View Full Version : Responses to: Shielding metallic interference for electromagnetictracking

10-11-2001, 11:06 AM
Hi all,

Thanks to those who responded to my question: Shielding metallic interference
for electromagnetic tracking. Following is my original question, a brief summary,
and the responses that I received.


Original post:
I am currently trying to collect some kinematic data from subjects walking on a
treadmill using an electromagnetic tracking system (Fastrak hardware with 6D
Research software). The problem is that the metal frame of the treadmill distorts
the coordinates. I have tried using a compensation mapping program
(SKMapper) that theoretically corrects for the metallic distortion, however in
some sensor positions the corrected coordinates are still distorted. As expected,
the worst distortion is of the foot sensor coordinates. I have tried moving the
transmitter around the treadmill without success.

Could anyone suggest a solution to this problem?

Could there be a way to shield the metallic frame of the treadmill so that it does
not interfere with the system?

Any responses whether they are from personal experience or from a theoretical
perspective would be appreciated.
Brief summary
There does not appear to be any way to shield metallic components in the
transmitters field. Some treadmills appear to be more problematic than others.
Most respondents suggested removing as much metal as possible from the
treadmill and replacing it with wood or another non- metallic compound. Two
companies were identified that have developed treadmills with no wooden parts
between the rollers, Biodex and Mobility Research. An email from Mobility
Research describing their "wooden" treadmill is included in the responses.

Responses to query (alphabetical order)
Jason Bodine wrote:

That is the behavior we generally get from metallic treadmills. I don’t know of a
coating that you could put over the metal to reduce its interference. You can try
contacting Polhemus’s technical support about that. Their tech guy is named
Dan Ratta, { HYPERLINK mailto:danr@polhemus.com }danr@polhemus.com.
The other solution we’ve done is to replace as much of the metal as possible
with wood. I know this is impractical sometimes, but if you can do it, it does help.

Jason Bodine
Software Engineer
Skill Technologies, Inc.
1202 E. Maryland Ave. Suite 1G
Phoenix, AZ 85014
(602) 277-7678
{ HYPERLINK mailto:jbodine@skilltechnologies.com }jbodine@skilltechnologies.com
Phillip Cheetham wrote:


Have you contacted our tech support? Chris Dible would be happy to discuss
this further ({HYPERLINK
“{ HYPERLINK "mailto:cdible@skilltechnologies.com" }mailto:cdible@skilltechnologies.com”}cdible@skill tech nologies.com). We have
successfully mapped several different treadmills.
Anyway I will give you my input as best as I can. We do have several systems
that are functioning well on a treadmill. It really depends on the treadmill and how
much metal there is in the base of the treadmill. If the platform under the belt is
sheet metal then you will probably not be successful but if it is wood you may
have better luck.
If you can modify the treadmill, try removing any metal cross bars that support
the base. Replace them with wood or plastic. Remove metal handrails and
replace with wood or plastic.

When mapping try to make the grid as dense as you can probably 2” to 4”
squares. Also position the transmitter on the side of the treadmill as close to the
persons legs as practical, at about knee height.

Unfortunately there is no magical shielding material. The best approach is to
move away as much distorting metal as you can and then map. We have now
developed a wooden treadmill will very few metal components that works very
nicely. This is available as a product that we sell. If you are interested let us

Phil Cheetham
Skill Technologies Inc.
Steven Cheetham wrote:

Dear Peter,
My brother Phillip passed on your email to me so I could give you the information
regarding the non-metallic (wooden) treadmill he had talked about.
There is a company in Tempe Arizona, called Mobility Research, that is
developing a wooden treadmill for use with Skill’s 6D-Research motion
measurement systems. Please contact them directly for product information
Mobility Research, Inc. -- 2211 W. First Street or P.O. Box 3141 (zip: 85280).
Tempe Arizona. 85281.
Atten: Amir Neregi. Tel: [1] 480-829-1727 Email:
{ HYPERLINK mailto:MoRe@LiteGait.com }MoRe@LiteGait.com
Stephen Cheetham
VP Sales and Marketing.
David Dychala wrote:

I am surmising that the power circuit to the equipment is properly grounded. If the
equipment ground is proper, you might try grounding the metallic shell to reduce
any stray signals that might be generated due to static electricty buildup at the
If there is a fault in the equipment ground, there is the possibility of a ground loop
between the treadmill and the tracking system.
Best Regards,
Dan Ferris wrote:
Hi Peter,

I went through a long search to work around this problem. The best answer I
found is that the amount of distortion depends on the treadmill. The amount of
metal in the treadmill, the dimensions of the treadmill, and the number of circular
loops of metal in the treadmill can all influence how well the mapping corrects for
the distortions. I would suggest trying some different treadmills (if possible) to
find one that minimizes the distortions. Even at the very best, we still had a very
limited space on the treadmill that allowed us to collect good data. If the subject
moved too far forward or too far back on the treadmill, we lost the sensors.

best of luck,
Richard Hinrichs wrote:

I’m afraid that you might not be able to get rid of the distortion. You need to use a
wooden or plastic treadmill! We have the same problem with our Skill
Technologies magnetic system.


