View Full Version : inclinometers (tilt sensors)? SUMMARY

02-23-2003, 10:49 AM
Thanks to all those that replied. The conclusion that I have reached based
on the replies and further investigation is that inclinometers by themselves won't work for measuring segment angles during gait. They are good for static situations, but not dynamic ones. However, it does seem possible to combine either inclinometers or accelerometers with gyroscopes to get accurate measurements during gait. I found the following web page particularly helpful:


My original posting follows along with the replies I received.

Dan Ferris
Human Neuromechanics Laboratory
University of Michigan

--On Thursday, February 20, 2003 1:50 PM -0500 "Daniel Ferris (University of Michigan)" wrote:

> I am looking for lightweight inclinometers or tilt sensors that I can
> mount on a person's legs during walking for real time feedback of torso,
> thigh, shank, and foot angles relative to vertical. It has to be real
> time (no post processing analysis) and has to measure segment angles
> relative to vertical, not anatomical joint angles. I have found one
> possible inclinometer that might work but they seem relatively expensive:
> N4 Inclinometers from Rieker Inc.($125.50 each).
> Does anyone else know of other possibilities that might be less expensive?
> Accuracy to the nearest degree would be acceptable.
> thanks,
> df
> _______________________________
> Dan Ferris, Ph.D.
> Human Neuromechanics Laboratory
> Department of Movement Science
> University of Michigan
> http://www-personal.umich.edu/~ferrisdp/
> mailing address:
> Dan Ferris
> Division of Kinesiology
> 401 Washtenaw Avenue
> Ann Arbor, MI 48109-2214
> e-mail: ferrisdp@umich.edu
> phone: (734) 647-6878
> fax: (734) 936-1925
> _______________________________


I might have seen something like that in the website of PCB Piezotronics.
Check it out.

Nick Stergiou, PhD
HPER Biomechanics Laboratory
University of Nebraska at Omaha
Omaha, NE 68182-0216
tel. 402-5543247
fax. 402-5543693
e-mail: nstergiou@mail.unomaha.edu

If you are looking for sensors of any kind look at
http://www.globalspec.com/. Just put Inclinometers in the search and
follow it into the specific search engine. You'll find a lot of them.
Good hunt.

Meir Bar-Tal


You can buy accelerometers from Analog Devices (www.analog.com) which
are capable of measuring both acceleration AND tilt. They sell them as
components but they have "demo" boards which help you interface them to
other equipment. If you are interested, drop me a note. My email
address is: rkumetz@hallux.com. (Don't hit "REPLY")

I have been looking at using accelerometers to monitor the movement and
performance of archers.

Ron Kumetz
Director of Engineering
The Hallux Group


We have used gravity sensitive inclinometers (about the size of a quarter) for
developing our Ergonomic dosimeter for a construction worker project. It
measures angles directly but you will have to calibrate it first. I forgot the
cost. The sensor (Micro ARC Transducers series 0728) was purchased from
Fredericks Co. Huntington Valley, PA.

If interested here is the ref of our article:

Bhattacharya, A, Warren, J Teuschler, J., Dimov, M and Lemasters, G.,
"Development and evaluation of a microprocessor based ergonomic dosimeter for
evaluating carpentry tasks" Applied Ergonomics Vol. 30: 543-553, 1999.

Best of luck.
Amit Bhattacharya, PhD, CPE
Professor, Environmental Health
Adjunct Professor, Biomedical Engineering,
Industrial Engineering & Physical Therapy
Biomechanics-Ergonomics Research
M.L. #056
3223 Eden Ave
Department of Environmental Health
University of Cincinnati Medical College
Cincinnati, OH 45267-0056

Tel: 513-558-0503
Fax: 513-558-8860
Lab Home Page: www.uc.edu/bert



Have you considered angular rate sensors, as well? I have used them in the
past (4 years ago) to obtain angular position information (via integration).


Carmen P. DiGiovine, PhD
Postdoctoral Research Associate
-- University of Illinois at Chicago
Assistive Technology Unit
Institute on Disability and Human Development
Department of Disability and Human Development
Department of Bioengineering
-- 312-413-3113 (direct)
312-413-1555 (main)
312-413-3709 (fax)
312-413-1554 (tty)


Just out of curiosity, do you know the dynamic range of the inclinometer? My
impression is that it only works for the static situation.

