On Mon, 3 Aug 2009 19:46:20 -0700, "Stu Fields"
wrote:


wrote in message
.. .
A bit of extra info on orthogonal balancing.

With the balance weights mounted close to one side of each
wheel the system balances out the radial unbalance component of
each wheel. However it takes no account of unbalance in the axial
direction i.e. the wobble if the wheel is not properly mounted
square to the shaft.

This is not normally a problem but, to avoid possible
trouble, be sure that the wheel mounting washers are accurately
square to the shaft and also dress true both the wheel periphery
AND the wheel sides before balancing.

In my initial experiments I loaded the voice coil with a
5/8" dia x 5/16" slug. This gave me a lot of trouble with the
100Hz torque variation signal from the single phase motor. This
is because I had accidentally made the "accelerometer" resonant
at 100Hz! I doubled the slug thickness, this dropped the resonant
frequency to about 75Hz and I had no further trouble.

Your speaker may have a different suspension stiffness so
it's worth checking the loaded resonant frequency. The easiest
way is to lightly tap the slug and to use the 'scope to monitor
the damped sinusoid that results.

Jim
We have three articles on balancing on the website www.experimentalhelo.com
One of the "No Phase" articles was written by Larry Meidel who is a well
known expert on helicopter vibrations. He showed how a balancing was done
with 4 readings on a three bladed helicopter. I modified his work for a two
bladed helicopter and one of the subscribers to the Experimental Helo
magazine, after reading the articles, balanced his tail rotor using just a
dial indicater. All of these articles are available on our website. They
are found on the left of the page down a little under special articles.

Stu

Thanks for the reference - these are useful articles. The
helicopter group three axis "amplitude only" method uses the
same basic approach as the orthogonal system. Pursued to a
conclusion both methods should yield exactly the same results.

The helicopter group method has the advantage that the result
can be a single balance weight or two weights at freely chosen
spacing - this gives the least total added weight. The orthogonal
system aims to finish up with 2 weights at 90 deg which may be,
up to root 2, heavier than the equivalent single weight.

The 3 axis system achieves this by a careful series of
amplitude measurements followed by vector calculation of the
results. The penalty is that it loses the quadrature independence
of the orthogonal 90 deg balancing axes - fine tuning by
cheerfully adding a bit of weight to see if the balance gets
better or worse is no longer an option.

The 2" speaker "accelerometer" I used is not suitable at
the lower RPM of helicopter blades - more mass and longer travel
is needed. I did some initial work modifying one of the self
powered flashlights which use a large rare earth magnet shaken
within a coaxial coil to charge up a couple of lithium cells.
This has along available travel and lots of output. However I
finished up with the speaker because it's a simpler modification
and all that is needed at grinder RPM.

Jim

On Aug 5, 10:46*am, wrote:

* * * The 3 axis system achieves this by a careful series of
amplitude measurements followed by vector calculation of the
results. The penalty is that it loses the quadrature independence
of the orthogonal 90 deg balancing axes - fine tuning by
cheerfully adding a bit of weight to see if the balance gets
better or worse is no longer an option.

* * * * * * * * * * * * Jim

I have not tried either method yet. My instinct is that the 3 axis
system could do fine tuning by first doing a balance using a fairly
large test weight. And after completing the balancing, do it all over
again with a test weight about 1/10 of the original test weight.

Dan