Click for next page ( 314


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 313
D.~. Introduction / Background One important aspect of creating an ergonomic workstation in transit buses is choosing an appropriate seat. The most ideal seat would be one that adjusts to all the ranges needed in the workstation design, attenuate as much vibration as possible, and is comfortable for the operator. There are several methods to measure vibration exposure to a seated person, and thus indicate what seat is the "best", some of which are described in (Gilmore, 1995~. Further, the following methods will be used to determine which seat is acceptable for the workstation prototype and guidelines, all taken from (Griffin, 19901. There are two approaches, time domain analysis and frequency analysis. In time analysis, one of the most common measures used is the root-mean-square (r.m.s.) value: R M S = [N ia2'(i)] where N is the total number of data points and aw is the weighted acceleration value. The r.m.s. value is an average measure of the peak values, and therefore is not subject to one or two extreme values. However, in motions where shocks occur, the r.m.s. value may not be an appropriate measure. To quantify this, some definitions are necessary. The peak value is defined as the maximum deviation from the r.m.s. value in a time series. Also, the crest factor is the ratio of the peak value over the r.m.s. acceleration. If the crest factor is greater than six, then the r.m.s. value is not a good representation of the vibration D-]

OCR for page 313
levels. Therefore, another measure, the root-mean-quad (r.m.q.), accounts for more of the higher acceleration levels and is a better method for high crest factor motions. RMQ=[NiaiV(i)] Further, a method that measures the cumulative exposure of vibration is through the vibration dose value (V.D.V.) which is defined as: V D V = [N ~ a4,~i)1 where Ts is the duration of the motion being analyzed. Finally, the last time analysis method used will be the seat effective amplitude transmissibility (SEAT, pronounced 'see-at') value, which is a ratio of the VDV of the seat over the VDV of the floor. The SEAT value gives another indication of what vibration is passed from floor to human. A value of 1 00% indicates similar comfort to sitting on the floor, and values less than 1 00% indicate an improvement over the floor. in the frequency domain approach, the methods used will be the power spectral density (PSD) and the transfer function or transmissibility. The PSD indicates the dominant frequencies of the r.m.s. value of the data, and the transfer function reveals how much acceleration is being passed on and at what frequencies. The seated human has natural frequencies around 4 Hz and ~ Hz (Griffin, 1990), and the bus has natural frequencies at 1-2 Hz and 10-12 Hz (Boileau and Boutlin, 19901. Therefore, it is important to investigate how the seats attenuate the vibration at the critical frequencies. D-2

OCR for page 313
D.2. Problem Statement The objective testing of the seats is made up of two distinct phases. Phase ~ includes the static evaluation of the seats. This includes evaluating the seat features (armrests, height adjust, etc.), and experimentally finding various seat parameters (i.e. cushion stiffness, suspension damping, etc.~. Phase I] involves the dynamic testing in a bus on the PTT test tracks. The goal of the dynamic testing of the transit bus seats is to relatively compare the ability of the seats to attenuate vibration. From this comparison, the seat that best isolates the vibration to the bus operator will be used in the bus operator workstation prototype. There are several different factors which effect how these seats operate during a typical transit route, and the test should be devised to accurately account for all of these variables. In this experiment, the same bus will be used for all of the trials since this will provide a control for the experiment. This bus is the Chevy bus used in previous studies by PT] (Figure D-.. (Note: Unless otherwise noted, all figures are located in Sub- appendix Dl). Also, a human ride simulator will be used as a control for human physiology (Wambold, 19861. This simulator represents a 50th percentile male and a good correlation with actual human subjects, yet is not susceptible to psychological influences such as personal mood. The two natural frequencies of the simulator are 4.25 Hz and 7.5 Hz. D-3

