View Single Post
  #15  
Old March 13th 05, 08:14 AM
sleepy6
external usenet poster
 
Posts: n/a
Default

In article ,
says...


"Morgans" wrote in message
...

How does the notch help in telling the rotational position?
--
Jim in NC


Without giving too much away...
The notch depth is more significant than any vibration. If using a
proximity measurment, it is easy to determine when the notch occurs.
The notch is aligned with a known position (engineer inserts notch whe
n
pistons/crank is in known position). This gives a phase reference. Y
ou can
compare where in the cycle a disturbance (knock, rub, etc) occurs with

relation to the notch. On recip applications, often a multi-tooth whe
el is
used in addition to a phase reference for more accuracy.

Adam
N7966L
Beech Super III


I love puzzles.

Start out with a picture in your mind of a perfectly round crank flange
with a dial indicator rideing on the outside edge at the top as you
slowly rotate the crank. The indicator will not move because the
flange is perfectly round.

Under viberation, the crank will move up and down (within the limits of
the bearing clearance) and the dial indicator will show how far the
crank moves.

Now replace the dial indicator with a precision proximity detector.
You get a steady flat line electronic signal until you get movement up
or down. Then the signal will change at the same frequency as the
movement and change amptitude with the amount of movement.

Now add a notch across the edge of the flange so that the notch is seen
by the sensor when #1 piston is at top dead center.

That notch causes a much larger timing blip in the signal than any
crank movement will cause. By measureing the time between the timing
blips you can determine the amount of rotation after a timing blip that
you see an up or down crank movement event.

In other words if you see a minor blip exactly midway between timing
blips then you know the crank moved up or down when #1 piston was at
bottom dead center in our example. The amptitude of the minor blip
will tell you how far it moved. It could even be set up to tell you if
the crank moved up or moved down by the polarity shift of the signal.

Of course this simple setup will not show side to side movement. That
would require a second proximity detector 90 degrees from the first
one. The 2 signals would require integration but would then show all
movement in the rotational plane with the proper signal processing and
calibration.