I have made a simple piezo-energy harvester. The piezo strip is connected to a full-wave bridge rectifier, then into a capacitor for storage. The voltage across the capacitor increases as expected, when measured with a multimeter - however drops rapidly due to the allowed current flow.

I then connected an INA126 single instrumentation-amp to allow the reading of the capacitors' voltage without current flow. A voltage-divider from 9V source supplies the opamps' V+ of 5V, and V- is ground, the same ground as the battery, as are the opamps. The voltage-divider has a capacitor in parallel to ground, and the Vout runs into an OP07CP precision opamp as a unity-feedback buffer for voltage regulation.

My problem is the voltage measured from Vout of the instrumentation-amp and ground, is an inaccurate reading of the capacitor to a direct reading across the capacitors terminals.

For example, the capacitor currently has a voltage of 1V, and the output of the IA126 with ref. to ground is a stable 3.6V.. And the voltage-divider output is roughly 5V.

PS There is no Rg resistor, to add gain over 5..

Instrumentation amp IC AC voltage source directly from piezo-strip


The INA126 common-mode input voltage range when powered from ±15 V supplies is typically ±11.5 V. This means it struggles with accuracy for inputs close to either rail. If the negative rail is 0 V (as you state) then measuring voltages lower than +3.5 volts above ground is going to be a problem.

See also section Input Common-Mode Range - it says: -

The input common-mode range of the INAx126 is shown in Typical Characteristics. The common-mode range is limited on the negative side by the output voltage swing of A2, an internal circuit node that cannot be measured on an external pin.

Section 9.1 also states this: -

The INAx126 can be operated from a single power supply with careful attention to input common-mode range

I expect that on lower supplies this problem eases somewhat but you can rely on the fact that trying to measure small voltages close to the negative rail is going to be difficult.

You should use an IAmp that is rail-to-rail ideally.

  • \$\begingroup\$ Large resistors won't make the inputs work closer to 0V. Low offset voltage and drift, or common-mode rejection are not what I'm referring to. \$\endgroup\$
    – Andy aka
    Oct 26 '16 at 12:22
  • \$\begingroup\$ If I used V+ of 4.5V, and V- of -4.5V would that alleviate this? \$\endgroup\$
    – Rogan Josh
    Oct 26 '16 at 12:25
  • \$\begingroup\$ Yes it should!! \$\endgroup\$
    – Andy aka
    Oct 26 '16 at 12:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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