OCR for page 313
D.3. Equipment 50 Ib. weights P.T.T. Chevy Bus B.S.T. portable 486-66 Mhz computer with Labview software Dytran amplifier board, capability of up to six channels 5 Dytran 3 ~ 27A accelerometers and cables Seven seats from various manufacturers Air tank capable of at least ~ 00 psi Ride quality simulator Durability and economy tracks at the P.T.~. Bus Testing Facility D.4. Procedure In the Phase ~ testing, various seat parameters and general features of the seats are measured. The seat parameters include suspension mass, suspension damping ratio, cushion stiffness, and cushion mass. (The 'suspension' is defined as the part of the seat from the seat riser to the bottom of the cushion.) Natural frequency and stiffness of the seat suspension can be found using the data from Phase I! (See Section D.5.) In order to find cushion stiffness, a simple experiment is conducted. The cushion is removed from the seat, placed on a rigid surface, and the ride quality simulator base is placed on the seat. Fifty pound weights are placed on the base, and the displacement of the base is D-4

OCR for page 313
measured. Using linear regression, the slope of the force / displacement line is the stiffness of the cushion, assuming linearity. Next, to find suspension damping ratio, the seat is set up in its normal operating conditions (air hookup, etc.~. The ride dummy is placed on the seat, and the seat is given an initial displacement and released. One can then look at the acceleration response, calculate the maximum overshoot, and use that to calculate the damping ratio (a). Cushion mass is found by using a standard spring force scale, and suspension mass is obtained from manufacturer data. Finally, all the general construction features of each seat are observed, such as height adjust, fore-aft adjust, etc. In the Phase T! testing, the experiment must minimize the effect of outside influences on the evaluation of the seat. Consequently, each of the seats tested during this experiment must be subjected to the same conditions (road surfaces, speeds, driver position, etch. There are seven seats for the testing, numbered I-7 (Table D4.] below). At this time, this table should not and will not be released to any seat manufacturers or anyone who isn't affiliated with this project. However, the project investigators do intend to publish these results at a later date. Figure D-.2 - D-~.8 include photographs and general kinematic schematics of each seat and suspension linkage. Note that all seats are standard air suspension systems, except Seat #2, which is an active control air suspension (continuously varies air pressure to pneumatic spring) , and Seat #5, which is a height adjustable rigid support system. D-S

OCR for page 313
Table D4. I: Key for Seat Numbering TABLE INTENTIONALLY REMOVED. For each experimental trial, the seat is placed on an aluminum mount with a universal bolt configuration for all seven seats at the center of gravity of the Chevy bus. (See Figure D-.9 for schematics of the mount). It was found that the natural frequency of a 40 ft New Flyer bus chassis was very close to the natural frequency of the Chevy chassis (Belfiore, ~ 992~. Therefore, accelerations passed to the seat base at the center of gravity of the Chevy bus are similar to that of a 40 foot transit bus. As illustrated in Figure D- ~ . ~ 0, a total of five accelerometers are utilized, two for the seat and three for the ride simulator. One accelerometer (#!~ is placed on the top surface of the mount near the base of the seat. Another accelerometer (#2) is placed on the rigid support underneath the seat pan cushion and as close as possible to the location where the operator would sit as recommended in ISO 2361. The other accelerometers are placed on the two masses suspended in the human ride simulator (#4 on lower mass, #5 on upper mass) and on the rigid frame of the simulator near the molded base (#3~. Figure DO .l ~ shows a photo of D-6

