0
\$\begingroup\$

schematic

simulate this circuit – Schematic created using CircuitLab

A linear regulator is typically an error amp (OA_error), which changes how much the pass transistor Q_pass dissipates in order to bring Vout and V_ref to the same voltage.

I'm curious what improvements there are in topology to this basic linear amplifier design. For the moment, I'm mostly interested in improvements in accuracy and noise performance.

\$\endgroup\$
1
  • \$\begingroup\$ Its all in the implementation, is this an IC design on silicon? Is it discrete components? What is your load? \$\endgroup\$
    – Voltage Spike
    Commented Feb 15, 2016 at 23:55

2 Answers 2

1
\$\begingroup\$

All linear regulators are similar to this- there is noise due to the reference and due to the amplifier. The popular band-gap reference tends to be noisy. A buried zener reference is typically less noisy but requires a relatively high voltage. All references drift with temperature and time, and there will be some effect from the power supply voltage.

There is some error due to the offset voltage and finite gain of the amplifier, as well as the reference. Adding an integrator could make the error disappear, at the expense of some stability. The amplifier will also have some effect due to changes in supply voltage (line regulation).

\$\endgroup\$
2
  • \$\begingroup\$ There haven't been any new regulation topologies in the last 40 years that offer better accuracy/noise performance? \$\endgroup\$ Commented Feb 16, 2016 at 21:41
  • \$\begingroup\$ Of course those things have improved (especially on a cost basis) and implementations keep getting better (low power consumption, lower noise etc.) but the basic topology is still a feedback loop comparing the output voltage to a reference of some kind. What else could it be? For example, the ultra-low noise ADM7150: "Internally, the ADM7150 consists of a reference, an error amplifier, and a P-channel MOSFET pass transistor". . \$\endgroup\$ Commented Feb 16, 2016 at 21:48
0
\$\begingroup\$

There aren't really any improvements that can be made in the ideal sense such as the circuit you have shown here. Like @Spehro said, you can add a different feedback mechanism. This question is like saying, how can I make a refrigerator more efficient, without specifying what kind of cooling mechanism your using.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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