Shock Sensor Question

[Brian Dessent]

Steve.Flanagan at VerizonWireless.com wrote:

> The non contact displacement sensor is what I had mentioned as a potential
> optical to reflective device, I was just in the wrong frequency range.  If
> we can mount a device that outputs a signal to a reflective surface and then
> measure the time to return, we know the distance.  Sounds easy, however, it
> would require some technology that can calculate the time for the signal to
> travel?  How complicated is it to design a circuit that would output a
> voltage relative to a displacement?

If you do a Google search for "ultrasonic rangefinder" you'll get some
OK links.  However, the more I think about it I don't think this would
be the best way.  They probably don't have the resolution you would
need.  You said you're measuring a 4" shock travel, but what resolution
do you need?  Knowing this will help you choose the right solution.  The
ultrasonic transducers would probably only be able to give you a
measurement within an inch or two, using the stock electronics.  You
might be able to build a custom signal conditioner to get higher
precision but you could spent a long time working on such a thing.

> Linear Pots (the cheap ones) will not work at the Freqs that the shock
> moves, I have been told to record between 500 and 1000 Hz for these
> measurements.

Well, do what you wish, but that seems WAY to high to me.  The sprung
mass resonates at about 1 Hz and the unsprung mass resonates at around
10-20 Hz.  You could roughly estimate these frequencies with
sqrt(k/m)/(2*Pi), where k = spring rate and m = mass (make sure to use
consistant units.)  Vibrations above the resonant frequency are absorbed
almost completely by the tires.  See:

http://www.umtri.umich.edu/erd/roughness/every.pdf

I would measure DC (i.e. 0 Hz) to 50 Hz at most.  Since you said you are
interested in 1/4 mile times, I take this to mean you want to know
front/rear weight transfer, and this would be a very LOW frequency
signal since you're accelerating hard pretty much the whole time.

> How exactly does a strain gauge work?  Does the output voltage change
> depending on how hard you pull on it?

It measures strain.  :-)  Seriously, take any elastic material and apply
a force (stress) and it reacts by deforming (strain.)  For solids,
stress is proportional to strain and the constant that relates them is
called the modulus of elasticity.  A strain gauge is usually a thin
piece of conductive material that changes its resistance with strain. 
Used alone, it's pretty worthless.  What's so neat about it is you can
bond it to whatever structure you're interested in and it will measure
the amount of strain, and by doing some calculations you can relate this
to the stress in the structure, which in turn will tell you when/if it
will fail.  For example, you could bond a strain gauge to a steel i-beam
in a skyscraper to see how much it flexes in the wind to know if your
building is going to fall over in a hurricane.  Anyway, what I was
talking about would be to buy a small strain gauge, epoxy it to part of
the coilspring, build/buy a signal conditioner (simple instrumentation
op-amp circuit, the full scale signal is probably 20mV or so), and then
datalog the resulting signal.   It would be proportional to deflection
of the coil spring if you choose the right location for the strain
gauge.  Go to www.omega.com and find their section on strain gauges,
they should have lots of background and application hints.

Bernd's idea of mounting a camera on the car with a view of the
suspension and then analyzing it with a computer is looking more and
more attractive.  The hard part would be finding room to mount it where
it could record a picture of the suspension.

Brian

----- End of forwarded message from owner-diy_efi at diy-efi.org -----
----------------------------------------------------------------------------
To unsubscribe from diy_efi, send "unsubscribe diy_efi" (without the quotes)
in the body of a message (not the subject) to majordomo at lists.diy-efi.org