Tweet
Colleagues -
I posted a message about a month ago regarding force plate mounting
techniques that would mitigate the noise-to-signal problem we're
getting in trying to run very small animals across our equipment. As
promised (though a little late) here is a summary of responses.
Unfortunately, due to the constraints of our plate (Kistler 9281B) and
our lab facilities, it looks like the more viable solution will be
building our own strain-gauge transducer force plate.
1. Acquire a large steel plate to mount the force plate onto. It was
also suggested that we put a very thin piece of compressible rubber
beneath this mounting plate, in order to smooth out any imperfections
in the floor. We did try putting rubber washers under the contacting
feet of our current mounting frame, but this did not mitigate the
noise.
2. Epoxy a new mounting frame to the floor and then screw the force
plate to this frame. Here is an epoxy company that was recommended:
www.prosetepoxy.com.
3. Try mounting the plate in sand: "The sand is contained in a wood
frame, a little larger than the force plate, with sides lower than the
height of the force plate. This works much better than placing it on a
concrete floor, especially because floors are never flat enough, which
creates problems because one of the 4 legs of the force plate is then
not firmly in contact, and creates spurious transients."
4. Use a triaxial accelerometer to record the building vibrations near
the plate and remove them later from the signal.
5. Use a vibration isolation table: "These are often used in Mechanical
or Electrical engineering departments by experimenters working with
MEMS. They come in various sizes, but are typically around 30"x40".
When "on" the entire surface is lifted on a cushion of air and isolated
from the external environment."
And my personal favorite:
"I suggest that you drill the holes, secure the force platform properly
and don't tell anybody. It is easier to ask for forgiveness than to
ask for permission. That was your first mistake.
If you don't buy into that, buy or make a massive concrete block in
your lab and drill into that. The problem will come when you no longer
want a massive block of concrete in the lab. So be sure to build in
some slots that will allow the concrete block to be picked up with a
fork-lift/pallet lifter."
Thanks to everyone who responded,
Jesse Young
______________________________________________
Jesse W. Young
http://gibbon.anat.sunysb.edu/IDPAS/index.php?page=students/young
Interdepartmental Doctoral Program in the Anthropological Sciences
Stony Brook University
Stony Brook University Functional Morphology Group
http://gibbon.anat.sunysb.edu/FunMorG/index.htm
"He who sees things from their beginning will have the finest view of
them."
- Aristotle
I posted a message about a month ago regarding force plate mounting
techniques that would mitigate the noise-to-signal problem we're
getting in trying to run very small animals across our equipment. As
promised (though a little late) here is a summary of responses.
Unfortunately, due to the constraints of our plate (Kistler 9281B) and
our lab facilities, it looks like the more viable solution will be
building our own strain-gauge transducer force plate.
1. Acquire a large steel plate to mount the force plate onto. It was
also suggested that we put a very thin piece of compressible rubber
beneath this mounting plate, in order to smooth out any imperfections
in the floor. We did try putting rubber washers under the contacting
feet of our current mounting frame, but this did not mitigate the
noise.
2. Epoxy a new mounting frame to the floor and then screw the force
plate to this frame. Here is an epoxy company that was recommended:
www.prosetepoxy.com.
3. Try mounting the plate in sand: "The sand is contained in a wood
frame, a little larger than the force plate, with sides lower than the
height of the force plate. This works much better than placing it on a
concrete floor, especially because floors are never flat enough, which
creates problems because one of the 4 legs of the force plate is then
not firmly in contact, and creates spurious transients."
4. Use a triaxial accelerometer to record the building vibrations near
the plate and remove them later from the signal.
5. Use a vibration isolation table: "These are often used in Mechanical
or Electrical engineering departments by experimenters working with
MEMS. They come in various sizes, but are typically around 30"x40".
When "on" the entire surface is lifted on a cushion of air and isolated
from the external environment."
And my personal favorite:
"I suggest that you drill the holes, secure the force platform properly
and don't tell anybody. It is easier to ask for forgiveness than to
ask for permission. That was your first mistake.
If you don't buy into that, buy or make a massive concrete block in
your lab and drill into that. The problem will come when you no longer
want a massive block of concrete in the lab. So be sure to build in
some slots that will allow the concrete block to be picked up with a
fork-lift/pallet lifter."
Thanks to everyone who responded,
Jesse Young
______________________________________________
Jesse W. Young
http://gibbon.anat.sunysb.edu/IDPAS/index.php?page=students/young
Interdepartmental Doctoral Program in the Anthropological Sciences
Stony Brook University
Stony Brook University Functional Morphology Group
http://gibbon.anat.sunysb.edu/FunMorG/index.htm
"He who sees things from their beginning will have the finest view of
them."
- Aristotle