-Ge Wu

Hi Dan,

I have been using a pair of inclinometers for some of the measurements in
our lab. They are from a commercial system for functional capacity
evaluation. I believe the internal sensor in the devices is from HP.
Unfortunately they would probably be quite expensive to purchase, so I too
would be interested in hearing about what you discover from your inquiry.

Incidently, one of the things I have investigated with our existing
inclinometers is the inertial sensitivity of the devices. The sensors are
clearly susceptible to inertial effects associated with acceleration and
deceleration of the limb segments they are attached to, and they can greatly
affect the output. There is a damping mechanism present to limit the affect,
but you do need to beware. I would be happy to discuss it further if you are



Peter V. Loubert PhD, PT, ATC
Associate Professor of Physical Therapy
Central Michigan University
Mount Pleasant, MI 48859 USA

Phone: 989-774-2396
Fax: 989-774-2908
Email: Peter.V.Loubert@cmich.edu


Hi Dan,

If you are looking for +/-1 deg accuracy with respect to the vertical during
walking I am pretty sure an inclinometer won't work well for you. This is
due to the linear accelerations during any motion (especially walking and
running) will add to the acceleration due to gravity and 'pollute' your
inclinometer measurement, resulting in large errors during e.g. swing.

If you need to reach this accuracy you will have to mix accelerometers (tilt
sensors) with rate of turn sensors (gyroscopes). If you are interested you
can have a look at our motion sensors (http://www.xsens.com/index_mt9.htm),
they could give you the 1 deg accuracy in real-time, but they are more
expensive than inclinometers.

Best regards,
Per Slycke (per@xsens.com)


Unidirectional accelerometers aligned vertically on the segments might do
the trick.
-- Dr Gabor Barton (MD)
Senior Lecturer in Biomechanics
The Research Institute for Sport & Exercise Sciences,
Liverpool John Moores University
Room 2.51 Henry Cotton Campus, 15-21 Webster Street, Liverpool, L3 2ET
Tel: +44 (0)151 231 4333/4321 Fax: +44 (0)151 231 4353
E-mail: G.J.Barton@livjm.ac.uk


Dear Dr. Ferris

My idea is using ADXL202 from Analog devices. These are quite nice light
weight tilt sensors. Each should be about 10 USD$ each. You can ask for a
sample from AD. I asked but they didn't send to Turkey.

I can help you for its micropocessor controlled applications. Are you
planning a real time telemetric application?

If required range of tilt is out of the specifications of the sensor, you
may use 2 parallel circuits.

Murat Pehlivan MD, PhD
Ege University Medical School
Dept. of Biophysics
Bornova, Izmir Turkey

E-Mail: pehlivan@med.ege.edu.tr
phone&fax: (90-232) 3398020


Dear Daniel,

I think you can go to http://www.xsens.com and have a look. If you are
interesting, you can contact Mr Casper Peeters to have financial

Casper Peeters
Henglosestraat 705
P.O. 545
7500 AM Enschede
The Netherlands



go to http://www.select-controls.com/ and look at the disturbance and tilt

Good luck.

Al Vangura Jr.


Hi Dan,

In answer to your query, we bought an inclinometer for my robot project with Dave Shipman in SF. We had pretty good results with it. The only problem you might have is that it has approximately first-order low pass filtering effects. When we combined it with our gyro data, you will see that we were able to reconstruct the actual angle in space pretty well. Generally if you want real time angle with respect to vertical, inclinometers will do it but they will have a settling time after impact with the ground. Accelerometers will also work but then you have to figure out where the g vector is. A state estimator combining an inclinometer and accelerometers would work well, but it would take some time and effort to integrate them.


"2-D inclinometer (Applied Geomechanics Model 902). The inclinometer is mounted to the back of the sensor box. It directly measures roll and pitch
angles up to +/- 45 and outputs voltages in the 0-5V range. Although quite accurate, it is very sluggish, with a nominal time constant on the order of 150 msec. Furthermore, it cannot be trusted when subjected to violent jarring as occurs during the collision of the robot's foot with the
ground. Its output can however be combined with the gyro outputs when accelerations are not too large. An inclinometer is required since the rate gyros cannot determine their absolute orientation and integrating the gyros alone introduces a slow drift in orientation error."


To unsubscribe send SIGNOFF BIOMCH-L to LISTSERV@nic.surfnet.nl
For information and archives: http://isb.ri.ccf.org/biomch-l