OCR for page 313
the ride quality simulator in a seat. Also, for those seats which have an adjustable stiffness in the air spring, the seat will be adjusted to a midride setting. This will be midway between the minimum and maximum heights when the air in the spring (spring stiffness) is varied. Finally, all seats will be placed at an air pressure of ~ 00 psi, which is in the operating range for all the seats. The T/4 mile durability track at P.T.~. is used for the rough road surface simulation, and the ~ mile fuel economy track is used for the smooth road simulation. A trial on the durability track consists of one lap counterclockwise around the track (Lane ~ and Lane 6, elements ~ through 14) at the normal posted speeds (See Figure D-~.12 - D- ~ . ~ 8 for durability track elements and speeds), all starting at a common point, marked in Figure D- ~ . ~ 2. On the economy track, a trial consists of one lap counterclockwise around the track at thirty-five mph, all starting from a common point (Figure D-~.191. The testing consists of three trials on the durability track and three trials on the economy track. Each of the seven seats are tested with the ride quality simulator on each seat for all trials. To help insure repeatability in the experiment, the same person drives the bus for all trials. Further, the seats are tested in as small a time frame as possible in order to eliminate the effect of temperature variability on the road surface and to a certain extent, tire pressure. After the data is collected, it is run through two filters. The first filter is a low pass Chebychev Type ~ filter at 30 Hz to filter out the high frequency data. Because of the human resonances at 4.25 Hz and 7.5 Hz, only frequencies up to 20 Hz are important for this study. The second filter is a frequency weighting filter as presented in ISO 263 I. D-7

OCR for page 313
From the resulting acceleration data, several methods of analysis are used, including both time series and frequency domain approaches. The time series approach will include measures of r.m.s., peak value, crest factor, r.m.q., vibration dose value (VDV) , and SEAT%. For frequency domain analysis, transmissibilities for the accelerometer locations relative to the floor and power spectral densities of each accelerometer location are investigated. This allows a comparison of the seats to determine what seat isolates the most vibration from the body at the various locations and at the frequencies 4.25 Hz and 7.5 Hz, indicating the "best" seat to use in the bus operator workstation prototype. The following table (Table D4.~) summarizes the procedure for the experiment, with a listing of accelerometer locations on which each method will be applied. D-8

OCR for page 313
Table D4.2: Summary of Data Analysis Seat Model: Trial: (1, 2, 3) Durability Track Economy Track Measurements l (Accel. Channel (Accel. Channel l l Application) I Application) 1~1~ - acre! vs Lime 1; 2; 3; 4; 5 1; 2; 3; 4; 5 RMS 1 1;2;3;4;5 1 1;2,3;4;5 i Peak Value 1 1;2;3;4;5 1 1;2;3;4;5 Cmsr ~r I: 2 : 4: ~1; 2; 3; 4; 5 j RMQ 1 1;2;3;4;5 1 1;2;3;4;5 i VDV 1 1;2;3;4;5 1 1;2;3;4;5 S~19o I,' 1~3 PSD* ~1 1; 2; 3; 4; 5 1 1; 2; 3; 4; 5 Transfer Function* l 1,2; 1,3; 1,4; 1,5 | 1,2; 1,3; 1,4; 1,5 * For the durability track' the frequency analysis procedures can only be applied to one specific element of the track at a time. See Section D.5. l D-9

OCR for page 313
D.5. Results / Discussion The results from Phase T are shown below in Table D5.l. Various features such as tore-aft adjustment, seatbelts, armrests, etc. are listed, as well as seat parameters such as cushion stiffness, suspension damping, etc. Note that these criteria are important, but doesn't automatically rule out a certain seat, because the potential exists for retrofitting a seat to meet the required adjustment (i.e. fore-aft adjust, etc.). Table D5.1: Seat Comparison T 1 T 2 ~ Seat 3 1 4 y y N - - N N Auto 1~.3 ~ 1 7 C Feature 5 1 6 N y Lap y 3 y - 9.52 11.4 36.29 7 - N Lap y Armrest | Headrest Adjustable r Seatbelt Seat Back Tilt Lumbar Air Chambers _ Seat Back Side Air Chamber Pan Side Air Chamber _ Height Adjust (cm) _ Fore-Aft Adjust (cm) _ N y Lap y 2 y N 12.4 24.i 38.56 1.59 37.1 0.741 8.48 _ 2.36 _ 7 N - Lap - y - Lap - N y N y N N 2 1 2 2* y N , _ l _ | N _ 12.7 24.0 40.82 N N 6.99 15.2 38.10 9.21 30.3 38.56 8.32 13.~ Suspension Mass (kg) _ Number of Shocks _ Cushion Mass (kg) _ _ Cushion Stiffness (kN/m) _ Susp.Damping Ratio (if) _ Suspension Stiffness (kN/m) Natural Frequency (Hz) 1 2 1 1 1 3.15 20.9 0.440 3.18 .49 8.9 2 1.80 20.1 0.434 7.37 _ 2.20 8.9 1.36 ~ . ~ {. 1 0.59: 0.89 2.60 _ 5.1 _ 3.15 3.15 19.6 2.27 18.0 ~1 8.38 2.36 11.4 _ 30.4 _ _ 0.50: _ _ 8.60 _ _ 2.62 8.9 Cushion Thickness (cm) * Non-air lumbar chambers D-10 8.9

