2
\$\begingroup\$

I want to use section A of an LM358 OpAmp to limit the current and section B to convert Arduino PWM to analog, and then connect output B to non-inverting input of input A as reference voltage.

The LM358 VCC is not going to be lower than 12V or higher than 16V.

Is there something like "leakage" of current/voltage or anything else between two sections that can damage the Arduino or two sections are isolated from each other?

\$\endgroup\$
  • \$\begingroup\$ As an aside, it's probably better to use an ADC if you're already using external components. \$\endgroup\$ – TLW Sep 21 at 22:55
  • \$\begingroup\$ @TLW Did you mean "DAC" instead of "ADC" ? \$\endgroup\$ – Nick Alexeev Sep 25 at 14:21
4
\$\begingroup\$

There is a very slight interaction between the two amplifiers at low frequencies that is likely equivalent to a few uV of Vos shift.

That is in part due to the shared bias network. Changing dissipation from the output section of one amplifier can cause temperature gradients across the die as well has heating, as @analogsystemsrf mentions in a comment, which will cause Vos changes in the other amplifier.

If both amplifiers are working with signals measured in volts it won’t likely be noticeable, let alone problematic.

\$\endgroup\$
  • \$\begingroup\$ side question; is it a good decision to use arduino PWM as a reference voltage after converting it to analog? this way i can achieve milivolts which i can't with get with a voltage regulator. \$\endgroup\$ – newbie Sep 21 at 18:30
  • 1
    \$\begingroup\$ At best it will be no better than the supply voltage in span tolerance. There may be small zero shifts due to current passing through the chip's GND pin etc. Just roughly a few percent accuracy and maybe 1% stability probably okay. If you need more like 0.1% accuracy, use something like a LM4040. I \$\endgroup\$ – Spehro Pefhany Sep 21 at 18:34
3
\$\begingroup\$

You should be ok. The two op amps are mostly isolated from each other. However, because they are both share the same die, if you were to exceed the maximum limits on one of the op amps, you may screw up the operation of the other. ( Example...driving one of the inputs of one op amp at a voltage greater than Vcc may very well effect the operation of the other op amp.)

\$\endgroup\$
  • 2
    \$\begingroup\$ heat in one will affect the input offset voltage of the other. \$\endgroup\$ – analogsystemsrf Sep 21 at 17:16
1
\$\begingroup\$

Considering the resolution of PWM and possible error to desired target and the jitter error of a PWM to LPF average voltage, I think the crosstalk error of -120dB is irrelevant.

\$\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.