1
\$\begingroup\$

I was trying to amplify the signal of a piezo, which is very low and amplify it so that I could use it to latch a flipflop. I had the LM358 opamp so I made a differential amplifier by a gain of 50 so that the output would reach 5 volts which is a high logic level.

I got the design of it from internet, I just changed the gain. I was measuring the output with my multimeter (I do not own an osciloscope,) but the output was never what was expected. The maximum voltage I measured was 1volt, but it should have been 5volts.

No matter what I tried, the output was same. I do not know if the opamp is not working or if I am measuring wrong.

Please tell me what am I doing wrong or what the problem is.

enter image description here shematic

\$\endgroup\$
3
  • \$\begingroup\$ Piezos don't produce an output voltage on their own. Were you aware of that? \$\endgroup\$
    – Andy aka
    Commented Nov 17, 2020 at 17:48
  • 1
    \$\begingroup\$ @Andyaka Actually, they do, as that is pretty much what the piezoelectric effect is about. One need only connect a piezo transducer to a high impedance scope to see it. A charge imbalance is created; the structure has a capacitance, therefore you see a voltage. See a barbecue grill lighter for a more extreme example able to exceed the breakdown voltage of a spark gap. \$\endgroup\$
    – Chris Stratton
    Commented Nov 17, 2020 at 19:20
  • \$\begingroup\$ Try adding another 500K resistor from the inverting input to VCC. You'll then probably find it is actually too sensitive and the real issue will be modifying it to only report things of interest. \$\endgroup\$
    – Chris Stratton
    Commented Nov 17, 2020 at 19:30

2 Answers 2

Reset to default
4
\$\begingroup\$

Aside from obvious questions as to whether your piezo is the same as whatever some random site on the internet might have used, and whether yours is being hit as hard, there is really no way to measure the output signal with a multimeter. The pulse will be brief.

If you are seeing 1V on a multimeter it may well be railing at +3-4V (the most the op-amp can manage, assuming a +5V supply- you don't specify) briefly. Try it on the flip-flop.

\$\endgroup\$
1
  • 1
    \$\begingroup\$ Thanks for the suggestion. It worked , I hooked the out put to the flipflop and it worked like a charm. I think the opamp was amplifying the signal after all. Maybe the pulse width was too short for the multimeter. I think the signal peak was logic level high as expected.Thank you again for your kind time and idea. Have a good day. \$\endgroup\$
    – nuren
    Commented Nov 18, 2020 at 5:04
0
\$\begingroup\$

If you look up the max input offset voltage error in the datasheet, and combine that with the worst-case tolerance of the resistors you are using, you will see that even if you remove the piezo device and connect the two 10K resistors (reference designators!!!) together, it is possible for the output to be very different than expected

To do what you want, you need both a better opamp and much tighter tolerance resistors. Also, note that the 358 output cannot swing all the way up to Vcc nor all the down to GND.

\$\endgroup\$
2
  • \$\begingroup\$ How much does the resistor tolerance affect the output? \$\endgroup\$
    – Spehro 'speff' Pefhany
    Commented Nov 17, 2020 at 18:03
  • \$\begingroup\$ You shouldn't need a better op-amp, but rather a design which is not particularly sensitive to offset. Remember the goal here is just to detect an "event" - linearity, even magnitude are non-issues. \$\endgroup\$
    – Chris Stratton
    Commented Nov 17, 2020 at 19:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.