OCR for page 313
Table D5. I: Seat Comparison (cont.) Seat | Cushion Bottom l 1 No rigid bottom, steel plate on seat suspension 2 No rigid bottom, steel plate on seat suspension 3 No rigid bottom, steel plate on seat suspension 4 Metal plate Steel skeleton around edge 3 belts along bottom Steel skeleton around edge 3 belts along bottom l Plywood Misc. - Electronically controlled Pan can extend forward Extra 5.08 cm of fore-aft when at full height Pneumatic control of fore-aft lever Height adjusts from weight Pan rear & fore raise/lower Pan fore raise (~2.54 cm) Rigid Suspension Pan fore raise (~2.54 cm) For Phase TI of the testing, several methods were used, as stated above. The time analysis methods include rms, rmq, vdv, and SEAT values. Figure D-.20 - D-.24 show line graphs of the values obtained for each seat trial per location (floor, frame, etc.) for the I" random chuck holes element (Figure D-. of the durability track. Since each element is designed to invoke a specific response, using the entire durability track trial is not appropriate. The chuck holes element was used because it most closely simulates rough road conditions and results in nearly vertical motion of the seat and ride simulator. The other elements invoke more roll and pitch motion in the system, which complicates the results. Next, the economy track line graph results are shown in Figure D-~.25 - D- ~ .29. These two sets of data show at a certain location, such as dummy frame (essentially the vibration felt by the seated person), what acceleration levels are present for each seat. Dot!

OCR for page 313
10' 10 lo-l 1ol loo lo 1 SuspensionIFloor it, 1o1 10 10-1 [: ummy Lower MassIFloor 0 ~ 10 15 0 ~10 15 Dummy FramelFloor . I . ~_\ 'O ~ Frequency (Hey 1o1 10 Dummy Upper MassIFloor 1~ ~~\ 1 10-' ~. I 10 16 0 ~ 10 Frequency (Hz) Figure D-~.47: Seat 2 Average Transm~ssibilities, Economy D- 72

OCR for page 313
Floor 0.04 O. 02 w o 0.04 0.01 I 0.005 O. Dummy Lower Mass 'O ~ 10 Suspension . . 0.02 ~ ~: ~ odor ~ 15 0 ~ 10 15 Dummy Upper Bless . . 0.04 ;4002k~;3 002~;~ 0 ~ 10 15 Dummy Frame ,~ : O ~ Frequency (Hz) O ~ 10 15 Frequency (Hz) 10 15 Figure D-~.48: Seat 3 Average PSD Functions, Economy D - 73

OCR for page 313
10 10 lo-l 10' 100 lo l SuspensionIFloor . 1 ~ - ~ ~,:~ . . O ~ Dummy FramefFloor 10 100 10-' 10 15 0 Dummy Lower h1assIFloor it, ~ 10 15 Dummy Upper MassIFloor 1 ~ 10'1 - 1~ 10~ 1.0 ~ 0 ~ 10 15 0 ~10 Frequency (Hz) Frequency (Hz) Figure D-1.49: Seat 3 Average Transmissibilities, Economy D - 74

