1
\$\begingroup\$

I intend to design a CMOS voltage regulator that takes 2.5V supply and outputs 1.2V. I am starting off with an opamp+pass transistor design.

Will there be a specific advantage to using an NMOS pullup device vs PMOS pulldown for the pass transistor?

Additional details: The regulator supplies this constant voltage but is also supposed to sink a lot of current (~5 mA)

\$\endgroup\$
  • 1
    \$\begingroup\$ This is an IC design or a discrete components design? \$\endgroup\$ – The Photon Jun 13 '13 at 23:19
  • \$\begingroup\$ Sorry. It is an IC design question. \$\endgroup\$ – BobLobLaw Jun 13 '13 at 23:21
1
\$\begingroup\$

The primary advantage of an NMOS transistor is that the higher mobility of electrons means that you can use a physically smaller transistor to pass a given amount of current. Of course, the disadvantage is that if you use an NMOS pass transistor then the maximum \$V_{GS}\$ you can provide is 1.3V, but if you use a PMOS transistor the maximum \$V_{GS}\$ is 2.5V. Since you want to sink as well as source current the typical way of doing this is to add a fully complementary buffer (i.e. a big inverter) at the output of the op amp, with a PMOS to 2.5V and an NMOS to ground.

\$\endgroup\$
  • \$\begingroup\$ I do not have to source any current. The output of this regulator will only be switched in when the receiving circuit (high cap load) needs to sink current while expecting a constant voltage for its bias. I am assuming this would mean NMOS would be preferable since using a PMOS would mean no path for the current to sink to. \$\endgroup\$ – BobLobLaw Jun 19 '13 at 19:10
  • 1
    \$\begingroup\$ Yes, you certainly want an NMOS. What you are describing would not normally be called a voltage regulator with a pass transistor, it's a constant voltage load. \$\endgroup\$ – Joe Hass Jun 19 '13 at 19:35
  • \$\begingroup\$ Would that be called a constant voltage load even if the load itself is variable? I was under the impression this would be a shunt regulator or just a general voltage regulator that sinks instead of supplying current. I see why the NMOS wouldn't be a 'pass' device as we are not trying to pull the voltage to ground but I am not clear on the nomenclature. \$\endgroup\$ – BobLobLaw Jun 19 '13 at 20:50

protected by clabacchio Jun 14 '13 at 6:02

Thank you for your interest in this question. Because it has attracted low-quality or spam answers that had to be removed, posting an answer now requires 10 reputation on this site (the association bonus does not count).

Would you like to answer one of these unanswered questions instead?

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