Richard N. Hinrichs, Ph.D.
Dept. of Exercise Science and Physical Education
Arizona State University
Box 870404
Tempe, AZ 85287-0404
(1) 480-965-1624 (voice)
(1) 480-965-8108 (fax)
{ HYPERLINK mailto:hinrichs@asu.edu }hinrichs@asu.edu (email)
Steve Martinez wrote:

We have developed a wooden treadmill that has been successfully tested
with the Skill equipment at Samaritan Rehab Institute in Phoenix, AZ. The
treadmill is a custom order and is a completely modified version of our
GaitKeeper. There are no metal parts between the rollers. We will be targeting a
selling price of approx US$8,000. This model would have a speed range of 0 to 4
mph, in increments of 0.1 mph. Belt width is 18 inches. We are also planning on
having available at a later date a model with a speed range of 0 to 7.0 mph, and
a belt width of 20 inches.
This model will sell for approx $1,000 more. We do not have any type of
Australian approvals and I am not sure if that would be necessary before we
could sell there. Thank you for your interest.

Best regards,
Steve Martinez
Mobility Research
{ HYPERLINK http://www.LiteGait.com }www.LiteGait.com
Andre Przybyla wrote:
Dear Peter,
I’ve just started my PhD in Bristol. I am sorry but I can not help you
with your problem. However, I have also plan to use Fastrak in my future
research. Could you just give me some short info about that 6D Research
software which have you mentioned in your mail?

Thanks in advance
Steven Sprigle wrote:
hi; tough problem
a few thoughts based upon our experiences

1. enclosed metal is the worst, for example the rim of a wheelchair wheel; if the
treadmill has a handrail that makes a frame, perhaps you can replace it with
some sturdy PVC

2. we minimized distortion of metal by positioning the transmitter in a manner
so metal is never between the sensors and transmitter; since we work with
seated individuals, we had a few options, including mounting ‘in space’ and
played around with it above head and just anterior to the abdomen; perhaps you
could mount it on a non-metallic bar that crosses the treadmill, at waist height

3. keeping the sensors within a single hemisphere or even quarter sphere also
helped us; you have a larger range so this might not be possible

good luck, keep plugging away

stephen sprigle
helen hayes hospital
new york
Bryan St. Laurent wrote:

I have a great deal of experience with electromagnetic tracking systems
(Polhemus and Ascension) as well as experience with the software products from
Skill Technologies.

There is no way to shield the field that I ever discovered. I went so far as to
acquire materials that are used in nuclear power plants to shield the electronics
from the fields present there.

If you would like to talk I would be happy to help. As you are in Australia, perhaps
it would be best to communicate over Microsoft MSN Messenger. If you are
interested, send me an e-mail with a list of 3 times (specify if they are local times
for you or me in Arizona, USA) that you would like to chat and I shall be sure to
be on-line. Otherwise, if you would like to talk, you may telephone me.

Let me know if you are interested and I’ll send you my Messenger address
and/or my phone number.

Bryan St. Laurent

Have you tried getting support from Skill Technologies Inc?



You may want to place a call to Biodex. I had some preliminary
discussions them about two years ago regarding the construction of
electromagnetic (EM) “friendly” treadmills. I believe they have created
such a thing for about five or six groups around the United States.
Again, if I may be of any further assistance, please do not hesitate to contact me.

Best of luck,
Matthew Walker wrote:
Hello - I would be interested in seeing any potential solutions to your problem
since we are having a similar problem with an electromagnetic probe we have
made. When in the vicinity of any metal the coordinates make no sense
whatsoever. I am also interested in the SKMapper software that you have
mentioned - how does it work? Is it available as trial software? Anyhow - best of
luck, unfortunately I have no solution for the problem.

Take care,
M. Walker
MSc candidate
University of Guelph
Guelph, Canada
Nancy Womack wrote:
I worked with Flock of Birds EM system at the University of Michigan. Our
solution, and the only one that seemed to work, was to remove all the metal from
our testing setup. We rebuilt a passenger car seat with wood. This may not be a
viable solution for you, but that’s what we did.
Also, we found that the sensors were unreliable in a certain orientation
regardless of where they were in the EM field. It was when the sensors were
straight up and down, where the angles flipped around the 180 degree mark. For
that we changed the orientation of the sensors.
I feel your pain in trying to get rid of the “noise.” I spent weeks on this with little
results. Hope you get it worked out.

Nancy Womack

Peter Mills
Biomechanics-Dynamics Group
School of Physiotherapy and Exercise Science
Griffith University - Gold Coast
PMB 50 Gold Coast Mail Centre 9726
Queensland Australia
Ph: +61 7 5552 8357
Fax: +61 7 5552 8674
Mobile: 0410 503 415
Email: P.Mills@mailbox.gu.edu.au

"The path to success is always under construction"

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