accelerometers. (Pedulous Rectification?)

Mark,

Thanks for your reply and it makes sense.

The MEMS accelerometer we've been looking at is a capacitive type.

Would you have any idea what could cause the second order non-linearity in
the MEMS accelerometer? (or at least point me in a direction where there
might be some papers or research concerning this?)

Since some of the applications for MEMS accelerometers is for tilt sensing,
this phenomenon is an issue.  Especially for high precision tilt sensing.

Regards,

Mike

-----Original Message-----
From: Mark Lemkin 
To: 'Mike Falck' 
Date: Tuesday, November 10, 1998 3:12 PM
Subject: RE: DC-bias shift in the presence of strong vibration in MEMS
accelerometers. (Pedulous Rectification?)


>Mike,
>
>It sounds like there is a second order (or even-order) nonlinearity
>in the accelerometer. Second order nonlinearity will result in
rectification
>of input vibrations into a dc offset. You can prove this by looking at
>what happens to a sine wave which goes through a ^2 function.
>
>Look at the sinusoidal vibration output of the accelerometer when subjected
>to an
>off-axis acceleration. There should be a component at the vibration
>frequency due to cross-axis sensitivity. Second-order nonlinearity would
>give you a component at the double-frequnecy. If the components at both the
>double and
>exitation frequencies are comparable, the off-axis vibration may be leaking
>into the output directly thorugh some sort of second-order nonlinearity
>coupling
>mechanism. Otherwise, the bias shift may be due to a 2nd order nonlinearity
>in
>the forward path.
>
>If you get a bias shift when you put a test vibration along the sensitive
>axis,
>you may be dealing with a nonlinearity in main signal path.
>
>Just out of curiosity, what type of accel. are you testing?
>
>Mark Lemkin
>
>-----Original Message-----
>From: Mike Falck [mailto:[email protected]]
>Sent: Monday, October 26, 1998 7:59 AM
>To: [email protected]
>Subject: DC-bias shift in the presence of strong vibration in MEMS
>accelerometers. (Pedulous Rectification?)
>
>
>This is a multi-part message in MIME format.
>
>------=_NextPart_000_0016_01BE00B6.7C745840
>Content-Type: text/plain;
> charset="iso-8859-1"
>Content-Transfer-Encoding: quoted-printable
>
>We are observing a DC-bias shift in the presence of strong vibration in =
>the non-sensitive axis of a MEMS accelerometer.  We have heard that it =
>is or could be related to a phenomenon called "Pedulous Rectification".  =
>We would like any information regarding this issue.
>
>The experiment we have performed is to take a MEMS accelerometer and =
>vibrate it in its non-sensitive axis from 50Hz-2KHz at 1.5g peak sine =
>wave.  The DC-offset or bias will change as much as 60mV at certain =
>frequencies.  The sensors we are using have a 0g bias at ~2.5V and have =
>a sensitivity of 2V/g.
>
>Since we are using the accelerometers in a precision tilt application =
>the 60mV appears as a significant error.
>
>Thanks in advance,
>
>Mike Falck
>Crossbow Technology
>Ph:  (408)965-3315.
>Email: [email protected]
>
>
>------=_NextPart_000_0016_01BE00B6.7C745840
>Content-Type: text/html;
> charset="iso-8859-1"
>Content-Transfer-Encoding: quoted-printable
>
>
>
>
>
>http-equiv=3DContent-Type>
>
>
>
>
We are observing a DC-bias shift in =
>the presence=20
>of strong vibration in the non-sensitive axis of a MEMS =
>accelerometer.  We=20
>have heard that it is or could be related to a phenomenon called =
>"Pedulous=20
>Rectification".  We would like any information regarding this=20
>issue.
>
 
>
The experiment we have performed is =
>to take a=20
>MEMS accelerometer and vibrate it in its non-sensitive axis from =
>50Hz-2KHz at=20
>1.5g peak sine wave.  The DC-offset or bias will change as much as =
>60mV at=20
>certain frequencies.  The sensors we are using have a 0g bias at =
>~2.5V and=20
>have a sensitivity of 2V/g.
>
 
>
Since we are using the =
>accelerometers in a=20
>precision tilt application the 60mV appears as a significant =
>error.
>
 
>
Thanks in advance,
>
 
>
Mike Falck
>
Crossbow Technology
>
Ph:  =
>(408)965-3315.
>
Email: href=3D"mailto:[email protected]">[email protected]
>
 
>
>------=_NextPart_000_0016_01BE00B6.7C745840--
>
>
>