truck frame, is shown in Figure 4.52 for the three cases tested.


0.00E+00

2.00E-03

4.00E-03

6.00E-03

8.00E-03

1.00E-02

1.20E-02



MR ActiveMR 3AOriginal





the first bending mode of the truck frame is shown in Figure 4.53 for the three cases tested.



0.00E+00

1.00E-03

2.00E-03

3.00E-03

4.00E-03

5.00E-03

6.00E-03

7.00E-03

8.00E-03

9.00E-03

1.00E-02



Ave

rage

Pea

k In

tens

ity (

m/s

^2-H

z)

MR ActiveMR 3AOriginal



77


the wheel hop frequencies of the three tractor axles is shown in Figure 4.54 for the three cases

tested.


Average Peak Intensity for frequency Band 9-14 Hz

0.00E+00

5.00E-03

1.00E-02

1.50E-02

2.00E-02

2.50E-02



MR Active

MR 3A

Original




the second bending mode of the truck frame is shown in Figure 4.55 for the three cases tested.

78



0.00E+00

2.00E-03

4.00E-03

6.00E-03

8.00E-03

1.00E-02

1.20E-02

1.40E-02



MR Active

MR 3A

Original



4.6.1 Results of Time Domain Analysis of Steady State Tests

In order to clearly show the effect that the MR dampers had with the skyhook control policy as

compared to the original dampers, the percent increase in RMS acceleration is plotted in Figure

4.56. Since this is a percent increase, a negative number represents a decrease in the RMS value

of the measured acceleration.

Position MeasurementAT driver side front, frameAB driver side front, axleBT passenger side front, frameBB passenger side front, axleET driver side rear, frameEB driver side rear, axleFT passenger side rear, frameFB passenger side rear, axleY B-post rollZ B-post heaveX B-post pitch

RMS Acceleration of MR Active Case vs. Original Case

-60.0

-50.0

-40.0

-30.0

-20.0

-10.0

0.0

10.0

20.0



Per

cent

Incr

ease

Figure 4.56. Percent Increase of RMS Acceleration: MR Active vs. Original (Steady State)

Channels 1 and 2, corresponding to the frame of the truck in the front show that the use of the

MR dampers with the skyhook control policy has reduced the RMS acceleration at these

positions by close to 50% on both sides of the truck. Channels 6 and 8 show that the RMS value

79

of the acceleration as measured on the rear axle was also significantly reduced. Further,

channels 9,10, and 11 show that the use of the MR dampers with the skyhook control policy on

the primary suspension was effective at reducing the RMS value of the acceleration seen by the

cab of the truck.

4.6.2 Results of Frequency Domain Analysis of Steady State Tests

In order to show the effectiveness of the MR dampers with the skyhook control policy versus the

original dampers, the percent increase in the average peak intensity was plotted. This is shown

in Figure 4.57.

Average Peak Intensit of MR Active Case vs. Original Case (Steady State)

-100.0000

-50.0000

0.0000

50.0000

100.0000

150.0000

200.0000



Per

cent

Incr

ease

1-4 Hz4-9 Hz9-14 Hz14-19 Hz


A T d r i v e r s i d e f r o n t , f r a m eA B d r i v e r s i d e f r o n t , a x l eB T p a s s e n g e r s i d e f r o n t , f r a m eB B p a s s e n g e r s i d e f r o n t , a x l eE T d r i v e r s i d e r e a r , f r a m e

E B d r i v e r s i d e r e a r , a x l eF T p a s s e n g e r s i d e r e a r , f r a m eF B p a s s e n g e r s i d e r e a r , a x le

Y B - p o s t r o l lZ B - p o s t h e a v e


Figure 4.57. Percent Increase of Average Peak Intensity: MR Active vs. Original (Steady State)

At most of the measurement locations, the MR dampers with the skyhook control policy showed

an increase in the average peak intensity of the measured acceleration. This points to larger

amplitude accelerations with shorter duration than with the original dampers.

chapter 4 vehicle testing - virginia tech

Documents