Reducing noise in voltage regulators

[jcarr]

Artnyos:

>The circuit is complicated enough as it is now; and more parts complicate stability, lengthen signal paths (and complicate ground current return paths which are critical) and make PCB layout more difficult.<

You are right, I do tend to ignore issues of complexity and difficulty when designing circuits and pcb layouts.

But a circuit wouldn't be fun if it didn't present a worthy challenge ;-).

regards, jonathan carr

[artnyos]

I do tend to ignore issues of complexity and difficulty when designing circuits and pcb layouts.

I noticed..... PCB layout is a little more complicated when the parts are actually on the board.

I got tied of 3Dcircuits when doing mods on circuit developement on telcom boards using SMT parts. I have done a few under a microscope. I try to do point to point with the parts in one plane. Much easier to measure, fix, or modify. But as they say, "Whatever floats your boat."

Art

[artnyos]

Sulzer type regulator. Veiw of positive half. Negative regulator half not finished.

Art

[ALW]

[QUOTE]ALW Andy: Looking at the schematic that you posted on the first page of this thread, how about doing a version of your regulator with a push-pull output?[QUOTE]

The schematic I'm working with is closer to that at the top of p6.

I don't feel PP offers any great benefits, and brings with it extra complication. For the power levels we're talking about for preamps, the reg. o/p can always be pre-loaded with a resistive load, in order to provide some current sinking capability.

I've used this scheme in a number of apps.

A.

[artnyos]

http://hitechnetworks.net/bwaldron/a..._regulator.htm

Art

[jwb]

This is probably the most informative and important thread on the whole of diyaudio.com. Thanks for so much great information.

My question is what to do about 5V and 3.3V supplies. If I need e.g. 5.5V for the analog side of a DAC, I can't use any of these designs that include LM329, because that is a 6.9V reference. If I feed the output of a Sulzer regulator into a regulator such as LT1962, will the LT1962 simply add back in a bunch of noise? How much noise do standard regulators generate internally?

I realize that a 5V supply filter was discussed in the very first page of this thread, but it doesn't answer the question of how to apply the Sulzer or similar regulator to a logic-level supply.

[ALW]

Here's my solution...

Jung reg, THS4271 op-amp (3nV/rt Hz, 1000V/us, 1.4GHz GBW!), TL431 set to it's reference voltage of 2.5V.

THS4222 could be promising for 3.3V, but is noisier @ 13nV/rt Hz.

A.

[borisov57]

Has someone tried made a sulzer type rugulator with high speed comparator (like ad8561) ?
I guess they much faster than opamps, but have lower gain and i do not know about noise voltage.

Have a nice day,

borisov57

[Optical]

Originally posted by artnyos
http://hitechnetworks.net/bwaldron/a..._regulator.htm

Art

how much current can be constantly drawn from one of these supplies??
Also i cant find those transistors in my area, will any low noise NPN/PNP pair work?

[jwb]

Voltage regulators ... they always come up in every project! Here's the thing:

1) Building yet another DAC
2) Some logic is 3.3V
3) Some logic is 5V
4) The analog side of the DAC needs ??
5) The clock also needs 9V

So I'm in the position where I need 3.3V, 5V, and 9V regulators, where the 9V needs to be quiet out to the clock speed of ... 24.576MHz. I like these Sulzer and Jung regulators designs, but I like the Sulzer better because it runs from the bulk supply, not the regulated supply. This allows me to run all the opamps from +15V bulk, instead of trying to find a good opamp that can run from a 3.3V single supply.

For the 9V supply, I can easily use the bog standard Sulzer with LM329 reference. To get the bandwidth I want I could go with AD797, it has serious specs. The problem I see in the Sulzer schematic is the huge AC feedback capacitor. Why 4.7uF? This rolls off the feedback at a very low frequency. The 797 has plenty of open-loop gain left to burn at 1MHz, and the closed-loop gain of the regulator would be only 2.3, so why not use 100pF for the AC feedback? Am I looking at the wrong impedance spec on the capacitors?

For the 3.3V and 5V regulators I believe I will use the solution ALW posted above. Thanks for the tip on the THS4271.

-jwb