The Traffic Accident Reconstruction Origin -ARnews-
|
The Traffic Accident Reconstruction Origin -ARnews-
|
|
I thought my comment might spark a question of two but wow what a response!
OK here's the deal. I have noticed quite a bit of controversy within the
reconstruction community with regard to the use of "sleds." Witness several
comments within this and last year's threads. Well, I've had some problems
with consistency in sled values versus the values from other methods too. So
I decided to do some serious research into sleds and, should it still be
worthwhile, maybe get my data into book form.
Boy I've found quite a bit of data that either we've all forgotten or probably
never been told or otherwise never heard. Firstly, Ed Phillips was kind enough
to supply me with a copy of the orginal Broshears "Drag Sled" manual (Drag Sled
by the way is a registered trademark)to get me started.
I, like many others, simply thought that you get any convenient object with
some considerable weight, glue some tire to the bottom, and get a farmer's
spring scale to pull it with and get a value. Or maybe get an old tire,
fill it with concrete.. and you know the deal. Just follow the Coulomb
model for friction. Well... it just ain't that easy!
I've found that two of the most prolific mistakes involve (1) failing to
determine and pull a flat-bottomed type of sled at its "proper pull angle," and
(2) assuming that a sled will automatically mimic the results of other methods
like fifth-wheels, accelerometers, test skids, etc.
As to number one.. Think about it... if you apply a force at any point which is
distant from a pivot point, you introduce a torque! Most drag devices have
pull eye-bolts located some distance above the pivot point (the leading edge
of the tire pad). That extraneous torque causes an extra downward force. The
extra downward force manifests itself into a higher drag factor value! I
didn't think this would be that big a problem until I did the calculations and
tried it. Determination of the "proper pull angle" solved not only a large bit
of the inconsistencies I had experienced but also got rid of a large part of
the "hopping," etc. (The spring-scale version of the original Drag Sled did
have such engineering built in but its manual simply mentions it.. it didn't
explain why or how to do it yourself!)
And then everybody tries to "calibrate" their sled with some other method like
a VC-2000 or whatever. The problem is that almost no two of the available
drag factor determination methods match precisely. Most differ because of
mechanical delay (pre-lockup braking inefficiency duration)... some incorporate
it in their values for drag factors, some don't. (See Jerry Eubank's SAE paper
which compares many of the methods.)
A drag device cannot show the effects of mechanical delay in its values and,
as a result, will probably give results which are a little higher than methods
or devices which do (like a MacInnis 5th Wheel or VC-2000 equiped with the
brake light activation harness).
I've devised what I call a "pull truck" which simply allows the user to
control the pull angle consistently. This in turn allows the user to VERY
closely and consistently mimic whatever method or device is chosen as
a reference.
Then there's tire hardness to consider.... and the effects of hydroplaning...
and then slope when all tires are not locked... My research has dug up a lot.
My book should soon be finished and maybe I can get the guys at IPTM to publish
it.
Ed Livesay
elivesay@alltel.net
For example, to continue this discussion look for a thread titled
Coefficient of Friction survey.
If this thread does not exist in the current archive, you can begin another one by using that title.