0
\$\begingroup\$

I only have one power supply and I only have the LF353N opamp. Is it possible to design a non-inverting amplifier circuit for this op-amp with only a single power supply?

Otherwise, what are some single power supply op-amp alternatives could I use? I'm using a non-inverting amplifier to amplify a 1V pk-pk 10kHz sine wave to 5V pk-pk.

\$\endgroup\$
3
\$\begingroup\$

Op-amps are not inherently single or dual supply orientated. OK there may be a tiny majority of exceptions that have a "ground" or "0 volt" pin but the vast majority have a positive power pin, a negative power pin, 2 inputs and an output.

The rules are simple: -

  • Supply enough power voltage differential to exceed the lower power limit voltage
  • Don't supply more than the maximum power rail
  • Keep inputs within the upper and lower power rail as dictated by the "input common mode voltage range" in the data sheet
  • Don't expect the output to swing to or beyond either power rail

So, the LF353A expects to be run from a power rail differential greater than 10 volts and less than 36 volts. If you want to call that +/- 5 volts and +/- 18 volts that's up to you.

The typical input common mode voltage range is typically -12 volts to +15 volts (on a +/- 15 volt supply) and this translates to +3 volts to +30 volts on a single ended +30 volt supply

OR

+3 volts to +20 volts on a single +20 volt supply.

However, the data sheet only guarantees +/- 11 volts on a +/- 15 volt supply so if you don't want to take risks this translates to +4 volts to + 26 volts on a single +30 volt supply.

There are graphs in the data sheet that give more detail such as figure 6 and 7.

There are also specifications and graphs for output voltage swing.

In short >99% of all op-amps do not understand there power supply regime - they are only interested in the differential supply voltage being within workable limits.

\$\endgroup\$
  • \$\begingroup\$ I think the only problem with me just sticking in say for an example 0V and 12V into the power rails is that my signal will be clipped at the trough. I will have to DC bias my signal in order to see the whole signal. \$\endgroup\$ – Guy Lee May 4 '17 at 9:08
  • \$\begingroup\$ @GuyLee yes, if your input signal sits around 0 volts then you need to bias it up or use split rails. \$\endgroup\$ – Andy aka May 4 '17 at 9:34
0
\$\begingroup\$

It is certainly possible but you will need a low impedance voltage at mid supply. Voltage reference or buffered voltage divider depending on your DC specifications.

\$\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.