On Thu, 06 May 2004 09:22:52 -0600, mikem wrote:

Toecutter and John, thanks for the informative responses. I am
interested in the methods. I have a few more ??

Sure thing.

Are the accelerometers two axis (X-Y) or single axis?

The units I am familiar with are all single axis.

If single axis, are they mounted so as to be sensitive left-right or
up-down?

Mounting is the semi-tricky part. The instructions say to mount
securely as close to the front of the engine as possible, with the
"axis" pointing as directly at the center of the crank, perpendicular
to it. I have mounted them off-set somewhat, just makes figuring out
where to add the weights (and the amount of weight) a little trickier.

Do all balancers use the second accelerometer on the accessory case?

The ones that I am familiar with are two channel units. They can be
used one sensor per engine on a twin, or as front/rear, just depends
where you bolt them on.

Do the engine mounts allow more up-down, or left-right shake if
the engine-prop are out of balance?

Here's the deal, with a strobe-type unit, the engine rpm is dialed
into the box (after fine-tuning it is an accurate way of checking the
ship's tach) and the strobe pointed at a piece of reflective tape
placed on the back of the prop blade.

With the engine running at the desired rpm, the stobe will "stop" the
tape somewhere in the circle of rotation, and the box will indicate
the measured amplitude of the vibe.

When the engine is shut down, the prop is rotated to the same position
it was in with the blade "stopped". The sensed "heavy spot" is now
aligned with where the velocimeter is mounted, the weight is added 180
degrees (directly opposing) the velocimeter.

Since the heavy spot is rotating 360 degrees during engine operation I
see no relationship in regard to left/right up/down vibration and the
engine mounts.

A Lycoming "dyna-focal" mounting is supposed to make it easier for the
engine to vibrate "around" it's center of mass, front to rear.
but we're sorta getting in over my head.

Is the criteria for "best subjective" balance some combination at what
the front and rear accelerometer see?

With regard to 1st order vibration, a qualifed yes.

In other words, if all you had to work with was the front accelerometer,
and you balanced to minimise its amplitude at the fundamental shaft
rotation, could you create a situation that the pilot may interpret as
making things worse?

Not typically, I've always assumed that since the front and rear are
for the most part flying in formation with the vibes 180 degrees out
of phase, reducing the front side will tend to reduce the rear side
level also. I know this is an elementary way to look at it, but I'm
kind of a simple guy.

Every once in awhile this is not the case. I am really not sure in the
case of 1/2 order vibration that it gets "worse", but
removing/reducing the 1st order vibe can make others that are present
seem "worse", or at least overall vibe level apparently un-changed.

The quickest way to do that is to reduce the vibe level at the front
(single velocimeter installed) to below .1 IPS. It is also the most
practical/economical method.

Is this "good enough" in most cases?

Yes. In most cases, when a single that shows up with anywhere from a
..4-.9 IPS level can be reduced to below .1 (sometimes is can be
reduced to next to nothing) using a front-mounted sensor, there will
be a definite change in the overall vibration level in the cabin. As I
indicated before, if the apparent vibration level stills seems high, a
rear sensor and/or a specturm analyzer can help point a finger in the
right direction, but a dy-bal probably ain't gonna help much.

..1 IPS is the maximum "balanced" level, if on the initial survey run
the level is above 1.0 IPS, adding weights/performing a dy-bal is NOT
recommended. Have only seen this a half dozen or so times. Yanking the
prop and checking the static balance is the first step, then a more
thorough inspection of the prop/engine until a problem is found.

The only "mechanical" explanation I can imagine for 1/2 RPM vibration
modes is that there is a weak cylinder, which in a 4 cycle engine only
fires every other shaft rotation?

That was my thinking also, but personally never came across a flat
cylinder, or any other combustion cycle-related mechanical problem to
back it up this theory. To clarify, I am not saying that you are
incorrect.

Makes perfect sense. I am an engineer and know enough about electronics,
instrumentation, mechanics, & digital signal processing to be totally
dangerous.

Heh, sounds suspiciously like me walking across the ramp pushing a
toolbox carrying a cordless screwdriver with a full charge (totally
dangerous).

Is the 337 necessary to do the ground testing, or only testing during
flight, or to drill a hole and put a bolt into the spinner backing
plate?

Um, er, the flight-testing is gray-area stuff.

The way I understood it at the time, is that if no procedure was in
place, adding (or removing) weights in any manner required a 337.

The thought is that the engine is required to be balanced within
certain limits via the repair/overhaul specs, likewise the propeller.

Since a typical dy-bal will likely put the static balance of the
propeller assy outside of published limits, it is a major alteration.
Likewise as you indicated, if no mechanical provision for securing the
weights is provided, this is also an alteration.

Since I was lucky enuff to have a prop shop next door, I had access to
current technical data (and parts) which allowed me to use the
propeller provisions for static balance weights for a lot of balance
jobs.

I have run into some balancer equipment salesmen at airshows who make
all sort of claims about "you dont have to know anything to use our
equipment" which automatically set off my "bull**** detector".

Heh, tend to have a relatively low tolerance for it myself.

Have only demo'ed the new digital stuff. If you've got a decent local
guy that thinks dy-bals are something he/she wants to get into, ACES
useta keep a couple of demo machines available for demo use. Am not
pimping for ACES, was just one of the digital demos that I remember
doing (was either the 2015 or the 2020). Just did a web search, and
here is a link to dload their operating manuals, if you're interested
in more info on how-to:

http://www.acessystems.com/manuals.htm

Again, it's been awhile since I used them, but they use a photo tach
pickup to spot/stop the reflective tape. After the survey run, the
machine tells you the moment/angular location of the correction weight
needed. The tape is lined up with the optical pickup, and the angular
reference is from this point. A calculator function allowed you to
input the arm (distance from the center of rotation) of the desired
weight mounting point, and would calculate the weight required. It
also allowed you to "split" the location, inputting the angular
locations of the available mounting points if one was not available in
the proper location. The solution would tell you how much weight to
put at each location. A second run would confirm the final solution.

For a basic single velocimeter balance job to below .1, the machine
will do just about all of the heavy-lifting. But beyond that, the tech
has to kinda know what's going on.

Appreciate your experience and willingness to share it

Any time. If I don't hang around here and throw in a little now and
them, I'm going to forget more about what I've learned in GA than what
I already have...

TC