View Full Version : Summary: Vibrations from the Icicle

Adam Shortland
01-07-2002, 07:00 AM
Thanks to all who replied to my recent message concerning vibration
pollution from a prospective building site close to our gait laboratory.
The problem is a little rhetorical by all accounts - it appears we will not
have the definitive answer until the building work commences. Some of the
replies highlight areas of concern that had not passed between my eardrums!
The original question is repeated below along with those kind and considered
replies. Thanks also to Forrest who phoned me from AMTI in the US to discuss
the problem.


Original Question:
On a site less than 10 metres from our clinical gait lab at Guy's Hospital
in London there are plans to construct the tallest building in Europe. The
architectural realisations depict a huge chard of ice pointing towards the
heavens (apologies to humanists/atheists). The building at 62 storeys with a
large radiator on top (the last word in Eco-design) will be twice as tall as
our hospital (the tallest hospital in Europe).
Necessarily, there will be considerable noise, traffic disturbance and dust,
but our main problem will come from the vibrations of the building work. I
understand the concrete pillars that form a part of the base of the building
will be helically-driven to reduce vibration pollution.
I like vibrations as much as the next person except when they interfere with
my equipment! My questions are:
If the magnitude of the vibration at source was known, could we estimate the
level of vibration within our laboratory given some material specifications?
What level of vibration would interfere with the operation of our
cameras/forceplates (some cameras are fixed to plasterboard, others are on
Would long-term vibrations have a deleterious effect on our gait lab

We are in the process of making a representation to the local planning
committee and need as much information as possible to construct a case. I
know it's not a Biomechanics question but I'd be grateful for your thoughts.


>From Garry Allison:
This is a great opportunity to negotiate a sponsorship deal:-)))))
one small step next to the tallest building ... hmmm has to be
something in that for the marketing.... anyway back to biomechanics...

The impact on data acquisition is dependent on many factors (i.e the
resonance freq of the set up - floors - subjects on floor, the
support structures of cameras etc...). The strategy may be that you
document [very well] the noise in the system prior to the development
(ask them to pay for this quantification:-)). Repeated frequency
domain analysis around the clock on repeated occasions. This will
provide a basis for quantifying the vibration "problem" if one is to
develop following / during construction.

Ideally your case will become much stronger if you can present a
threshold where the use of the equipment becomes futile - i.e what
are the clinically significant changes you are trying to observe. If
you are able to define what is clinically significant then you will
be able to establish a threshold of accuracy at which point the
vibration makes the equipment defunct.

The concept of negotiations then becomes about the utility of the
equipment not the fact that the building site creates vibrations. In
this sense compensation and negotiations are related to loss of value
utility not just that the background RMS of the forceplate noise at
67Hz has increased by 117mV.

The real issue I think is the usefulness of the data and how the
vibration affect the usefulness....not just the fact that there is
more noise in the system after construction than before.
This opens a greater questions of clinical meaningful observations in


>From Joan Deffeyes:
Ten meters away?? Your equipment is the least of your worries - you
and your staff won't be able to stand the noise! I worked in a
building some years ago where a new building was being built about
100 meters away. It wasn't that big of a building, but because of the
sensitive equipment that was going to be put into the building, it
had to be built to be very vibration resistant. They pounded support
beams down to the underlying rock, and that made quite a racket. I
just had an office there - so I have no experience with what would
have happened to the lab equipment.

You might have better luck finding noise data from this type
operation, estimating what noise levels might be expected in your
lab, and seeing how that compares to worker safety regulations (these
would be OSHA regulations in the US - you need to find the equivalent
British organization that has jurisdiction at your location. The
people who do lab safety for you would probably know about this, but
might have to dig a little to find the noise regulations if that's
not something they normally deal with). You might be able to show
your planning committee that worker safety noise regulations would
not allow you to continue working while construction was taking
place. You might also work up a cost estimate for moving your lab to
a different location while the new building is being built - you
really don't want to be there if they are going to be doing the same
type of pile driving that was done near the building where I used to

As far as the vibration goes - I suspect that the vibrational
transmission may depend greatly on the type of rocks and soil between
you and the construction site - like earthquake damage depends on how
well the local geologic features transmit vibrational energy. You may
need to do some assessment of that before you can get an answer, or
find somebody who has. Have you tried talking to somebody in the
geology department at a nearby university? Usually you can find
information on the Internet about research at the department, and you
can look for someone doing seismology on a related topic.

Good luck!
Joan Deffeyes
>From Alan Morris

One solution might be to actually mount a very sensitive accelerometer to
concrete near your force-plate. This will measure the vibrations across all
frequencies. You should be able to apply a digital filter to remove these
vibrations from your signals using simple (i.e. bandpass filter) or complex
(i.e. wavelet) methods.

The other suggestions are the more typical vibration isolation suggestions
such as mounting the plate on a vibration-isolating materials (rubber pads)
that are sold for use with industrial manufacturing or measuring components.

I would say any possible fix depends on the bandwidth and the amplitude
of the external vibrations.


Alan Morris, M.A.Sc., P.Eng.
>From Neil Mansfield:

Usually I am posting here regarding biomechanical responses to vibration so
it makes a nice change to consider psychological responses, which is also a
research area for me!

Your best way forward is to consider a British Standard for assessing
vibration in buildings. BS6472 is designed to allowing prediction of human
responses (i.e. is does not consider possible structural damage, or noise).
It basically works by classifying vibration into 'impulsive', 'intermittent'
and 'continuous' and by considering the location: i.e. 'critical working
areas (e.g. some hospital operating theatres. precision labs)',
'residential', 'office', 'workshops'. The criteria are then related to a
multiplier to perception thresholds.

I suggest that you get a copy of BS6472 (you might have online access at
www.bsi-global.com if you have an Athens login). Another useful reference
might be K.C. Parsons Applied Ergonomics 31 (2000) 581-594 Environmental
ergonomics: a review of principles, methods and models which has a section
on building vibration.


>From Jim Walton:

Just a suggestion ...

You can probably put two or more markers on FIXED landmarks in
unobtrusive locations on the other side of the room and measure
the vibration being introduced into your video signals. Moreover,
you could quite easily subtract-out the vibration component if it
become a problem. All it would take is a "filter" to extract the
"apparent" 2-D motions of the reference targets before you try to
perform any tracking. Obviously, this is not the "preferred"
solution, but if the vibration appears to be a problem, the extra
effort may be worth the work.

BTW, we use this technique when we have several cameras on a moving
reference frame (a vehicle, for instance) ... it works. Back in
"the old days" :-) we used the same technique to "register" 16-mm
frames that passed through a sloppy film transport.


>From Arthur Magill
Hi Adam,

I don'y honestly know the answers to your problem, but it's certainly an
interesting one.
I doubt you'll be able to usefully estimate the amplitude of the vibration
from the source levels,
as it will depend so much on the structure of your building, and local
structure of the ground,
buildings nearby, etc. But what would be interesting is to measure it.
Make some background
measurements using the forceplates before work starts, and then repeat when
all hell breaks
loose, and let us know what happens. As to long term damage, I'd be very
surprised if you had
any trouble.

If you have any imaging labs on your site (MRI, CT) or microscopists, they
might have some idea.

Good luck,


Adam Shortland PhD,
One Small Step Gait Laboratory,
Guy's Hospital

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