Accelerometers, etc.

[Carl Meyer]

This is a multi-part message in MIME format.

---------------------- multipart/alternative attachment

  ----- Original Message -----=20
  From: Delwin D Fandrich=20
  To: pianotech@ptg.org=20
  Sent: Thursday, January 17, 2002 12:21 PM
  Subject: Accelerometers, etc.

  Recently the following statement appeared in one of the posts on =
soundboards:
  =93An accelerometer transduces variations in pressure to a=20
  voltage.=94 =20

  I agree that a better understanding of the testing methods would be =
helpful to those who don't have an electronic background.  Thanks for =
that post.  I'd like to add to that understanding.
  . When used to detect acceleration in a vibrating body or system the =
accelerometers signal can also be electronically integrated to indicate =
both the velocity and displacement of the vibrating object.

  An accelerometer is basically a microphone. A specialized one that =
first of all is a contact mic and usually calibrated for critical =
measurement.

  You alluded to this, but let me stress that an accelerometer is =
generally  an AC coupled device and a strain gauge is a DC coupled =
device.  AC coupled can also be called capacitive coupled, so no static =
measurements can be made.  For example, if you mount a  strain guage and =
an accelorometer to a beam and then suddenly bend the beam to produce 1 =
volt output from each sensor, the strain guage will stay at 1 volt as =
long as the bend remains, but the accelerometer output will decay to =
zero at the rate of the rc time constant.

  Attaching accelerometers to anything becomes problematic since the =
device affects the resonance of the item under test and adds mass.  The =
bonding is also critical as you say.  At some point errors can be =
introduced and its not easy to identify the problem.

  I hope this doesn't add to the confusion.  I have an accelerometer and =
a four trace scope.  I'll get some test made some day.

  By the way, the magnetic pick up for the sat is a form of =
accelerometer.  The plate moves.=20

  Carl Meyer  Assoc. PTG
  Santa Clara, California
  cmpiano@attbi.com=20

   principle accelerometers are quite simple. They consist of a base, a =
slice of piezoelectric material=97usually quartz or an artificially =
polarized ferroelectric ceramic=97and a seismic mass. The crystal is =
placed between the base and the seismic mass. When the assembly is set =
in motion the seismic mass (wishing to remain stationary) induces a =
mechanical stress in the crystal which causes it to generate an =
electrical charge across its pole faces. This electrical charge is =
proportional to the applied force. Obviously this charge is generated =
only when there is acceleration. When the accelerometer has achieved =
some steady state=97whether that be stationary or at some steady =
velocity=97there is no electrical charge generated.

  Accelerometers do not measure or indicate force or pressure. Force or =
pressure applied to any part of an accelerometer will have no effect on =
its output unless there is some acceleration involved. Or, I suppose, if =
the force is great enough to cause physical damage to the accelerometer =
housing in which case there will be some signal output but it may be =
quite brief.

  When appropriately mounted to the top of a vibrating body=97such as =
the bridge of a piano as it is being played=97an accelerometer will =
produce an electrical signal proportional to the physical vibratory =
motion of the bridge at the mounting point of the accelerometer. Since =
accelerometers are unidirectional they indicate acceleration (or =
velocity or displacement) in only one direction. Most tests I have done =
on the mobility of the bridge/soundboard assembly have involved studying =
the vertical motion (in a grand) of the bridge. Usually this involved =
mounting the accelerometer just beside the note in question. To measure =
the for-and-aft motion of the bridge the accelerometer is simply turned =
on its side so its axis is aligned in the direction you want to =
investigate. (Or you can spend a small fortune and purchase a triaxial =
accelerometer.

  There are several ways to mount an accelerometer on a test object. The =
least destructive is to simply glue it in place. The glue can be as =
benign as beeswax or as intrusive as some epoxy or CA adhesive. In =
trying to figure out the motion of the agraffe, a CA adhesive was used =
to mount an accelerometer to the top of a prepared agraffe (the top had =
been milled flat to provide a good mounting surface for the =
accelerometer). I=92ve used beeswax to temporarily bond accelerometers =
to the tops of bridges=97removing the strings from one adjacent unison =
usually provides enough room=97and in various places around the rim and =
structure. I=92ve also used studs screwed into holes drilled into the =
bridge between unisons such that the accelerometer ended up locked down =
tight with its base just above the strings. This lowered the resonant =
frequency of the accelerometer some but (according to the manufacturer) =
not enough to worry about for the tests I had in mind. Magnetic bases =
are available from most accelerometer manufacturers but I=92ve not used =
them. When looking at vibrations in plates I=92ve simply glued the =
accelerometer to the plate surface using beeswax.

  Del

  Delwin D Fandrich
  Piano Designer & Builder
  Hoquiam, Washington  USA
  E.mail:  pianobuilders@olynet.com
  Web Site:  www.pianobuilders.com

---------------------- multipart/alternative attachment
An HTML attachment was scrubbed...
URL: https://www.moypiano.com/ptg/pianotech.php/attachments/7c/5d/6e/26/attachment.htm

---------------------- multipart/alternative attachment--