Negative LDOs with high PSRR?

[langwadt]

If you have a positive supply, you don't need a negative reference, just an op amp and a PNP.
(Assuming you don't need fancy current limiting/thermal protection, which you shouldn't at 200mA.)

You mean by using the positive rail as a reference?

kinda, you know that midpoint between +/- supply is gnd

http://www.seekic.com/uploadfile/ic-circuit/s201159215141626.gif

[magic]

Maybe you could make a discrete regulator with a LM3411 and a transistor?
The LM3411 is a kind of 'negative TL431' with reference tied to V+.

This is an interesting part, but the internal feedback network with ±25% tolerance is not helping when you need custom closed loop voltage. Without this rubbish it could be a viable negative TL431 substitute.

There is supposed to exist adjustable LM385, which is more like a true complement of 431. But it's fairly low voltage only, so further parts would be required to deal with that.

If you have a positive supply, you don't need a negative reference, just an op amp and a PNP.
(Assuming you don't need fancy current limiting/thermal protection, which you shouldn't at 200mA.)

You mean by using the positive rail as a reference?

Looks like the cheapest way.

Make sure the opamp isn't doing something stupid when the positive rail collapses (or either of the upstream supplies disappears) and inputs may go outside normal common mode input range.

Watch out for PSRR of the opamp itself.

[SiliconWizard]

Maybe you could make a discrete regulator with a LM3411 and a transistor?
The LM3411 is a kind of 'negative TL431' with reference tied to V+.

This is an interesting part, but the internal feedback network with ±25% tolerance is not helping when you need custom closed loop voltage. Without this rubbish it could be a viable negative TL431 substitute.

There is supposed to exist adjustable LM385, which is more like a true complement of 431. But it's fairly low voltage only, so further parts would be required to deal with that.

If you have a positive supply, you don't need a negative reference, just an op amp and a PNP.
(Assuming you don't need fancy current limiting/thermal protection, which you shouldn't at 200mA.)

You mean by using the positive rail as a reference?

Looks like the cheapest way.

Make sure the opamp isn't doing something stupid when the positive rail collapses (or either of the upstream supplies disappears) and inputs may go outside normal common mode input range.

Watch out for PSRR of the opamp itself.

Yes if we already have a positive supply and a regulated positive rail, we don't need an extra reference.

I have played a bit with simulation to see what could be achieved. What I did here was to set up the opamp as an inverter rather than follower and use the positive rail as a reference. It works fine.

The potential issues are the usual - possible oscillation, PSRR and what you mentioned above.
And, if we want reasonable protection against overcurrent, it adds another layer. Did I mention I was looking to save some space.

But it's interesting to work on.

[edavid]

If you have a positive supply, you don't need a negative reference, just an op amp and a PNP.
(Assuming you don't need fancy current limiting/thermal protection, which you shouldn't at 200mA.)

You mean by using the positive rail as a reference?

Right!