Datalogging RPM signal

[Brian Dessent]

Steve.Flanagan at VerizonWireless.com wrote:

> What inputs are used to calculate the HP / TQ numbers?  And how is the drag
> from wind added to the equation.

For what it's worth, Hugh MacInnes talks about the "road dyno" method in
his _Turbochargers_ book.  Basically, you need:

1. Weight of car + occupants
2. Tachometer
3. Speedometer
4. Accelerometer

Note that these should all be calibrated if possible.  I think all but
the accelerometer are pretty easy to find/calibrate.  I know you can buy
simple rugged accelerometers these days pretty cheap.  I also know of a
method to use a two-filament automotive incandescent light bulb as an
accelerometer.  (See end of post for details.)  You might be able to get
a rough idea of acceleration by taking the first derivative of the
speedometer signal, but it would not be as accurate as a real
accelerometer.

You can then get a torque / HP value for any RPM while on the road.  The
idea is to note the accelerometer reading at the desired RPM, presumably
while at WOT.  You then coast in neutral back past that same RPM and
note the deceleration reading; this corrects for wind, friction, road
incline, etc.  You add the two magnitudes to get the total
acceleration.  Repeat this for several different values of RPM to get a
curve.  Presumably you would also repeat each point many times so that
your numbers are averages.  Then the physics:

Force = Mass * Acceleration  (and)  Power = Force * Velocity

In English units (for non-US folks, please forgive me):

HP = Weight * Accel * Speed / 375

Weight is in pounds, Accel is in "G"s, Speed is in MPH.  The 375 comes
from converting all the units.  MacInnes gives the following example:

Weight of car + occupants: 3650 lbs
Road speed: 60 mph
Engine speed: 2800 rpm
WOT acceleration: 0.12G
Neutral decelration: 0.06G
(total acceleration = 0.18G)
Hence HP = 105.1 @ 2800 rpm.  And of course you can convert to torque
with the familiar:

Torque = HP * 5250 / rpm

-----

Light bulb accelerometer:

I can't remember all the details, but it goes something like this: one
filament is run at a constant voltage/temperature.  This causes a
convection current to be set up in gasses present in the bulb.  The
other filament is used in the same way as in a MAF: this convection
current inside the bulb causes a certain rate of cooling of the second 
filament.  This is measured by maintaining a constant current and
measuring voltage (and hence resistance, which is proportional to
temperature), or maybe the reverse, I can't remember.  Anyway, the
really neat thing is that if the bulb is accelerated in the right
direction relative to the orientation of the filaments, the path of the
convention currents changes, and the rate of cooling of the second
filament changes, and you can sense this change.  I will go back and see
if I can find the details if anyone's interested.  It wouldn't be the
most accrate thing in the world, but you could calibrate it with gravity
since it's only sensitive to acceleration in a particular axis.  My
source of this info was a lecture in a graduate level MEMS class by a
professor at UC Berkeley.  He mentioned it because you can use the same
concept except on a micron-scale to make accelerometers, as an
alternative to the standard cantilever-beam-flexture with capacitive
sensing.  But he actually demonstrated this with a 99 cent light bulb
and some instrumentation, specifically a frequency generator such that
the change in resistance would manifest as a changing pitch so the whole
class could hear.  Sure enough when he waved it around you could hear it
respond to acceleration.

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