|
|
|
Project 4: Engineering
Solutions for Optimal Wheelchair Suspension
Researchers:
Grigor Kerdanyan, MSME, PE
Craig Newsam, DPT
Philip S. Requejo, PhD
Robert Waters, MD
Somboon Maneekobkunwong, MSME
Project Goals:
To objectively measure wheelchair comfort and the effect
of the chair's design on wheelchair comfort, and use obtained knowledge to design a suspension
chair.
Summary:
Wheelchair comfort has been of growing concern as users
become increasingly active over a longer lifespan. Wheelchair ride quality has a great
effect on back and neck pain for wheelchair users. It is also suspected that the added
loads of a harsh ride can contribute to pressure sores. Wheelchair manufacturers have
addressed this concern with designs using innovative frames and suspension systems. Our
intent is to develop an objective way of measuring wheelchair comfort, to compare the
performance of different wheelchair designs. This will also give us the ability to compare
ride quality using different chair setups such as tire pressure. This acquired knowledge
will also be used to design an optimized wheelchair suspension system.
Progress:
Instrumented
chairs:
We secured generous manufacturer donations of Quickie GPV and XTR, Colours Boing, rigid and
suspended Invacare A4, and TiLite chairs, and subsequently instrumented these chairs to collect
data. The instrumentation collects forces and moments acting on the rider through the seat and backrest in all (x,y,z) degrees of freedom. The chair seat is connected to the chair entirely
by load cells, as seen in Figure 1. To collect acceleration data, we have three 2-axis
accelerometers: one on a bicycle helmet worn by the test subject, one mounted on the seat,
and one mounted close to the hub of the wheel. This setup provides information of the disturbance
entering the chair through the wheel hub, reaching the seat, and finally reaching the user. The
data is collected using a laptop computer equipped with data acquisition cards, providing mobile
acquisition potential.
Test Fixture:
A fixture was developed to perform two types of controlled disturbance tests, Figure 2. The
first test simulates uneven pavement. In this test, a ball/socket joint restrains the front
of the instrumented chair while the rear wheels are placed over a rotating drum with a small
obstacle fixed to it. The speed, size, shape, and orientation of the obstacle can be varied.
A transfer platform was built to elevate the subject in their chair to the level of the
instrumented chair. A second disturbance test utilizes this transfer platform to simulate curb
drops. The platform was designed to drop and hit the floor with controlled and measured speed
while the subject is on the platform in the instrumented chair.
click image for a larger version.
Testing
Procedure:
During each test session we test the same rigid chair against one of the suspension chairs.
This way we always have the data from the rigid chair as a reference, which allows comparison
of data from different test sessions. The subject is weighed, and transferred on to the test
chair. The Initial test chair (rigid or suspension) is alternated between subjects. First, we
find a drum speed simulating their own pushing speed on uneven surfaces and collect the data
for about twenty seconds. Following the drum test, the subject moves back on to the elevation
platform for the drop test. We set a drop speed that is comfortable for the subject and collect
data for five drops. If the subject is comfortable going down curbs, we collect the data for five
curb drops. The subject is instructed to keep each drop as identical as possible. For testing of
the second chair the drum speeds, the drop speed, and the curb height are unchanged.
Test Results:
We have tested eleven subjects so far with the Colours Boing and Quickie GPV. The preliminary
data analysis have given the following results.
The drum test simulating uneven pavement demonstrates a clear performance difference between the
two chairs. Figure 3 and Table 1 show vertical force and vertical head acceleration, for two
bumps, collected with the Boing and GPV, with identical cushions, tires and tire pressure. The
obstacle was hitting the tire at 2.5 mph. The weight of the subject on the seat was 134 lb, the
remainder is on the foot rest. (Subject's total weight was 149LB). A separate drum test was
performed for the rigid chair in which the tire pressure was varied. For a 155 pound subject
experiencing a given bump, the peak forces ranged from 183 lb at 50 psi, to 194 lb at 100 psi.
This test confirmed, repeatably, both the instrument sensitivity and the importance of tires as
suspension elements.
click image for a larger version.
The curb drop simulation
does not provide as obvious a difference as in the drum test. The reason for this is that in the
drum tester, or in an actual curb drop, the rear wheels take most, if not all, of the impact, while
in our simulation the chair is sitting on the platform with all four wheels, and the load is
distributed between the rear wheels and the front casters. For this reason, the outcome of the test
was affected by the seating position of the subject. Also, some subjects could tolerate very low
drop speeds, .5 mph impact velocity range, so that these tests produced very similar loads and
accelerations between the two chairs, as the small forces are not sufficient to overcome friction in
the suspension components. However, in the tests where subjects could tolerate higher drop speeds,
1.0 mph range, the difference in force and acceleration could be seen in Figure 4. There is about a
67 % difference in force and an 87% difference in acceleration between the two chairs.
click image for a larger version.
The results of a real
curb drop test were the most puzzling, since visual observation of the test clearly indicated
that the uspension was working, yet the measured peak force was very close between the rigid and
suspended chairs. More careful analysis, calculating area under the force vs. time plot to yield
momentum, revealed that subjects were going of the curb more aggressively with the suspension chair
than with the rigid chair. Thus, analysis showed that the impact momentum of the Boing chair is
higher than that of the Quickie, Figure 5. This suggests that the subjects may have a peak force
level that is tolerable, and that during the test, instead of keeping the drops identical, riders
may tend to drop so as to produce the same force level. The data of subject 6 indicates that the
suspension chair produced the same peak force even though the impact momentum in the suspension
chair was 89% higher than in the rigid chair.
click image for a larger version.
Suspended System
Design:
We have been collaborating with Colours, in redesigning the going suspension system. The goal was
to simplify manufacturing and reduce the weight of the chair. Most of the design work has been
completed Figure 6 illustrates a computer rendering of the new design
click image for a larger version.
Future Goals:
Complete the testing of Boing and start testing the Quickie XTR,
Invacare A4, and TiLite chairs.
Develop a systematic way of analyzing and evaluating the
collected data and comparing the different chairs.
Create computed dynamic model of
chair/rider system to simulate test results and use the model as a tool to assist in designing of
chairs.
Build a prototype of the Boing redesign and possible other suspension concepts.
Rancho REP ||
What's New? ||
About Us ||
Our Projects ||
REP Publications
Contact Us ||
Frequently Asked Questions ||
REP Links
Web Administrator
Last modified: Jul.
6, 2004
|