OCR for page 313
Floor 0.04 at, 0. 02 ~ 0.01 0.02 0.02 ~ 0.01 . . O __ 0 ~ 10 15 Suspension 0.04 0.02 0.1 0.05 Dummy Lower Mass ,~, 10 15 Dummy Upper Class few 0 ~ 10 15 0 Dummy Frame -1 ' ' 1 ~ 10 15 Frequency (H.) ~ 10 15 Frequency (Hz) Figure D-1.50: Seat 4 Average PSD Functions Economy D- 75

OCR for page 313
1 o 1 10 10 sol 10 1 o -1 SuspensionIFloor 10' 10 -1 Dummy Lower h~lassIFloor ~w ~v~x 0 ~ 10 15 0 ~10 15 Durnnny FramefFloor O ~ Frequelloy (Hz) 1ol 10 10 15 0 [: ummy Upper MassIFloor ; An: -11 ~ 10 15 Frequency (Hz) Figure D-1.5 1: Seat 4 Average Transmissibilities, Economy D- 76

OCR for page 313
0.02 ~ 0.01 o I 0.02 0.01 o 0.02 ;4 0.01 i Floor 'a O ~ :: ~ ~ . _ O ~ Dummy Frame . ., Dummy Lower Mass 0.04 10 15 0 Suspension 002 J: A o ~ 10 15 Dummy Upper Mass 0.1, 0.05 o 10 15 0 ~ O ~,~ 0 ~ 10 15 Frequerley (Hz) At 10 Frequency (Hz) Figure D-1.52: Seat 5 Average PSD Functions, Economy 15 D- 77

OCR for page 313
Dum~ny Lower hdassfFloor 10' 100~ 10 ' _ O S;uspensionfFloor Jo 10 10-1 10 15 0 ~ lol 10 10.1 Dummy FramelFloor , fit -. 1 0 ~ 10 15 Frequency (Hz) Figure D-1.53 10' .: I, 10 15 Dummy Upper MassfFloor 10 1Q ' ' o ~ 10 15 Frequency (Hz) Seat 5 Average Transm~ssibilities, Economy D - 78

OCR for page 313
Floor ~ 0.02 I ~ 0.01 0.02 ~ 0.01 o 10 Suspension it. O ~ Dummy Frame 10 15 0 005 ~. ~ O ~ Dummy Lower Mass 0.04 15 -O 0.05 10 15 Frequency (H~) . . 0.02 /~` ~ 10 15 Dummy Upper Mass ~ 10 15 Frequency (Hz) Figure D-.54: Seat 6 Average PSD Functions, Economy D - 79

OCR for page 313
10' 104 10 10' 10 10 SuspensionfFloor . . 'my 14 , O ~ Dummy FramelFlcor 10' . 0 ~ 10 Frequency (Hz) 10' 10 10-1 Dummy Lower 191assIFloor ' '' '1 ~~N 10 15 0 ~ 10 15 Dummy Upper MassIFloor 10 10 ' 16 0 ~ 10 Frequency (Hz) Figure D-1.55: Seat 6 Average Transmissibilities, Economy D - 80 .

OCR for page 313
Floor 0.05 I W o 0.04 ;4 0. 02 0.02 ~ 0.01 o Dummy Lower Bless art 0 ~ 10 Suspension :1 l o Dummy Frame ~ ~ - ~ O ~ 0.04 002 15 0 0.05 Lo _: - 10 15 Dummy Upper Mass 1 {~ o 10 15 ~0 ~ Frequency (Hz) 10 15 Frequency (Hz) 10 15 Figure D-:.56: Seat 7 Average PSD Functions, Economy D- 81

OCR for page 313
10 104 10-1 1ol 10 SuspensionIFloor ~ f O ~ 10 15 Dummy FramelFloor J r 10' 10 10 10' 100 _ . 10-1 ~ 0 ~ 10 15 ~ 0 Frequency (Hz) Dummy Lower hlassIFloor - ~ 10 15 Dummy Upper MassIFloor it. .: . _ L ~ 10 Frequency (Hz) Figure D-~.57: Seat 7 Average Transm~ssibilities, Economy D - 82 15