0
\$\begingroup\$

So this is kind of vague question. I'm wondering what the approach to biasing MOSFETs in saturation is.

For the following circuit, I really have no specifications - I'm just trying to play around with it and get the highest gain I can.

Here's my usual approach for biasing:

  • Set supply voltage (Vdd) to some reasonable value like 1.8V or 1.5V
  • Have all MOSFETs ratios at default (I'm literally just trying to get them into saturation first)
  • Start off by trying to bias the PMOS current source - one transistor at a time
  • To do so, I would adjust the gate voltage for those PMOS devices starting off at something like 1V. But the issue arises that my drain source voltage for that device would be far too low and I would never be able to get it into saturation.

enter image description here

Again, I'm not looking for an exact answer here - just some guidance into what sort of approaches/mindset you have when biasing. Majority of the textbooks I've seen really seem to just gloss over biasing and I haven't actually seen a concrete trial/error example for it. They just assume it's always biased properly.

\$\endgroup\$
  • 1
    \$\begingroup\$ Use the circuit tool and draw a better schematic \$\endgroup\$ – laptop2d Mar 15 at 21:31
  • 1
    \$\begingroup\$ @laptop2d done. \$\endgroup\$ – AlfroJang80 Mar 15 at 21:51
  • 1
    \$\begingroup\$ You picked a circuit where the (DC) biasing isn't that easy. I suggest that you start with a circuit similar to a common emitter stage, so a common source: electronics-tutorials.ws/amplifier/mosfet-amplifier.html Understand what is done there to bias the transistor. \$\endgroup\$ – Bimpelrekkie Mar 15 at 21:56
  • 1
    \$\begingroup\$ The circuit you have is very unpractical on its own as getting the biasing voltages right is tricky. Sometimes such a circuit is used on a chip but to bias it properly I would need a lot of extra transistors. That's probably why the books just assume the biasing is correct, it would be too complex to explain how to do that properly. The books usually just want to explain how the circuit behaves wrt how it amplifies a signal. So don't worry too much about the biasing just now. Once you learn more how to bias the circuit will become more clear. \$\endgroup\$ – Bimpelrekkie Mar 15 at 21:58
  • 2
    \$\begingroup\$ Small anekdote: Many years ago I followed a course about "Systematic design of feedback amplifiers". In that course we were taught to first get the signal path right (so gm, load impedances etc.) and simply force the proper biasing using current and voltage sources etc. Then when the signal path is set, the DC biasing is added. I still think that that is a good method as it teaches to separate the two (signal and bias) which makes designing amplifiers much more straightforward. \$\endgroup\$ – Bimpelrekkie Mar 15 at 22:03
0
\$\begingroup\$

Consider this

schematic

simulate this circuit – Schematic created using CircuitLab

\$\endgroup\$
0
\$\begingroup\$

The most common way of biasing this type of amplifiers I have encountered is like this:

schematic

simulate this circuit – Schematic created using CircuitLab

The amplifying transistor M3 is configured as a diode-connected transistor via \$R\$ to make it self-biasing. At high frequencies, the capacitor \$C\$ will couple the signal to the gate of M3 which will be amplified. More advanced schemes probably exist but rely on the same principle.

\$R\cdot C\$ is chosen large to have a passband at low frequencies. \$R\$ should be very large to avoid resistively loading the output which can decrease the gain.

\$\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.