Dear list members,
I posted a question concerning Force covariation with speed.

>We are working with force measuring horse shoes and conducted several
studies on the changes of the >force pattern/magnitude with speed of running
(trotting). There are may reports on the changes of >magnitude in the human
litterature, but I have not found any describing covariation/cross
correlation of the >vertical, fore-aft or medial-lateral. Can anyone help
me? A follow-up question is whether the cross-talk >between the transducers
of the forceplate (as most commonly used) changes with increasing force. I
am >aware of the problem with plate deformation and the effect on the point
of application, and thus more >interested in the behavior of the transducers.

I received several responses and wish to thank all.

A short summary will follow here.

In doing some work with our force measuring horse shoe on a variable speed
treadmill, we found that the vertical and horizontal forces tend to
correlate (though not to a large extent) with increasing speed. The horse
shoe which is a composite of may materials, among of which is polyurethane
(80 degree Shore A) between the two 3 mm shoes. This is to keep the weight
and height of the shoe to be equivalent to a normal horse shoe. Static and
dynamic test (though at a low frequency) demonstrate linearity. Cross talk
can then easily corrected for. The question arises concerning more rapid
loading to the shoe. We thought about buying or constructing a force plate
to in vivo calibrate the horse shoe when mounted on the horse's hoof.
Before making such an investment though we were interested in knowing if the
force plate is that much better in similar circumstances. Horses weighing
say 600 kg can easily exert a vertical force of 10 kN in a 200 ms period at
a slow trot. Force plates used in equine studies are usually large, and it
would be expected to deform to this loading. Would we expect a linear
deformation of the plate with speed and different rates of loading during
the stance? Are the transducers linear at high rates of loading?

Several responses referred to the need to control the elements of the
measuring system (from the strain gages to the multiplexing..). These
controls have been done and accounted for. Others commented on the need of
an accurate calibration matrix, which also has been developed (Instron
machine). Thirdly, the question of natural frequency also arose, which in
this construction is above 1000 Hz. A problem with calibrating the force
shoe with a force plate may also be due to the dynamical organ which it is
attached, the hoof as pointed out by Young Hui Chang.

Assuming the force plate/shoe is sufficiently rigid at low loading rates
can one also assume the same is true at higher rates, such as heel strike in

No comments were offered on the true correlation of the various forces. I
would however assume that the geometry and movements of the limb and body
would result in periods of high and low correlation. These periods would be
at higher rates of loading, thus returning full circle to the question of
calibration of force plates and our shoe and loading rates.

If my assumptions are incorrect or my questions are not properly formulated
please feel free to comment. Secondly if anyone has any further comment
please contact me. If further comment is of interest I can continue this
summary at a latter date, hopefully addressing the question of calibration
and loading rates.

Thank you again for your interest


Chris Johnston, DVM
Equine Biomechanics Lab.
School of Veterinary Medicine
P O Box 7011
750 07 Uppsala, Sweden