The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> opamp for ldo
https://designers-guide.org/forum/YaBB.pl?num=1156252990

Message started by grosser on Aug 22^nd, 2006, 6:23am

Title: opamp for ldo
Post by grosser on Aug 22^nd, 2006, 6:23am

hello

Can you recommend an opamp for LDO regulator which is suitable to work under low voltage (Vin=1V-3.3V)?

if you have, upload the schematics please

regards

Title: Re: opamp for ldo
Post by loose-electron on Aug 22^nd, 2006, 7:50am

two questions -

Input voltage range?
What are you driving for a load?

Jerry

Title: Re: opamp for ldo
Post by grosser on Aug 22^nd, 2006, 9:04am

loose-electron wrote on Aug 22^nd, 2006, 7:50am:

two questions -

Input voltage range?
What are you driving for a load?

Jerry

input range 1V-3.3V

Iload 1mA-100mA

Title: Re: opamp for ldo
Post by loose-electron on Aug 23^rd, 2006, 8:25am

That large an output current, you will need a source follower type buffer for an output driver.
Also, your input voltage is not going down to ground so a NMOS differential pair input should work.

Add an NMOS source follower to the attached and it should get the job done. The compensation here uses pole splitting, there are other options as well.

Title: Re: opamp for ldo
Post by Visjnoe on Aug 23^rd, 2006, 10:55am

Hi,

I would just like to comment on the OTA topology given above and note that:

1. It has a certain systematic offset, which might be kept low enough if you have enough gain in the first stage.
2. It has a very bad PSRR. Insert a cascode to improve this.

Kind Regards

Peter

Title: Re: opamp for ldo
Post by loose-electron on Aug 23^rd, 2006, 6:35pm

cascode will help with the PSRR a bit yes, agreed.

The systematic offset will probably be overshadowed by the differential pair mismatch.

An LDO usually ends up getting trimmed (usually the bandgap reference) to get to the right operating point output, so that should fall out of the equation.

Title: Re: opamp for ldo
Post by grosser on Aug 24^th, 2006, 12:36am

hi

i have tried this simple topology, but it is not enough because of the source follower.
For the lowest Vin=1V it can drive the full load (100mA) but can't drive low load current.

I'm looking for something with simple architecture and good slew-rate current.

regards

Title: Re: opamp for ldo
Post by loose-electron on Aug 24^th, 2006, 10:07am

grosser -

Not sure exactly what the problem is, if you could define the problem a bit more perhaps I can help.

Jerry

Title: Re: opamp for ldo
Post by grosser on Aug 24^th, 2006, 1:18pm

loose-electron wrote on Aug 24^th, 2006, 10:07am:

grosser -

Not sure exactly what the problem is, if you could define the problem a bit more perhaps I can help.

Jerry

the problem is that for low input Voltage=1V when i use source follower with nmos input opamp can drive large load, but can't drive low load and can't shutt off when it's needed.

When i use source follower with pmos input i can't drive large load.

i know this topology is not suitable for low voltage and i want to check other possibilities.

Regards

Title: Re: opamp for ldo
Post by richard88 on Aug 24^th, 2006, 10:02pm

grosser,
I think you have to specify more info for an LDO :
What is your VIN (or VDD), Vref, Vout ? is the feedback factor=1 (no resistor divider, direct feedback) ?
Richard

Title: Re: opamp for ldo
Post by loose-electron on Aug 25^th, 2006, 8:40am

Quote:

the problem is that for low input Voltage=1V when i use source follower with nmos input opamp can drive large load, but can't drive low load and can't shutt off when it's needed.

When i use source follower with pmos input i can't drive large load.
i know this topology is not suitable for low voltage and i want to check other possibilities

Ah, if that is the case, you can use an OTA type output, with lots of gain. You do this in a Class-AB type structure with a PMOS and NMOS outputs being controlled. With lots of gain it give the capability for either the PMOS or NMOS output to be driven triode.

That will get your control for the PMOS power transistor to go from ground to power (or pretty bloody close)

in concept (sorry dont have a picture)

2 NMOS differential pairs

2 current mirror structures above the differential pairs

opposite polarity on the high gain output of each differential pair

one output drives the PMOS output of the OTA
other output drives a PMOS current source, that mirrors into a NMOS (diode connected) which drives the NMOS output of the OTA.

There are other ways of doing this as well, with one differential pair, and you may need additional gain before getting to the NMOS/PMOS outputs.

The key thing is the A/B output stage.

Jerry