0
\$\begingroup\$

I've been working on building a circuit which uses a wheatstone bridge to find the value of a resistance temperature sensor (Danfoss AKS 21) and then passes that to a differential amplifier the output of which is to be read by an ADC.

When I read the voltage across the bridge with a multimeter without the op-amp circuit connected, I get values that I expect (at room temp around 0.87V). However, when I connect the circuit to the op amp, the voltage across the bridge drops to around 0.5 V.

I've included a schematic I made of the circuit below. Would really appreciate some help figuring out where I'm going wrong.

I'm using the same 3.3 VCC to power the op-amp as the rest of the circuit.

Thanks in advance everyone, Al Circuit Diagram

\$\endgroup\$
3
  • \$\begingroup\$ The series combination of R4 and R7 is in parallel to the PTC. A similar thing happens to R3. \$\endgroup\$ – Mario Feb 11 '18 at 14:51
  • \$\begingroup\$ Your op-amp setup has input resistance, in this case around 1.5 kΩ. Your multimeter has input resistance in the G Ω range, or more. - You might want to buffer the inputs to the op-amp. \$\endgroup\$ – Harry Svensson Feb 11 '18 at 15:43
  • 1
    \$\begingroup\$ BTW there's no need at all for a Wheatstone bridge to measure a PT1000 \$\endgroup\$ – carloc Feb 11 '18 at 19:31
3
\$\begingroup\$

You want a circuit with high impedance inputs so as not to load the bridge. You can use this circuit if you want to stay with the LM358 (requires 1.5 chips):

enter image description here

The offset voltage and TCVos is rather high on the LM358 though (maybe 3-10uV/K) so you may wish to use a better amplifier such as a 'zero drift' instrumentation amplifier.

Also 3.3V is barely enough voltage to power an LM358 so you will get very limited input common mode range (as bad as 0V to 0V over temperature), so in fact it is most certainly not guaranteed to work- you are lucky it's functioning at all. If you must use this kind of amplifier at least give it 5V or more to work with.

\$\endgroup\$
1
  • 1
    \$\begingroup\$ This worked for me, I used 150kOhm resistors (thats what I had) With a 2k pot. i used a 5v supply with negative rail to ground \$\endgroup\$ – peterretief Dec 5 '19 at 8:23
0
\$\begingroup\$

Strange Behaviour? Not at all.

For a start R4 and R7 (in series) are across your PTC to ground. And whatever voltage is at the junction of R4 and R7 is projected to the inverting input of the LM358 via negative feedback hence R3 is being sapped of current by R5.

Micro-cap 11 student version is free and you'll never look back once you get the hang of SPICE simulation.

If you want a better solution try an instrumentation amplifier - like a diff amp with high impedance input buffers.

\$\endgroup\$

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.