Differences between voltage regulator topologies

[bloguetronica]

Hi,

I'v seen this article on Wikipedia: https://en.m.wikipedia.org/wiki/Error_amplifier_(electronics). It uses a PNP transistor as the pass element, and the error amplificador in this topology has its inverting input connected to the voltage reference.

However, I've been using a different topology: a NPN transistor as the pass element, controlled by an op-amp where the non-inverting input is connected to the voltage reference (VDAC) and the inverting input is used for feedback. Please, see the schematic in the attached image.

My questions are:
- Can the op-amp still be called an error amplifier in the latter topology?
- I see this PNP transistor topology very often. What are the advantages of using such topology?
- Which one has better load regulation? And line regulation?

Thanks in advance.

Kind regards, Samuel Lourenço

[Kleinstein]

There are mainly 2 types of voltage regulators with different types of power stages:
A) with a power-stage with low output impedance, like an emitter follower
B) with a power-stage with high impedance, like a transistor in emitter circuit. This output-stage is more current controlling.

Type A are the classical regulators like the 78xx series. Usually they have quite some drop from a Darlington output stage.
The low drop voltage regulators are thy type B), often with an PNP or P-MOSFET as an output element. 

The different stages need a different compensation: the emitter follower version can get away with a simple single pol compensation. This makes it relatively easy and ideally it can get away with little capacitance at the output. For very tight regulation it may still use a more complicated compensation.
The low drop regulators essentially needs quite some capacitance at the output, often with a defined ESR. The compensation needs more than just a single pole.

How good the regulation is depends on the implementation. Especially the higher frequency performance depends a lot on the output capacitors and also the layout. There is a slight tendency for the conventional regulators to be better with voltage regulation, but worse with current regulation. However the need for the output cap can compensate some of this. NPN transistors tend to be a little fast, which helps a little for better regulation.

[bloguetronica]

Hi,

I'm confused. So A would be NPN, and B would be PNP? What about line regulation?

Kind regards, Samuel Lourenço

[Kleinstein]

For positive output the type A) would use NPN, the type B) would use a PNP or P-MOSFET.

For modern regulators line regulation is often very good. There is not really a principle difference here, at least for the low frequency part. The higher frequency part PSRR may be slightly better for the classical regulators, but it depends - there are good and not so good low drop regulators and the output capacitor can make a huge difference.

For integrated regulators much of the load regulation depends on the pin resistance. Not many regulator chips have separate sense pins.

[David Hess]

- Can the op-amp still be called an error amplifier in the latter topology?

Yes, the operational amplifier is an error amplifier in both configurations.  The difference is that the function of the inverting and non-inverting inputs is reversed when the PNP output transistor is used because it inverts the output of the operational amplifier.

- I see this PNP transistor topology very often. What are the advantages of using such topology?

The PNP transistor does not require a bias voltage above the positive supply to operate so is commonly used in low dropout positive regulators.

- Which one has better load regulation? And line regulation?

For a positive voltage regulator, the NPN output transistor generally performs better and is easier to use but either can be made to work very well.

[bloguetronica]

Thanks for your answers so far!

Kind regards, Samuel Lourenço