In a message dated 09/08/2005, shammon1@............. writes:

In 1991 one of the PSN members build a number of  piezo devices. He had some 
issues with the output because if you  "tap" it hard, the output can be quite 
high and damage the  electronics.

 
Hi There,
 
    For anyone who wants to try out a loaded piezo  disk, the Digi-Key 
102-1170-ND at $1.62 looks suitable. It is 44mm OD, has a  nominal capacity of 70 nF 
and wire leads. It's natural frequency is 600 Hz.  Mouser also stock a 
similar item 665 KBI 4406.
 
    I use Holdtite Acrylic Adhesive ST3295 in a twin  syringe pack. This 
adhesive stays a bit flexible and the set time is about 5  min. It does NOT go 
brittle like epoxies. I glue a screw head to the centre of  the brass disk and 
then screw a weight onto that. You crack fewer disks that  way.... You can also 
use twinpack polyurethane adhesive.
 
    If you try to solder a screw head onto the disk,  the heat depolarises 
the PZT element. It is also likely to crack it. I use  weights of about 1 to 3 
oz -- 30 to 100 gm, typically ~50 gm. My disks are a bit  stiffer than the 
above ones, so 1 to 1.5 oz should be OK.
 
    You DO need to connect two signal diodes across the  piezo element to 
protect the input to the opamp - see above.
 
 
The devices worked best for very high freq and did  not work at all for low 
freq.

    My disk works out to it's 4 Sec limit OK. You  depend on the CR time 
constant to enable you to see the longer periods, but the  output decreases, as 
you would predict for an accelerometer. 70 nF across  47 M Ohms = 3.3 seconds.
 
    Regards,
 
    Chris Chapman
 






In a message dated 09/08/2005, shammon1@............. writes:
<=
FONT=20
  style=3D"BACKGROUND-COLOR: transparent" face=3DArial color=3D#000000 size=
=3D2>In 1991 one of the PSN members build a number o=
f=20
  piezo devices. He had some issues with the output because if you=
=20
  "tap" it hard, the output can be quite high and damage the=20
  electronics.
 
Hi There,
 
    For anyone who wants to try out a loaded piezo=20
disk, the Digi-Key 102-1170-ND at $1.62 looks suitable. It is 44mm OD, has a=
=20
nominal capacity of 70 nF and wire leads. It's natural frequency is 600=
 Hz.=20
Mouser also stock a similar item 665 KBI 4406.
 
    I use Holdtite Acrylic Adhesive ST3295 in a twi=
n=20
syringe pack. This adhesive stays a bit flexible and the set time is about 5=
=20
min. It does NOT go brittle like epoxies. I glue a screw head to the centre=20=
of=20
the brass disk and then screw a weight onto that. You crack fewer disks that=
=20
way.... You can also use twinpack polyurethane adhesive.
 
    If you try to solder a screw head onto the disk=
,=20
the heat depolarises the PZT element. It is also likely to crack it. I use=20
weights of about 1 to 3 oz -- 30 to 100 gm, typically ~50 gm. My disks are a=
 bit=20
stiffer than the above ones, so 1 to 1.5 oz should be OK.
 
    You DO need to connect two signal diodes across=
 the=20
piezo element to protect the input to the opamp - see above.
 

<=
FONT=20
  style=3D"BACKGROUND-COLOR: transparent" face=3DArial color=3D#000000 size=
=3D2>The devices worked best for very high freq and=20=
did=20
  not work at all for low freq.
    My disk works out to it's 4 Sec limit OK. You=20
depend on the CR time constant to enable you to see the longer periods, but=20=
the=20
output decreases, as you would predict for an accelerometer. 70 nF acro=
ss=20
47 M Ohms =3D 3.3 seconds.
 
    Regards,
 
    Chris Chapman