←
→
eftychia: Me in kilt and poofy shirt, facing away, playing acoustic guitar behind head (Default)
Add MemoryShare This Entry

Posted using TxtLJ

posted by [personal profile] eftychia at 12:36am on 2007-09-02
If you wanted to build a recording digital accelerometer, would a piezo sensor per axis be a reasonable start?
←
→
There are 4 comments on this entry. (Reply.)
 

(no subject)

posted by [identity profile] skreidle.livejournal.com at 04:41am on 2007-09-02
Do they make piezo accelerometers? The piezo sensors that come to mind could only measure acceleration if attached between one static object and one movable object.

Might I recommend the torrent of accelerometer acquisition/usage info that is Google? :)
[link] [reply]
 

(no subject)

posted by [identity profile] garnet-rattler.livejournal.com at 12:40am on 2007-09-04
The short answer is yes, but ...

The normal way its done is to set a piezo oscillating and read off the average voltage level over a ~lot of cycles. Thus, for no accelleration in that particular plane you get zero, but if there was a delta a, then you get a direction-signed envelope with instantaneous amplitude directly proportional to the accelleration. Adding a mass to the free end of the crystal (or middle if supported on the sides) increases sensitivity at the expense of range (usually spec'd in 'g's) and a cut in resonant frequency.

Several integrated, IC-packaged units are out there, even in the surplus market, for one, two and three axis, and with g ranges of +-2 and +-10 that I know of. Probably numerous others if you look harder. I also have a pile of small piezo elements lying about that you could use if you wanted to make (and calibrate ... less fun, but more educational) your own.
[link] [reply]
 

(no subject)

posted by [identity profile] djwildstar.livejournal.com at 04:45am on 2007-09-04
Even better, there are a bunch of pre-packaged, integrated digital acellerometers out there that you can get, hook up to some sort of microcontroller, and write the software. I believe that these are mostly based on a capacitative sensor rather than a piezo sensor ... but since the analog to digital translation and calibration are already done for you, it doens't matter.

The ones I'm familiar with come in a variety of parts with different sensitivities from +/- 2G to +/- 200G full scale. They generally emit a serial pulse stream of counts, so all the microcontroller has to do is count the pulses and then translate the accumulated count into the number of interest.

Take a look at Silicon Designs model 1010 or Analog Devices iMEMS (ADXLxxx) parts. The Analog Devices line has parts designed for use in airbag deployment sensors, among other things.

I've you're nice and polite (and can fake some letterhead), you may be able to get one or two of the parts as engineering samples.
[link] [reply]

Links

January

SunMonTueWedThuFriSat
1
1
 2
1
 3
1
 4
1
 5
1
 6
1
 7
1
8
1
 9
1
 10
1
 11
1
 12
1
 13
1
 14
1
15
1
 16
1
 17
1
 18
1
 19
1
 20
1
 21
1
22
1
 23
1
 24
 
 25
 
 26
 
 27
 
 28
 
29
 
 30
 
 31